UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Charismatic leadership : effects of leadership style and group productivity on individual adjustment… Howell, Jane Mary 1986

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1986_A1 H69.pdf [ 16.97MB ]
Metadata
JSON: 831-1.0097286.json
JSON-LD: 831-1.0097286-ld.json
RDF/XML (Pretty): 831-1.0097286-rdf.xml
RDF/JSON: 831-1.0097286-rdf.json
Turtle: 831-1.0097286-turtle.txt
N-Triples: 831-1.0097286-rdf-ntriples.txt
Original Record: 831-1.0097286-source.json
Full Text
831-1.0097286-fulltext.txt
Citation
831-1.0097286.ris

Full Text

CHARISMATIC LEADERSHIP: OF LEADERSHIP STYLE AND CROUP PRODUCTIVITY INDIVIDUAL ADJUSTMENT AND PERFORMANCE By JANE MARY HOWELL B .A. , The University of British Columbia, 1976 M.A. , The University of Western Ontario, 1980 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Faculty of Commerce and Business Administration) We accept this thesis as conforming to the required standard EFFECTS ON THE UNIVERSITY OF BRITISH COLUMBIA January 1986 © Jane Mary Howell, 1986 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Commerce and Business Admi n i s t r a t i on The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date February 23, 1986 DE-6(3/81) ii ABSTRACT The present study examined the effects of three leadership styles (charismatic, structuring, and considerate) and two levels of group productivity (high and low) on individuals' adjustment to and performance on an ambiguous decision making task. One hundred and forty-four Commerce undergraduates participated in a simulated organization (The Mackenzie Institute) which was obstensibly designed to assess their practical business skills. They completed an in-basket exercise directed by a manager (an experimental confederate) who portrayed a charismatic, structuring, or considerate leadership style. Participants individually worked on the exercise in the presence of two other Commerce students (also experimental confederates) who advocated to them either high or low productivity on the task. The participants subsequently completed a questionnaire measuring their adjustment to the task, the manager, and the two student confederates. Univariate analyses of variance generally indicated that individuals with charismatic leaders had significantly higher task performance, task adjustment, and adjustment to the leader when compared to individuals with considerate or structuring leaders. The group productivity data indicated that individuals in the high productivity group reported a significantly greater task satisfaction, lower role conflict and higher adjustment to the group than individuals in the low productivity group. Croup productivity norms had no significant effect on individual task performance. iii The interaction between leadership style and group productivity re-vealed that charismatic leadership, regardless of the directionality of group productivity norms, produced high individual task performance, task adjustment, and adjustment to the leader and to the group. In contrast, the impact of structuring leadership on individuals' task adjustment was modified by group productivity norms: individuals who worked with a structuring leader and in a high productivity group reported higher task satisfaction and lower role conflict than individuals who worked with a structuring leader and in a low productivity group. Individuals with a considerate leader and in a high productivity group had significantly higher task satisfaction than those with a considerate leader and in a low productivity group. Multivariate analyses of the data revealed a similar pattern of results. Explanations and implications of the results are discussed and directions for future research are presented. iv To CAMERON, for his continual support and encouragement. V T A B L E O F C O N T E N T S P a g e A B S T R A C T i i D E D I C A T I O N i v T A B L E O F C O N T E N T S v L I S T O F T A B L E S i x L I S T O F F I G U R E S x L I S T O F A P P E N D I C E S x i A C K N O W L E D G E M E N T S x i v C H A P T E R I - S C O P E O F T H E S T U D Y A N D L I T E R A T U R E R E V I E W I n t r o d u c t i o n 1 O v e r v i e w o f t h e P r e s e n t S t u d y 5 O r g a n i z a t i o n o f t h e D i s s e r t a t i o n 6 A L i t e r a t u r e R e v i e w o f L e a d e r s h i p S t y l e 7 O v e r v i e w 7 T h e o r y a n d R e s e a r c h R e l a t e d to C h a r i s m a t i c L e a d e r s h i p 12 O v e r v i e w 7 W e b e r ' s C o n c e p t u a l i z a t i o n o f C h a r i s m a t i c A u t h o r i t y 14 A n E v a l u a t i o n o f W e b e r ' s C o n c e p t u a l i z a t i o n o f C h a r i s m a t i c A u t h o r i t y 17 P o l i t i c a l S c i e n c e V i e w s o f C h a r i s m a 20 S o c i o l o g i c a l V i e w s o f C h a r i s m a 25 S i t u a t i o n a l C i r c u m s t a n c e s F o s t e r i n g C h a r i s m a 26 T h e E x i s t e n c e a n d D i s t r i b u t i o n o f C h a r i s m a i n O r g a n i z a t i o n s 30 P s y c h o l o g i c a l V i e w s o f C h a r i s m a 32 H o u s e ' s T h e o r y o f C h a r i s m a t i c L e a d e r s h i p 36 A n O v e r v i e w o f t h e T h e o r y 36 R e s e a r c h E v i d e n c e 37 B a s s ' T h e o r y o f T r a n s f o r m a t i o n a l L e a d e r s h i p 41 • A n O v e r v i e w o f t h e T h e o r y 41 R e s e a r c h E v i d e n c e 45 T h e o r y a n d R e s e a r c h R e l a t e d to S t r u c t u r i n g a n d C o n s i d e r a t e L e a d e r s h i p S t y l e s 48 T h e B e h a v i o u r a l A p p r o a c h 48 T a s k O r i e n t e d a n d S o c i o e m o t i o n a l O r i e n t e d L e a d e r s h i p 49 I n i t i a t i n g S t r u c t u r e a n d C o n s i d e r a t i o n 50 T h e C o n t i n g e n c y A p p r o a c h 53 F i e d l e r ' s C o n t i n g e n c y T h e o r y o f L e a d e r s h i p 53 H o u s e ' s P a t h - G o a l T h e o r y o f L e a d e r s h i p 56 L e a d e r s h i p U n d e r C o n d i t i o n s o f S t r e s s 61 vi T A B L E O F C O N T E N T S ( c o n t i n u e d ) P a g e A L i t e r a t u r e R e v i e w o f C r o u p P r o d u c t i v i t y 64 O v e r v i e w 64 T h e I n f l u e n c e o f t h e Wo r k G r o u p on N e w c o m e r s ' A d j u s t m e n t to t h e O r g a n i z a t i o n 65 G r o u p N o r m s f o r P r o d u c t i v i t y 67 D e t e r m i n a n t s o f C o n f o r m i t y to P r o d u c t i v i t y N o r m s 68 Low P r o d u c t i v i t y N o r m s 71 H i g h P r o d u c t i v i t y N o r m s 73 P e r f o r m a n c e 74 A d j u s t m e n t 76 T a s k A d j u s t m e n t 76 I n t e r p e r s o n a l A d j u s t m e n t 78 D e f i n i t i o n o f T e r m s 80 E x p e r i m e n t a l H y p o t h e s e s 82 M a i n E f f e c t s 83 L e a d e r s h i p S t y l e 83 T a s k P e r f o r m a n c e a n d T a s k A d j u s t m e n t 83 A d j u s t m e n t t o t h e L e a d e r 86 C r o u p P r o d u c t i v i t y 88 T a s k P e r f o r m a n c e 88 A d j u s t m e n t to t h e T a s k a n d to t h e G r o u p 88 I n t e r a c t i o n E f f e c t s 90 L e a d e r C h a r i s m a t i c B e h a v i o u r a n d G r o u p P r o d u c t i v i t y 90 L e a d e r S t r u c t u r i n g B e h a v i o u r a n d G r o u p P r o d u c t i v i t y 90 L e a d e r C o n s i d e r a t e B e h a v i o u r a n d C r o u p P r o d u c t i v i t y 91 C H A P T E R II - M E T H O D E x p e r i m e n t a l D e s i g n 94 A N o t e on L a b o r a t o r y S t u d i e s 94 E x p e r i m e n t a l D e s i g n 98 E x p e r i m e n t a l T a s k 100 O p e r a t i o n a l D e f i n i t i o n s o f t h e I n d e p e n d e n t V a r i a b l e s 102 L e a d e r s h i p S t y l e 102 C h a r i s m a t i c S t y l e 102 C o n s i d e r a t e S t y l e 106 S t r u c t u r i n g S t y l e 107 G r o u p P r o d u c t i v i t y 108 H i g h P r o d u c t i v i t y C o n d i t i o n 110 L ow P r o d u c t i v i t y C o n d i t i o n 111 E x p e r i m e n t a l S c r i p t s 112 M a n i p u l a t i o n C h e c k s 112 O p e r a t i o n a l D e f i n i t i o n s o f t h e D e p e n d e n t V a r i a b l e s 113 T a s k P e r f o r m a n c e 113 T a s k A d j u s t m e n t 117 I n t e r p e r s o n a l A d j u s t m e n t 120 vii TABLE OF CONTENTS (continued) Page Operational Definitions of the Individual Difference Variables 121 Need for Achievement 121 Tolerance for Ambiguity 122 Need for Affiliation 123 Experimental Procedure 125 Confederates 125 Selection of Leaders 125 Selection of Co-workers, Secretary, and Interviewer 127 Training 128 Validity Checks on the Confederates' Performances 130 Participants 134 Experimental Setting 136 Experimental Procedure 137 Administration of the Dependent Measures 143 Demand Characteristics 143 CHAPTER III - RESULTS Overview of Statistical Analyses 146 Data Screening 149 Preliminary Analyses 152 Instrumentation for the Independent Variables 152 Instrumentation for the Dependent Variables 153 Task Performance 154 Task Adjustment 156 Interpersonal Adjustment 159 Manipulation Checks 160 Univariate ANOVAs 164 Post Hoc Analyses 170 Leadership Style 170 Group Productivity 172 Leadership Style x Group Productivity 172 MANOVA 176 Discriminant Analysis 179 Leadership Style 179 Group Productivity 183 Leadership Style x Group Productivity 186 Supplemental Analyses 189 Individual Differences 189 Participant Gender 190 Performance of the Optional Task 191 viii T A B L E O F C O N T E N T S ( c o n t i n u e d ) P a g e C H A P T E R IV - D I S C U S S I O N O F R E S U L T S T e s t s o f H y p o t h e s e s B a s e d o n U n i v a r i a t e A n a l y s e s 192 A n O v e r v i e w 192 L e a d e r s h i p S t y l e 194 T a s k P e r f o r m a n c e a n d T a s k A d j u s t m e n t 194 A d j u s t m e n t to t h e L e a d e r 205 G r o u p P r o d u c t i v i t y 213 T a s k P e r f o r m a n c e 213 A d j u s t m e n t to t h e T a s k a n d to t h e G r o u p 216 L e a d e r s h i p S t y l e x G r o u p P r o d u c t i v i t y 219 N o n s i g n i f i c a n t R e s u l t s 227 M u l t i v a r i a t e A n a l y s e s 229 L e a d e r s h i p S t y l e 234 G r o u p P r o d u c t i v i t y 236 L e a d e r s h i p S t y l e x G r o u p P r o d u c t i v i t y 237 S u p p l e m e n t a l R e s u l t s 239 I n d i v i d u a l D i f f e r e n c e s 239 P a r t i c i p a n t G e n d e r 239 P e r f o r m a n c e o f t h e O p t i o n a l T a s k 240 C H A P T E R V - C O N C L U S I O N A S u m m a r y o f t h e P r e s e n t S t u d y 241 I m p l i c a t i o n s a n d S p e c u l a t i o n s 248 L e a d e r s h i p S t y l e 249 C h a r i s m a t i c L e a d e r s h i p 249 S t r u c t u r i n g a n d C o n s i d e r a t e L e a d e r s h i p 253 C r o u p P r o d u c t i v i t y 256 V a l i d i t y I s s u e s 257 I n t e r n a l V a l i d i t y 258 E x t e r n a l V a l i d i t y 259 D i r e c t i o n s f o r F u t u r e R e s e a r c h 261 C o n c l u d i n g C o m m e n t s 268 R E F E R E N C E S 269 A P P E N D I C E S 297 i x L I S T O F T A B L E S T a b l e D e s c r i p t i o n P a g e 1 E x p e r i m e n t a l D e s i g n 99 2 O p e r a t i o n a l D e f i n i t i o n o f L e a d e r s h i p S t y l e s 103 3 O p e r a t i o n a l D e f i n i t i o n o f C r o u p P r o d u c t i v i t y 109 4 E x p e r i m e n t a l P r o c e d u r e 124 5 S c a l e C h a r a c t e r i s t i c s , M e a n s , S t a n d a r d D e v i a t i o n s a n d R e l i a b i l i t i e s f o r t h e D e p e n d e n t a n d I n d i v i d u a l D i f f e r e n c e M e a s u r e s 161 6 P e a r s o n C o r r e l a t i o n C o e f f i c i e n t s f o r t h e D e p e n d e n t M e a s u r e s 162 7 M e a n s a n d S t a n d a r d D e v i a t i o n s f o r t h e D e p e n d e n t M e a s u r e s 165 8 C e l l M e a n s a n d S t a n d a r d D e v i a t i o n s f o r t h e D e p e n d e n t M e a s u r e s 166 9 A n a l y s e s o f V a r i a n c e S u m m a r y T a b l e f o r t h e D e p e n d e n t , M e a s u r e s 167 10 M u l t i v a r i a t e A n a l y s i s o f V a r i a n c e S u m m a r y T a b l e f o r t h e D e p e n d e n t M e a s u r e s 178 11 S t a n d a r d i z e d D i s c r i m i n a n t F u n c t i o n C o e f f i c i e n t s f o r L e a d e r s h i p S t y l e 180 12 R o y - B o s e C o n f i d e n c e I n t e r v a l s f o r P a i r w i s e C o m p a r i s o n s o f t h e Mean S t a n d a r d i z e d D i s c r i m i n a n t S c o r e s 182 13 S t a n d a r d i z e d D i s c r i m i n a n t F u n c t i o n C o e f f i c i e n t s f o r C r o u p P r o d u c t i v i t y 184 14 S t a n d a r d i z e d D i s c r i m i n a n t F u n c t i o n C o e f f i c i e n t s f o r L e a d e r s h i p S t y l e b y C r o u p P r o d u c t i v i t y 187 LIST OF FIGURES Description Interaction Effect for Leadership Style and Group Productivity on Role Conflict Interaction Effect for Leadership Style and Croup Productivity on General Satisfaction Interaction Effect for Leadership Style and Group Productivity on Specific Satisfaction Typology of the Centroids in Discriminant Space for Leadership Style Typology of the Centroids in Discriminant Space for Group Productivity Typology of the Centroids in Discriminant Space for Leadership Style by Group Productivity xi L I S T O F A P P E N D I C E S A p p e n d i x P a g e A T h e I n - B a s k e t E x e r c i s e 297 B E x p e r i m e n t a l S c r i p t s f o r t h e S e c r e t a r y , L e a d e r s , 330 C o - W o r k e r s a n d I n t e r v i e w e r C L e a d e r s h i p S t y l e M a n i p u l a t i o n C h e c k s 345 D I n d i v i d u a l S e l f - R a t e d P e r f o r m a n c e S c a l e 349 E G e n e r a l S a t i s f a c t i o n S c a l e 352 F J o b D e s c r i p t i v e I n d e x : J o b S a t i s f a c t i o n S u b s c a l e 354 G R o l e A m b i g u i t y S c a l e 356 H Ro l e C o n f l i c t S c a l e 358 I J o b R e l a t e d T e n s i o n I n d e x 361 J A d j u s t m e n t to t h e L e a d e r S c a l e 364 K G r o u p A t m o s p h e r e S c a l e 366 L P r o c e s s M e a s u r e C o m p l e t e d b y t h e C o - w o r k e r s 368 M D i a r y C o m p l e t e d b y t h e P a r t i c i p a n t 374 N N e e d f o r A c h i e v e m e n t S c a l e 376 O T o l e r a n c e f o r A m b i g u i t y S c a l e 380 P N e e d f o r A f f i l i a t i o n S c a l e 382 Q S u m m a r y T a b l e s 386 Q1 P i l o t T e s t f o r D i f f e r e n c e s B e t w e e n A c t r e s s 1 a n d 387 A c t r e s s 2 o n t h e L e a d e r s h i p S t y l e M a n i p u l a t i o n C h e c k s Q2 P i l o t T e s t M a n i p u l a t i o n C h e c k s f o r t h e L e a d e r s h i p 388 S t y l e C o n d i t i o n s Q3 V a r i m a x R o t a t e d F a c t o r M a t r i x f o r t h e L e a d e r s h i p 389 S t y l e M a n i p u l a t i o n C h e c k s Q4 U n r o t a t e d F a c t o r M a t r i x f o r t h e S e l f - R a t e d 391 P e r f o r m a n c e S c a l e I tems Q5 V a r i m a x R o t a t e d F a c t o r M a t r i x f o r t h e T a s k 392 P e r f o r m a n c e D e p e n d e n t M e a s u r e s Q6 U n r o t a t e d F a c t o r M a t r i x f o r t h e Ro l e A m b i g u i t y 393 S c a l e I tems Q7 U n r o t a t e d F a c t o r M a t r i x f o r t h e Ro l e C o n f l i c t 394 S c a l e I tems Q8 U n r o t a t e d F a c t o r M a t r i x f o r t h e R o l e C o n f l i c t 395 S c a l e I tems 2 a n d 4 to 8 XII L I S T O F A P P E N D I C E S ( c o n t i n u e d ) A p p e n d i x Q9 U n r o t a t e d F a c t o r M a t r i x f o r t h e Ro l e C o n f l i c t S c a l e I tems 4 to 8 Q10 U n r o t a t e d F a c t o r M a t r i x f o r t h e J o b R e l a t e d T e n s i o n S c a l e I tems Q11 V a r i m a x R o t a t e d F a c t o r M a t r i x f o r t h e T a s k A d j u s t m e n t D e p e n d e n t M e a s u r e s Q12 U n r o t a t e d F a c t o r M a t r i x f o r t h e A d j u s t m e n t t o t h e L e a d e r S c a l e I tems Q13 U n r o t a t e d F a c t o r M a t r i x f o r t h e C r o u p A t m o s p h e r e S c a l e I tems Q14 V a r i m a x R o t a t e d F a c t o r M a t r i x f o r t h e I n t e r p e r s o n a l A d j u s t m e n t D e p e n d e n t M e a s u r e s Q15 S t u d e n t ' s t T e s t s f o r D i f f e r e n c e s B e t w e e n A c t r e s s 1 a n d A c t r e s s 2 on t h e L e a d e r s h i p S t y l e M a n i p u l a t i o n C h e c k s Q16 G r o u p M e a n s a n d S t a n d a r d D e v i a t i o n s f o r t h e L e a d e r s h i p S t y l e M a n i p u l a t i o n C h e c k s Q17 M a n i p u l a t i o n C h e c k s f o r t h e L e a d e r s h i p S t y l e C o n d i t i o n s Q18 M a n i p u l a t i o n C h e c k s f o r t h e H i g h a n d Low C r o u p P r o d u c t i v i t y C o n d i t i o n s Q19 A S u m m a r y T a b l e o f N e w m a n - K e u l s P o s t Hoc T e s t s f o r S i g n i f i c a n t L e a d e r s h i p S t y l e M a i n E f f e c t s Q20 A S u m m a r y T a b l e o f N e w m a n - K e u l s P o s t Hoc T e s t s f o r S i g n i f i c a n t L e a d e r s h i p S t y l e b y C r o u p P r o d u c t i v i t y I n t e r a c t i o n s Q21 M u l t i v a r i a t e A n a l y s i s o f C o v a r i a n c e S u m m a r y T a b l e f o r t h e D e p e n d e n t M e a s u r e s Q22 S t u d e n t ' s t T e s t s f o r D i f f e r e n c e s B e t w e e n Ma le a n d Fema le P a r t i c i p a n t s ' P e r c e p t i o n s o f t h e L e a d e r s h i p S t y l e s Q23 F r e q u e n c y T a b l e f o r P a r t i c i p a n t s ' C o m p l i a n c e W i t h t h e L e a d e r ' s - R e q u e s t to P e r f o r m an O p t i o n a l T a s k Q24 F r e q u e n c y T a b l e f o r P a r t i c i p a n t s ' C o m p l i a n c e W i t h t h e L e a d e r ' s R e q u e s t t o P e r f o r m an O p t i o n a l T a s k i n t h e P r e s e n c e o f C r o u p P r o d u c t i v i t y E f f e c t s x i i i L I S T O F A P P E N D I C E S ( c o n t i n u e d ) A p p e n d i x P a g e R R e c r u i t m e n t o f E x p e r i m e n t a l P a r t i c i p a n t s : A 414 D e s c r i p t i o n o f t h e M a n a g e m e n t T r a i n i n g P r o j e c t b y t h e C o u r s e C o - o r d i n a t o r S P h o t o g r a p h s o f t h e E x p e r i m e n t a l S e t t i n g 417 T D e m o g r a p h i c D a t a F o r m 421 U I n t e r v i e w Q u e s t i o n s 423 x i v A C K N O W L E D G E M E N T S I w i s h to e x p r e s s my a p p r e c i a t i o n to t h e f o l l o w i n g p e r s o n s , w h o s e t i m e , e x p e r t i s e , a n d s u p p o r t g r e a t l y c o n t r i b u t e d to t h e d e v e l o p m e n t a n d e x e c u t i o n o f t h i s t h e s i s . E s p e c i a l l y , t o P r o f e s s o r P e t e r J . F r o s t , t h e c h a i r m a n o f my t h e s i s c o m m i t t e e , f o r a l w a y s b e i n g t h e r e . . . w i t h e n c o u r a g e m e n t , u n d e r s t a n d i n g , a n d i n s p i r a t i o n . P e t e r ' s t i m e , d e d i c a t i o n , c r e a t i v e e n e r g i e s , a n d c o p i o u s k n o w l e d g e o f t h e r e s e a r c h p r o c e s s e n s u r e d t h i s d i s s e r t a t i o n e x p e r i e n c e was v e r y s p e c i a l a n d v e r y m e a n i n g f u l f o r me. T o P r o f e s s o r s D o n G . D u t t o n , G e r r y G o r n , a n d V a n c e F. M i t c h e l l , t h e s i s c o m m i t t e e m e m b e r s , e a c h o f whom c o n t r i b u t e d i n a s p e c i a l w a y to my d o c t o r a l p r o g r a m m e a n d t h e s i s i n p a r t i c u l a r . T o D o n , a s my m e n t o r t h r o u g h o u t my u n i v e r s i t y e d u c a t i o n , f o r b e l i e v i n g i n me a n d e n c o u r a g i n g me t o p u r s u e my d o c t o r a t e . T o G e r r y , f o r t e a c h i n g me t h e a r t o f m e t h o d o l o g y ; h i s v a l u a b l e a n d s t i m u l a t i n g c o m m e n t s made me a p p r e c i a t e more f u l l y t h e s u b t l e t i e s a n d i n t r i c a c i e s o f e x p e r i m e n t a t i o n . T o V a n c e , f o r g u i d i n g me t h r o u g h my d o c t o r a l p r o g r a m m e a n d t h e s i s w i t h s a g e a d v i c e , r e a s s u r a n c e , a n d c a r i n g . T o P r o f e s s o r W a l t e r B o l d t f o r h i s p a t i e n t a n d e x c e l l e n t g u i d a n c e i n t h e d a t a a n a l y s e s . T o S a n k a r f o r h i s c o m p e t e n t a s s i s t a n c e i n a n a l y z i n g t h e d a t a . X V T o D a v i d C a w o o d f o r h i s w i l l i n g s p o n s o r s h i p o f t h e M a n a g e m e n t T r a i n i n g P r o j e c t . T o my a c t o r s a n d a c t r e s s e s f o r t h e i r d e d i c a t i o n to t h e p r o j e c t a n d t h e i r p r o f e s s i o n a l i s m i n e n a c t i n g t h e i r r o l e s . T o S h a u n T a u n e s s e e f o r h i s t e c h n i c a l e x p e r t i s e a n d a s s i s t a n c e i n v i d e o t a p i n g . T o A n n a C u r r i e , J e a n F i s h , B e t t y K e r s e y , a n d K e n W o y t a z f o r t h e i r c o m p e t e n t p r e p a r a t i o n o f t h e m a n u s c r i p t . T o t h e I n d u s t r i a l R e l a t i o n s M a n a g e m e n t f a c u l t y a n d P r o f e s s o r C h a r l e s B . W e i n b e r g f o r t h e s t i m u l a t i n g l e a r n i n g e x p e r i e n c e t h e y p r o v i d e d me t h r o u g h o u t my d o c t o r a l p r o g r a m m e . T o my h u s b a n d , C a m e r o n S p o o n e r , a n d my p a r e n t s , J o h n a n d D o r o t h y H o w e l l , f o r t h e i r e n c o u r a g e m e n t a n d s u p p o r t t h r o u g h o u t t h i s e n d e a v o u r . - 1 -C H A P T E R I S C O P E O F T H E S T U D Y A N D L I T E R A T U R E R E V I E W I n t r o d u c t i o n C h a r i s m a t i c a n d t r a n s f o r m a t i o n a l l e a d e r s h i p h a v e r e c e n t l y e m e r g e d as p o t e n t i a l l y i m p o r t a n t c o n c e p t s a m o n g o r g a n i z a t i o n a l l e a d e r s h i p s c h o l a r s . J a m e s M a c C r e g o r B u r n s ( 1 9 7 8 ) , f o r e x a m p l e , d i s t i n g u i s h e s b e t w e e n e x c h a n g e o r i e n t e d t r a n s a c t i o n a l l e a d e r s w h o r e w a r d f o l l o w e r s f o r r e a c h i n g e s t a b l i s h e d o b j e c t i v e s a n d t r a n s f o r m a t i o n a l l e a d e r s w h o i n s p i r e f o l l o w e r s to t r a n s c e n d t h e i r i n t e r e s t s f o r s u p e r o r d i n a t e g o a l s . In t h e i r b e s t s e l l i n g b o o k . In S e a r c h o f E x c e l l e n c e , P e t e r s a n d W a t e r m a n (1982) o b s e r v e t h a t a t some p o i n t i n t h e i r h i s t o r y s u c c e s s f u l l y m a n a g e d c o m p a n i e s h a v e h a d t r a n s f o r m a t i o n a l l e a d e r s a t t h e he lm to i n s t i l l p u r p o s e , s h a p e v a l u e s , a n d e n g e n d e r e x c i t e m e n t . T r i c e a n d B e y e r (1984) p r e s e n t a r i c h d e s c r i p t i o n o f t h e i n s t i t u t i o n a l i z a t i o n o f c h a r i s m a in t w o s o c i a l m o v e m e n t o r g a n i z a t i o n s : A l c o h o l i c s A n o n y m o u s a n d t h e N a t i o n a l C o u n c i l on A l c o h o l i s m . A n i n - d e p t h e x p l o r a t i o n o f t h e e v o l u t i o n o f c h a r i s m a t i c p o l i t i c a l l e a d e r s i n t h e t w e n t i e t h c e n t u r y i s p r o v i d e d b y W i l l n e r ( 1 9 8 4 ) . F i n a l l y , B a s s ( 1 9 8 5 ) , in h i s r e c e n t b o o k L e a d e r s h i p a n d P e r f o r m a n c e B e y o n d E x p e c t a t i o n s , a r g u e s t h a t t r a n s f o r m a t i o n a l l e a d e r s h i p is n e c e s s a r y t o p r o m o t e f o l l o w e r p e r f o r m a n c e b e y o n d t h e o r d i n a r y ' l i m i t s . T h e p u r s u i t o f e m p i r i c a l k n o w l e d g e r e g a r d i n g t h e n a t u r e a n d e f f e c t s o f c h a r i s m a t i c l e a d e r s h i p i n o r g a n i z a t i o n s may b e j u s t i f i e d on s e v e r a l - 2 -g r o u n d s . F i r s t , t h e r e i s an i n c r e a s e d r e c o g n i t i o n o f t h e c r i t i c a l n e e d f o r o r g a n i z a t i o n a l c h a n g e a n d s c h o l a r s n e e d to e x p l o r e a l l p o s s i b l e m e c h a n i s m s to p r o d u c e t h i s c h a n g e ( K a n t e r , 1983; T i c h y & U l r i c h , 1984; T r i c e & B e y e r , 1 984 ) . D r a w i n g on t h e w r i t i n g s o f M a x W e b e r ( 1 9 4 7 ) , c h a r i s m a t i c a u t h o r i t y r e p r e s e n t s a p o t e n t m e c h a n i s m f o r c h a n g e . In f a c t , some c o n t e m p o r a r y t h e o r i s t s c o n t e n d t h a t o r g a n i z a t i o n s a n d o r g a n i z a t i o n a l l e a d e r s w h o a r e a b l e to c a p t u r e t h e c h a r i s m a t i c l o y a l t y a n d d e v o t i o n o f t h e i r m e m b e r s a n d f o l l o w e r s w o u l d a p p e a r to h a v e a s t r o n g a d v a n t a g e in t h e a t t a i n m e n t o f o r g a n i z a t i o n a l s u c c e s s ( e . g . , B e n n i s , 1982; B e n n i s & N a n u s , 1985; O b e r g , 1972; Z a l e z n i k , 1 983 ) . S e c o n d , c h a r i s m a may be r e g a r d e d as a f a c e t o f t h e e x p r e s s i v e a n d c u l t u r a l c o m p o n e n t s o f o r g a n i z a t i o n a l l i f e ( T r i c e & B e y e r , 1 984 ) . B y c r e a t i n g n e w s e t s o f m e a n i n g s , v a l u e s , a n d b e l i e f s f o r t h e i r f o l l o w e r s , c h a r i s m a t i c l e a d e r s may c o n s t r u c t a n d s h a p e new c u l t u r e s ( e . g . , M o o r e & B e c k , 1984; P e t e r s S W a t e r m a n , 1982; P e t t i g r e w , 1979; S a s h k i n £ F u l m e r , 1985; S c h e i n , 1985; S e l z n i c k , 1957; S m i r c i c h S M o r g a n , 1982; T i c h y S U l r i c h , 1984; T r i c e & B e y e r , 1 9 8 4 ) . F i n a l l y , c h a r i s m a t i c l e a d e r s h a v e r e c e i v e d w i d e s p r e a d a t t e n t i o n i n t h e a c a d e m i c l i t e r a t u r e a n d p o p u l a r p r e s s . T h e y h a v e b e e n f o u n d e r s o f s o c i a l m o v e m e n t o r g a n i z a t i o n s p r o m o t i n g r a d i c a l , l a r g e s c a l e s o c i e t a l c h a n g e ( e . g . , M a h a t m a C h a n d i , M a r t i n L u t h e r K i n g ) a n d h a v e e m e r g e d in p o l i t i c a l a r e n a s ( e . g . . F r a n k l i n De f ano R o o s e v e l t , F i d e l C a s t r o ) , i n r e l i g i o u s s p h e r e s ( e . g . , J i m J o n e s ) , i n m i l i t a r y s e t t i n g s ( e . g . . G e n e r a l G e o r g e P a t t o n ) , i n e d u c a t i o n a l i n s t i t u t i o n s ( e . g . . P r e s i d e n t M o r g a n o f A n t i o c h C o l l e g e ) , a n d i n b u s i n e s s o r g a n i z a t i o n s ( e . g . , Lee l a c o c c a , M a r y K a y A s h , J o h n D e L o r e a n ) ( e . g . , B a s s , 1985; T r i c e & B e y e r , 1984 ) . - 3 -D e s p i t e t h e a p p a r e n t p o t e n c y a n d p e r v a s i v e n e s s o f t h e c h a r i s m a t i c p h e n o m e n o n in s o c i a l l i f e , i t h a s r e m a i n e d a l a r g e l y u n e x p l o r e d c o n c e p t e m p i r i c a l l y . A l t h o u g h t h e r e ha s b e e n some d e b a t e a s t o w h e t h e r c h a r i s m a t i c e f f e c t s c o u l d be c a p t u r e d o p e r a t i o n a l l y a n d r e p r o d u c e d i n t h e l a b o r a t o r y o r i n a s i m u l a t i o n , s e v e r a l r e s e a r c h e r s h a v e a r g u e d t h a t we s h o u l d be a b l e to d i s t i l l a n d s t u d y some o f t h e e l e m e n t s in t h e d y n a m i c i n t e r a c t i o n o f s u c h l e a d e r s w i t h t h e i r f o l l o w e r s ( B a s s , 1981; H o u s e , 1977; H . W e i s s , p e r s o n a l c o m m u n i c a t i o n , M a r c h 16, 1984 ) . A c c o r d i n g l y , t h e p r e s e n t s t u d y a t t e m p t e d to a s s e s s t h e i m p a c t o f c h a r i s m a t i c l e a d e r s h i p , r e l a t i v e to t h e t r a d i t i o n a l l y r e s e a r c h e d d i m e n s i o n s o f c o n s i d e r a t e a n d s t r u c t u r i n g l e a d e r b e h a v i o u r s , on i n d i v i d u a l s ' a d j u s t m e n t a n d p e r f o r m a n c e i n a new w o r k s e t t i n g . L e a d e r s h i p d o e s n o t o c c u r i n a s o c i a l v a c u u m ; i n t e r d e p e n d e n c e e x i s t s b e t w e e n l e a d e r s a n d m e m b e r s o f g r o u p s ( S c h r i e s h e i m , M o w d a y , & S t o g d i l l , 1979, p . 1 0 7 ) . " T h e g r o u p ' s d e f i n i t i o n o f i t s t a s k , g o a l s , a n d p a t h s to i t s g o a l s s t r o n g l y a f f e c t s w h a t a l e a d e r c a n a c c o m p l i s h i n t h e g r o u p . In t u r n , t h e l e a d e r o f t e n h a s a n i m p a c t o n g r o u p o u t c o m e s b y i n f l u e n c i n g t h e g r o u p ' s n o r m s a n d g o a l s " ( B a s s , 1981 , p . 4 2 9 ) . T h e r e f o r e , t h e s o c i a l c o n t e x t o f l e a d e r - g r o u p r e l a t i o n s n e e d s to be r e c o g n i z e d in a n a l y s e s o f l e a d e r s h i p e f f e c t i v e n e s s . R e c e n t l e a d e r s h i p l i t e r a t u r e h a s i n v e s t i g a t e d s e v e r a l d i m e n s i o n s o f l e a d e r - g r o u p r e l a t i o n s i n c l u d i n g c o h e s i v e n e s s , d r i v e , a r o u s a l , a n d p e r f o r m a n c e ( G r e e n e & S c h r i e s h e i m , 1980; P o d s a k o f f & T o d o r , 1985; S c h r i e s h e i m , 1 980 ) . H o w e v e r , t h e n o r m a t i v e s t r u c t u r e o f g r o u p s a n d i t s i n f l u e n c e o n m e m b e r s ' a t t i t u d e s a n d p e r f o r m a n c e w i t h i n t h e l e a d e r - g r o u p c o n t e x t r e m a i n s to be e x p l o r e d . _ n -A p a r t i c u l a r l y c r i t i c a l r o l e o f t h e g r o u p is to a c t a s a n o r m a t i v e r e f e r e n t f o r a p p r o p r i a t e t y p e s o f w o r k b e h a v i o u r , e s p e c i a l l y t h e l e v e l o f p r o d u c t i v i t y . G r o u p s c a n a c t i n c o n c e r t w i t h a l e a d e r ' s a ims a n d o b j e c t i v e s b y a d v o c a t i n g h i g h w o r k p r o d u c t i v i t y o r a g a i n s t t h e s e a ims a n d o b j e c t i v e s b y e n c o u r a g i n g low w o r k p r o d u c t i v i t y . F u r t h e r , g r o u p e f f e c t s on m e m b e r p e r f o r m a n c e may be a u g m e n t e d , w e a k e n e d , o r u n c h a n g e d b y t h e l e a d e r ' s i m p a c t . T h u s t h e i n t e r a c t i v e i n f l u e n c e o f l e a d e r s h i p s t y l e a n d g r o u p p r o d u c t i v i t y n o r m s on i n d i v i d u a l s ' a d j u s t m e n t a n d p e r f o r m a n c e was a l s o e x a m i n e d i n t h i s s t u d y . F i n a l l y , o ne o u t c o m e t h a t i s t y p i c a l l y e x a m i n e d i n l e a d e r s h i p r e s e a r c h is i n d i v i d u a l p e r f o r m a n c e . In b o t h l a b o r a t o r y a n d f i e l d s t u d i e s o f l e a d e r - s u b o r d i n a t e i n t e r a c t i o n s , r e s e a r c h e r s h a v e f o c u s e d o n b o t h t h e q u a n t i t a t i v e a n d q u a l i t a t i v e a s p e c t s o f j o b p e r f o r m a n c e ( e . g . , C i l m o r e , B e e h r & R i c h t e r , 1979; G r e e n e , 1979; K a t z , 1977; L o w i n , H r a p c h a k , & K a v a n a g h , 1969; S c h r i e s h e i m , M o w d a y , & S t o g d i l l , 1979 ) . A n e q u a l l y i m p o r t a n t o u t c o m e is i n d i v i d u a l s ' a d j u s t m e n t to t h e d e m a n d s o f a new r o l e . G i v e n t h e h i g h l e v e l s o f l a b o u r m o b i l i t y , t e c h n o l o g i c a l c h a n g e , o r g a n i z a t i o n a l r e t r e n c h m e n t s , a n d m e r g e r s a n d a c q u i s i t i o n s , w o r k r o l e t r a n s i t i o n s h a v e become a p r e v a l e n t a n d i m p o r t a n t a s p e c t o f l i f e i n m o d e r n o r g a n i z a t i o n s ( e . g . , L o u i s , 1980a; N i c h o l s o n , 1 984 ) . T h e r e f o r e , i t wa s c o n s i d e r e d n e c e s s a r y a n d a p p r o p r i a t e t o e x a m i n e b o t h i n d i v i d u a l a d j u s t m e n t a n d p e r f o r m a n c e o u t c o m e s i n t h i s i n v e s t i g a t i o n o f l e a d e r a n d g r o u p b e h a v i o u r . - 5 -O v e r v i e w o f t h e P r e s e n t S t u d y T h e p u r p o s e o f t h e p r e s e n t s t u d y was to e x a m i n e t h e e f f e c t s o f t h r e e l e a d e r s h i p s t y l e s ( i . e . , c h a r i s m a t i c , s t r u c t u r i n g , a n d c o n s i d e r a t e ) a n d two l e v e l s o f g r o u p p r o d u c t i v i t y ( i . e . , low g r o u p p r o d u c t i v i t y a n d h i g h g r o u p p r o d u c t i v i t y ) on p a r t i c i p a n t s ' a d j u s t m e n t to a n d p e r f o r m a n c e on a n a m b i g u o u s d e c i s i o n m a k i n g t a s k . S p e c i f i c a l l y , t h i s s t u d y a d d r e s s e d t h e f o l l o w i n g q u e s t i o n : How d o l e a d e r s h i p s t y l e a n d g r o u p p r o d u c t i v i t y n o r m s f a c i l i t a t e i n d i v i d u a l s ' a d j u s t m e n t to a n d p e r f o r m a n c e i n a new o r g a n i -z a t i o n a l s e t t i n g ? T h e s i g n i f i c a n c e o f t h i s q u e s t i o n i s t h r e e f o l d . F i r s t , t h e p r e s e n t s t u d y r e p r e s e n t s a p i o n e e r i n g a t t e m p t to i n v e s t i g a t e t h e p h e n o m e n o n o f c h a r i s m a t i c l e a d e r s h i p i n a l a b o r a t o r y s e t t i n g . S e c o n d , t h e s o c i a l c o n t e x t o f l e a d e r - f o l l o w e r r e l a t i o n s i s r e c o g n i z e d b y i n c l u d i n g g r o u p m e m b e r s . F i n a l l y , b o t h t a s k a n d i n t e r p e r s o n a l a d j u s t m e n t , u n d e r s t u d i e d a s p e c t s o f o r g a n i z a t i o n a l l i f e , a r e e x a m i n e d i n a d d i t i o n to p e r f o r m a n c e . In t h e p r e s e n t s t u d y . C o m m e r c e a n d B u s i n e s s A d m i n i s t r a t i o n u n d e r g r a d u a t e s w e r e r e c r u i t e d to p a r t i c i p a t e i n a p r o j e c t a s s e s s i n g t h e i r p r a c t i c a l b u s i n e s s s k i l l s . T h e p a r t i c i p a n t s c o m p l e t e d an i n - b a s k e t e x e r c i s e u n d e r t h e d i r e c t i o n o f t h e p r o j e c t m a n a g e r ( a n e x p e r i m e n t a l c o n f e d e r a t e ) w h o p o r t r a y e d a c h a r i s m a t i c , s t r u c t u r i n g , o r c o n s i d e r a t e l e a d e r s h i p s t y l e . T h e p a r t i c i p a n t s w o r k e d on t h e e x e r c i s e i n t h e p r e s e n c e o f two o t h e r s t u d e n t s ( a l s o e x p e r i m e n t a l c o n f e d e r a t e s ) w h o a d v o c a t e d e i t h e r low o r h i g h p r o d u c t i v i t y on t h e t a s k . A t t h e c o n c l u s i o n o f t h e e x e r c i s e , t h e - 6 -participants rated their adjustment to the task, the manager, and the other students. Organization of the Dissertation The organization of this dissertation is as follows. The remainder of Chapter I is composed of six sections. The first section reviews the literature relevant to charismatic, structuring, and considerate leadership styles. The group productivity dimension - low group productivity and high group productivity - is subsequently examined. The third section outlines the use of qualitative and quantitative task performance as outcome measures. In the fourth section, two components of adjustment -task and interpersonal - are described. The fifth section presents definitions of the terms used in the study. In the sixth section, hypotheses concerning leadership style and group productivity are deduced from the previously reviewed theories and empirical research. Chapter II outlines the experimental design and procedure. Findings of the study are described in Chapter III. Chapter IV discusses the results in light of previous research. Implications and speculations regarding the study's findings, validity issues, and directions for future research are presented in Chapter V. - 7 -A L I T E R A T U R E R E V I E W O F L E A D E R S H I P S T Y L E O v e r v i e w T h e s t u d y o f o r g a n i z a t i o n a l l e a d e r s h i p h a s e v o l v e d o v e r m a n y y e a r s ( B a s s , 1981; C h e m e r s , 1984; H u n t , 1984; Y u k l , 1 981 ) . In o r d e r to p r o v i d e a c o n t e x t u a l u n d e r s t a n d i n g o f t h e l e a d e r b e h a v i o u r s e x a m i n e d in t h i s s t u d y , t h e e v o l u t i o n o f o r g a n i z a t i o n a l l e a d e r s h i p a p p r o a c h e s i s b r i e f l y t r a c e d . T h e e a r l i e s t l e a d e r s h i p a p p r o a c h f o c u s e d on t r a i t s . It was p r i m a r i l y c o n c e r n e d w i t h t h e i d e n t i f i c a t i o n o f t r a i t s t h a t d i s c r i m i n a t e d b e t w e e n l e a d e r s a n d n o n l e a d e r s , e f f e c t i v e l e a d e r s a n d i n e f f e c t i v e l e a d e r s , o r l e a d e r s a t h i g h e c h e l o n s in t h e o r g a n i z a t i o n as c o m p a r e d to t h o s e a t l o w e r e c h e l o n s ( C h e m e r s , 1984; H o u s e & B a e t z , 1979; S t o g d i l l , 1948, 1 974 ) . Wh i l e ma jo r r e v i e w s o f t r a i t r e s e a r c h f a i l e d to s u b s t a n t i a t e t h e p r e m i s e t h a t c e r t a i n s a l i e n t l e a d e r c h a r a c t e r i s t i c s w e r e e s s e n t i a l f o r e f f e c t i v e l e a d e r s h i p ( e . g . , G i b b , 1969; M a n n , 1959; S t o g d i l l , 1948, 1 9 7 4 ) , H o u s e a n d B a e t z (1979 , p . 349 ) o b s e r v e d t h a t s e v e r a l t r a i t s s h o w e d c o n s i s t e n t l y h i g h a s s o c i a t i o n s w i t h l e a d e r s h i p i n c l u d i n g i n t e l l i g e n c e , d o m i n a n c e , s e l f c o n f i d e n c e , e n e r g y , a n d t a s k r e l e v a n t k n o w l e d g e . M o r e r e c e n t w r i t i n g s h a v e r e c o g n i z e d t h a t t h e v a r i a b i l i t y o f g r o u p s i t u a t i o n s n e c e s s i t a t e s d i f f e r e n t l e a d e r s h i p a p p r o a c h e s ( K e n n y £ Z a c c a r o , 1983; S c h n e i d e r , 1985 ) . T h u s l e a d e r s ' a b i l i t y to a c c u r a t e l y p e r c e i v e v a r i a t i o n s in g r o u p s i t u a t i o n s a n d to a d j u s t t h e i r b e h a v i o u r a c c o r d i n g l y may b e mo re i m p o r t a n t t h a n t r a d i t i o n a l p e r s o n a l i t y t r a i t s i n p r o d u c i n g l e a d e r e f f e c t i v e n e s s ( K e n n y & Z a c c a r o , 1 983 ) . - 8 -When it became evident that no stable and situationally invariant personality characteristics identified leaders, researchers explored the behavioural correlates of effective leadership. Studies conducted by the leadership group at Ohio State University identified two dimensions of leader behaviour: consideration which reflected interpersonal warmth, concern for feelings of subordinates, and use of participative two-way communication and initiation of structure which reflected directives, goal facilitation, and task related feedback (e.g., Fleishman & Peters, 1962). In a widely cited review of studies investigating the relationship between these leader behaviour patterns and group outcomes such as productivity and satisfaction, mixed results were reported (Korman, 1966). As an outgrowth of dissatisfaction with the behavioural approach, contingency leadership models were developed. These models suggest that desired subordinate behaviour can be elicited by adopting the appropriate leadership style or by altering contingencies in the work environment to make them compatible with a particular leadership style. Fiedler's (1967) contingency theory. House's (1971) path-goal theory, Vroom and Yetton's (1973) normative decision theory, Hersey and Blanchard's (1977) situational leadership theory, and Yukl's (1981) multiple linkage model of leadership effectiveness are representative of current contingency models. The contingency paradigm has been the dominant approach among leadership researchers since the late 1960s (Hunt, 1984, p.114). In recent years, however, researchers have increasingly questioned the adequacy of this paradigm. Some argue that the paradigm is reductionistic, static, and sterile; it is an incomplete representation of the numerous other - 9 -components of the leadership process (McCall & Lombardo, 1978). As Mintzberg (1973), Weick (1978) and others (e.g., Mitroff, 1978; V a i l l , 1978) have observed, the task of leadership is characterized by variety, complexity, and fragmentation. Moreover, leaders' actions are influenced by the wider environmental context within which they operate, including the economy, technology, labour unions and so on, as well as by the myths and traditions which permeate the organization (McCall & Lombardo, 1978). Other criticisms of the contingency model concern its variable predictive accuracy in terms of group and individual outcomes (Hunt, 1984). In response to these criticisms, alternative approaches to studying organizational leadership have been proposed. For example, there is increasing recognition that leadership is a reciprocal influence process involving oscillating cycles of interaction between leaders and followers (e.g., Fulk & Cummings, 1984; Wofford 6 Srinivasan, 1983; Zahn & Wolf, 1981). The vertical dyad linkage theory developed by George Graen and his associates (e.g., Dansereau, Graen, & Haga, 1975; Graen & Cashman, 1975) is illustrative. T h i s theory suggests that leaders develop different types of exchange relationships with different subordinates. This reciprocal exchange notion of leadership is further reflected in operant conditioning theories of leadership (e.g., Mawhinney £ Ford, 1977; Scott, 1977; Sims, 1977). These theories propose that the leader structures reinforcement contingencies in the work environment such that subordinates engage in appropriate responses to stimuli in order to elicit desired consequences or avoid undesirable consequences (Hunt, 1984). Subordinates' responses subsequently influence the leader's alteration of reinforcement contingencies (Sims, 1977). - 10 -Other researchers have recognized the need to extend the focus from the leader-follower dyad to the social network of peers, superiors, and subordinates within which the leader is embedded (e.g.. Cast, 1984; Salancik, Calder, Rowland, Leblebici, S Conway, 1975; Stewart, 1982; T s u i , 1984). For instance, the work of T s u i (1984) suggests that in order to gain reputational effectiveness, the leader needs to be responsive to multiple constituencies' role expectations. Similarly, Gast (1984) argues while discretionary actions enable leaders to mobilize their authority and personal power to influence others, such actions are restricted by social and political processes, organizational policies, structural limitations, resource constraints and so on. Another group of leadership scholars has focused on expanding the contingencies considered in leadership models (e.g.. Hunt S Osborn, 1982; T o s i , 1982). These "second generation" contingency models incorporate both macro and micro environmental variables that jointly affect superior-subordinate performance (Hunt, 1984). For example. Hunt and Osborn's (1982) multiple influence model of leadership asserts that environmental, contextual (size and technology), and structural complexity affects required and discretionary leader behaviours and operates in combination with such behaviours to influence work unit performance and satisfaction related outcomes. Despite these suggested refinements and extensions of the current contingency leadership paradigm, some scholars contend that we are overlooking the essence of the leadership phenomenon and have called for a radically different perspective, in Hunt's (1984) terms, for a paradigm - 11 -shift. One controversial perspective, based on attribution theory, suggests that leadership is essentially a perceptual construct; it is an inference based on observed behaviour accepted as evidence of leadership (e.g., Calder, 1977; Lord & Smith, 1983; McElroy, 1982; Meindl, Ehr l i c h , & Dukerich, 1985; Pfeffer, 1977). For example, Meindl and his colleagues (1985) have recently argued that due to biased preferences to understand important but causally indeterminant and ambiguous organizational events and outcomes, individuals have a tendency to ascribe excessively high levels of control and influence to leaders. Accordingly, a highly romanticized, heroic view of leadership has evolved. However, Meindl et al.'s (1985) views need to be reconciled with convincing empirical evidence suggesting that leaders can have substantive effects on organizational outcomes (e.g.. Smith, Carson, 6 Alexander, 1984; Weiner S Mahoney, 1981). Specifically, Weiner and Mahoney (1981), in their study of 193 manufacturing organizations over a 19 year period, reported that leadership accounted for more variance in organizational performance than did several environmental or organizational factors. More recently. Smith and his associates (1984) presented longitudinal data supportive of the hypothesis that effective leaders have a positive impact on organizational performance. The symbolic aspects of leadership are illustrative of the other recent thrust in organizational leadership. T h i s perspective suggests that the import of leadership lies in the ability to shape meanings and interpre-tations of organizational events and activities. Through the use of language, rituals, drama, stories, myths, and other symbolic forms, leaders develop social consensus around the activities being undertaken - 12 -and produce organized collective action (e.g., Pettigrew, 1979; Pfeffer, 1981; Pondy, 1978; Pondy, Frost, Morgan, & Dandridge, 1983; Siehl & Martin, 1^ 984; Smircich S Morgan, 1982). In particular, charismatic leaders are especially skilled in creating and managing meanings in organizations through their use of evocative imagery, compelling visions, expressive language, and dramaturgical ski l l s . To conclude, the traditional contingency approach, which often incorporates the dimensions of consideration and initiating structure or their analogues, has become a recognized and entrenched view of organi-zational leadership and offers a point of comparison for newer perspectives such as charisma. Accordingly, in the present study, charismatic leadership was examined in the context of the traditionally researched dimensions of considerate and structuring leader behaviour. The following sections review the theoretical literature and empirical research related to charismatic, s t r u c t u r i n g , and considerate leadership styles and their effects on individual performance in and adjustment to a new work setting. Theory and Research Related to Charismatic Leadership Overview A recent and often repeated theme in the leadership literature is that researchers have focused on the more mundane, readily observable leader-- 13 -subordinate relations and have ignored the profound aspects of leadership to be seen in the charismatic movers and shakers of our time (e.g., Avolio & Bass, 1985; Bass, 1981, p.609, 1985; Dubin, 1979; McCall & Lombardo, 1978; T o s i , 1982). Certainly one finds copious references in both the academic literature and popular press to the pervasive influence of charis-matic leaders in formal organizations and in society. To date, however, no scholarly consensus has emerged on the precise application of the concept of charisma (Tucker, 1970). Moreover the term charisma has been indiscriminantly applied to any individual who is immensely popular or who possesses personal charm. Such erroneous application of the term serves to dilute its potential, to obscure its meaning, and to debase its powerful effects (Apter, 1968). In addition, there is a paucity of integrated theoretical work and empirical studies on charismatic leadership. Charisma has been examined as a significant factor in historical and social change; its function in religious and political spheres has been explored; and its psychological dimensions and psychoanalytic origins have been discussed. Despite the definitional ambiguity and fractionated inquiries, insights into the nature of charisma can be gleaned from existing theoretical and empirical literature from several different disciplines. The thrust of this dissertation is the examination of the behaviour and effects of intra-organizational charismatic leaders from a psychological perspective. However, in order to provide the appropriate background and context for the discussion of the psychological view, it is necessary to review the relevant theoretical and empirical literature from the disciplines of political science and sociology. A frui t f u l beginning for this review is Max Weber's seminal ideas on charismatic authority. Accordingly, these ideas are _ 14 -presented and evaluated below. Subsequently, an examination of the relevant literatures will be conducted. Weber's Conceptualization of Charismatic Authority Max Weber's conceptualization of charisma serves as the starting point for most writers concerned with charismatic leadership. In his classic writing. The Theory of Social and Economic Organization, Weber postulated three ideal types of legitimate authority: traditional, rational or legal, and charismatic. Traditional authority is derived from "an established belief in the sanctity of immemorial traditions and the legitimacy of the status of those exercising authority under them" (Weber, 1947, p.328). Rational or legal authority is based on an established "belief in the 'legality' of patterns of normative rules and the right of those elevated to authority under such rules to issue commands" (Weber, 1947, p.328). Thus authority flows from traditional custom or occupancy of an office; it is impersonal and stable. In contrast, charismatic authority breaks with tradition or rational norms; it is personal, dynamic, and revolutionary. Specifically, in charismatic authority, individuals are followed and obeyed because of a special trust they induce, their special powers, and their unique qualities. These individuals claim their authority not through enacted position or traditional dignity, but owing to gifts of grace (Tucker, 1970). According to Weber (1947, pp.358-359), charisma is: a certain quality of an individual personality by virtue of which he is set apart from ordinary men and treated as endowed with supernatural, superhuman, or at least specifically exceptional powers or qualities. These are such as are not accessible to the - 15 -ordinary person, but are regarded as of divine origin or as exemplary, and on the basis of them the individual concerned is treated as a leader. Weber believed that the rise of charismatic leaders was most apparent at times of crisis in which the basic values, institutions, and legitimacy of the society were in question. A major breakdown in social and political order increases the likelihood that people will feel helpless, disturbed, and fragmented and will therefore eagerly accept the authority of leaders who appear to be uniquely qualified to lead them out of their acute distress. This potential for salvation from distress creates the special emotional intensity of the charismatic response.... Followers respond to the charismatic leader with passionate loyalty because the salvation, or promise of it, that he appears to embody represents the fulfillment of urgently felt needs (Tucker, 1970, p.81). In particular, by espousing a transcendent goal, inspirational mission, or "explosively novel" innovation (Shils, 1965, p.199) as possible solutions for overcoming distressful conditions, charismatic leaders come to be regarded as saviours. While charisma may be partially attributed to the extraordinary qualities of individuals and to the context of their mission, it is also necessary for these individuals to be acknowledged as exceptional by a following. What is alone important is how the individual is actually regarded by those subject to charismatic authority, by his 'followers' or 'disciples', ... It is the recognition on the part of those subject to authority which is decisive for the validity of charisma (Weber, 1947, p.359). - 16 -Consequently the sole source of legitimate authority for charismatic leaders lies in the regard of their followers. Thus charisma is defined in terms of followers' perceptions of and responses to such leaders (e.g., Rustow, 1970; Tucker, 1970; Willner, 1984). Indeed, the maintenance of their leadership is dependent upon f u l f i l l i n g , in a relevant and acceptable manner, the expectations of their followers. Thus there is a continual need for charismatic leaders to prove that they are the chosen ones. In summary, according to Weber's conception, charisma consists of several interrelated components including extraordinary personal qualities of leaders; a social c r i s i s or situation of distress; a transcendent course of action or inspirational mission which offers the hope of salvation from the c r i s i s ; followers who, out of love, passionate devotion and enthusiasm, willingly subscribe to charismatic leaders and their missions; and leaders' repeated demonstration of their charismatic powers in order to maintain their followers' devotion. In addition, Weber stresses that the crucial test of charisma is the response and perceptions of the followers (Tucker, 1970). It is not what the leader is but what the followers perceive the leader to be that is essential in fostering the charismatic relationship (Willner, 1984). Weber (1947, p.364) acknowledges that " i f [charisma] is not to remain a purely transitory phenomenon, but to take on the character of a permanent relationship forming a stable community.. .it is necessary for the character of charismatic authority to become radically changed". Thus, through the process of routinization, charisma becomes modified and attributed to an office, lineage, or clan. Charisma is no longer pure - 17 -charisma in the original sense discussed above; it is transformed from an extraordinary and purely personal relationship into an established authority structure that is no longer necessarily dependent upon personal charismatic qualities in the incumbent leader (Tucker, 1970, p.91). Thus the routinization of charisma "provides the basis and legitimacy for newly established traditional or rational-legal hierarchies and systems of domination, which, in the process of routinization, become conservative" (Bensman £ Civant, 1975, p.580). An Evaluation of Weber's Conceptualization of  Charismatic Authority Weber's writings on charismatic authority have stimulated considerable discussion in the political science, religious, sociological, psychological, and organizational literatures. According to Tucker (1970), some scholars are impressed with its power or potentiality as a device for analyzing historical leadership situations. Others, however, have questioned the relevance of charisma to contemporary social life (e.g., Fr i e d r i c h , 1961; Lowenstein, cited in Tucker, 1970). For example, Friedrich (1961) has argued for a restrictive interpretation of charisma. Since the term originally meant leadership based on a transcendent call by a divine being, Friedrich (1961) contends that Weber's broadening of the term to include secular and nontranscendent types of callings is inappropriate. However, Tucker (1970) has persuasively argued for the secularization of the concept of charisma citing its great explanatory power in analyzing political and social situations. - 1 8 -Other critics of Weber's conceptualization of charisma point to the lack of clarity of the term and the difficulty in operationalizing and applying it in practice (e.g., Madsen & Snow, 1983; Tucker, 1970; Willner, 1984). Two closely related criticisms are of particular importance. F i r s t , on the basis of Weber's various formulations, it is difficult to differentiate between leaders who really are charismatic and leaders who are not (Ratnam, 1964; Tucker, 1970). Second, critics have observed that Weber provided no clear statement of the personal qualities in charismatic leaders which create the special emotional bond with their followers (Ratnam, 1964; Tucker, 1970, p.732). Thus the Weberian conceptualization of charisma leaves some doubt as to which leaders are charismatic, how they become so designated, and what it is exactly that makes them so. Another key criticism of the contemporary application of the concept of charisma to modern society is the conditions under which charisma may arise. Several scholars (e.g., Apter, 1968; Blau, 1963; Dow, 1969; Friedland, 1964; Willner, 1984) have argued that Weber did not adequately specify the elements fostering the genesis of charismatic authority. For example, Blau (1963, p.309) states that Weber's theory of charisma "encompasses only the historical processes that lead from charismatic movements to increasing rationalization and does not include an analysis of the historical conditions that give rise to charismatic eruptions in the social structure." A further and very fundamental criticism concerns Weber's proposition that charismatic authority ultimately becomes "either traditionalized or rationalized, or a combination of both" (Weber, 1947, p.364). Many - 1 9 -contemporary sociologists, psychologists, and organizational behaviourists argue that charismatic leadership can and does exist in formal complex organizations (e.g., Dow, 1969; Etzioni, 1961; House, 1977; Katz £ Kahn, 1978; Oberg, 1972; Parsons, 1937; Shils, 1965; Tucker, 1970). In particular, Edward Shils attempts to refocus the discussion of charisma by emphasizing the elements in charisma that link it to established orders. Concerning Weber, he argues: He did not consider the more widely dispersed, unintense operation of the charismatic element in corporate bodies governed by the rational-legal type of authority... .Weber had a pro-nounced tendency to segregate the object of attributed charisma, to see it almost exclusively in its most concentrated and intense forms, and to disregard the possibility of its dispersed and attenuated existence. He tended indeed to deny the possibility that charisma can become an integral element in the process of secular institutionalization (Shils, 1965, p.202). Thus Shils suggests that charisma is not confined to charismatic organ-izations but may in fact emerge in any type of organizational setting. This contention has been supported by empirical evidence demonstrating that charismatic authority exists in bureaucratic organizations (e.g., Avolio £ Bass, 1985; Roberts, 1984; Scott, 1978). Since Weber's initial writings, scholars from several disciplines have refined, elaborated, modified and, in some cases, challenged his views on charismatic authority. In the following sections three different bodies of literature - political science, sociology, and psychology - which have theoretically and empirically examined the charismatic phenomenon will be reviewed. - 20 -Political Science Views of Charisma The investigation of the charismatic phenomenon from a political science perspective has proceeded in several different directions. For example, the concept of charisma has been fruitfully applied in examinations of post-colonial and totalitarian regimes. Specifically, Weber's notion of charisma has been invoked in analyses of modernization and political development in ex-colonial new states (e.g., Apter, 1968; Rustow, 1970; Willner & Willner, 1965). The thrust of these analyses is the historical tracing of the critical role of charismatic leaders during the transition from colonial ruled traditional society to politically independent modern society. In addition, the concept of charisma has been applied in analyses of totalitarian regimes as illustrated in Tucker's (1970) study of Lenin as founder and leader of the Bolshevik revolutionary movement. More recently, Schweitzer (1984) has sought to explain the causes and consequences of successful charismatic leadership under two different political systems - democracy and dictatorship. Other political scientists have systematically investigated specific components of Weber's formulation such as situational events and the routinization of charisma. For example. Cell (1974), using a social structural model, isolated five situational factors to account for charisma among 34 twentieth century heads of state: national social c r i s i s , disruptive youth, denial of leader's access to following, prepower following, and nationalism and nationalistic movements. Madsen and Snow (1983) have focused on the stages in the process of routinization of charisma based on the case of Juan Peron and the Peronist movement in - 21 -Argentina. Two stages in the evolution of the bond between the charismatic leader and his/her mass following were identified: 1. The development of a structure within the movement and 2. The dispersion of a mass charismatic response across that structure and its higher level personnel. Another thrust of political science inquiry is in-depth analyses of the charismatic phenomenon utilizing charismatic political leaders in the twentieth century as rich illustrations. Of particular interest is the work of two political scientists: James MacGregor Burns and Ann Ruth Willner. In his incisive analysis of political leadership, Burns (1978) distinguishes between two types of leaders: transactional and transformational. T r a n s -actional political leaders "approach followers with an eye to exchanging one thing for another: jobs for votes, or subsidies for campaign contributions. Such transactions comprise the bulk of the relationships among leaders and followers, especially in groups, legislatures, and parties" (Burns, 1978, p.4). T hus the relationship between transactional leaders and their followers is entrenched in a bargaining process wherein both parties to the exchange pursue their related purposes. Th i s relationship is circum-scribed; it is maintained as long as the respective needs of leader and follower can be met through a reciprocal exchange of rewards for services provided. As Burns (1978, p.20) observes, while a leadership act has occurred, it is not one that "binds leader and follower together in a mutual and continuing pursuit of a higher purpose". - 22 -In contrast to transactional leadership. Burns (1978, p.4) posits that the transformational leader "looks for potential motives in followers, seeks to satisfy higher needs, and engages the full person of the follower. The result of transforming leadership is a relationship of mutual stimulation and elevation that converts followers into leaders and may convert leaders into moral agents". Therefore transformational leaders appeal to higher order values that encompass followers more fundamental and enduring needs (Burns, 1978, p. 42). Accordingly, followers' goals and aspirations transcend their immediate self interests and are focused on the collective purpose. A comprehensive and insightful analysis of the origins and develop-ment of charismatic political leadership has been recently presented by Willner (1984) in her book The Spellbinders. Using Weber's work as a point of departure, she focuses on the genesis of charismatic political leadership in the twentieth century. Seven charismatic political leaders are selected to illustrate the explanation for the charismatic phenomenon: Castro, Gandhi, Hitler, Mussolini, Roosevelt, Sukarno, and Khomeini. In contrast to Weber's formulation, it is argued that social crisis and psychic distress; doctrines, messages, or missions which restore a sense of purpose and offer a vision of the future; and followers' susceptibility to charismatic appeals and domination are neither necessary nor sufficient to catalyze charisma. Rather, the leader, through a combination of personal attributes and actions and mode of public presentation, can be an active initiator of charismatic perceptions. Four catalytic factors are outlined which foster charismatically oriented impressions. - 23 -F i r s t , charismatic political leaders fortuitously or intentionally invoke cultural myths that are "linked to its sacred figures, to its historical and legendary heroes, and to its historical and legendary ordeals and triumphs" (Willner, 1984, p.62). In so doing, charismatic leaders become associated in the hearts and minds of their followers with such illustrious cultural heroes. For example, Fidel Castro deliberately invoked the memory of Jose Marti, the long idolized and revered father of Cuban independence, through his speeches, symbols, and deeds (Willner, 1984, pp.72-74). In agreement with Weber, the second factor precipitating the charismatic image is the performance of an extraordinary or heroic feat. Willner (1984) describes a multitude of elements which contribute to perceptions of an act as extraordinary such as the apparent risk entailed, the existence of major obstacles, and the suspense surrounding the act. Benito Mussolini, for example, gained the reputation of a "fearless champion of Italian rights and the restorer of Italian honour and prestige" by his successful handling of international affairs (Willner, 1984, p.104). To illustrate, in the long standing dispute between the Italian and Yugoslavian governments regarding the control of Fiume, an Adriatic settlement, Mussolini attained the seemingly impossible: officially incorporating Fiume into Italy as well as striking an alliance with Yugoslavia (Willner, 1984). A thi r d dimension of charismatic legitimation is the projection of the possession of exceptional personal attributes. Willner (1984) argues while followers' perceptions of the superhuman endowments of charismatic political leaders may arise from leaders' actual manifestation of specific - 24 -supernatural attributes, they can also arise from followers' generalized notions of such extraordinary capabilities. That i s , a spillover effect may occur whereby charismatic leaders who become associated with historical or mythical heroes or who perform outstanding feats may also be credited with personal powers they do not possess or display (Willner, 1984, p.129). According to Willner (1984, p.130) Mahatma Gandhi is the best exemplar of a twentieth century leader who, through a combination of qualitites, a syndrome of attributes and actions, and a style of life that closely matches cultural notions of the ideal person, generated perceptions of superhuman endowments. His prodigious personal qualities of intellect, energy, stamina, composure, and self confidence; his remarkable accomplishments; and his exemplification of some of the highest ideals of Hinduism in his lifestyle collectively contributed to followers' beliefs in his outstanding powers. The final factor promoting charismatic perceptions is outstanding oratorical sk i l l s . The eloquent and spellbinding rhetoric employed by charismatic leaders can create incredible emotional fervour on the part of followers. According to Willner (1984), the use of figurative language, such as similes, metaphors, and analogies, and of rhetorical devices related to sound, such as rhythm, repetition, and alliteration, are strongly conducive to charismatic affect. Of particular importance for charismatic attributions is the invocation of selected cultural symbols and their associated emotions through figurative expressions. For instance. Franklin Delano Roosevelt, "in rhetorically presenting himself as the leader of a crusade of the common people against fear and want, ... employed elevated Biblical language and cadences in his major addresses" (Willner, 1984, p.154). - 25 -Overall, Willner enriches our understanding of the elements critical to the creation of charismatically oriented perceptions of political leaders. By recognizing the active role of the leader in catalyzing and shaping followers' charismatically oriented perceptions through a myriad of substantive and symbolic means, Willner's intensive analysis moves us beyond Weber's original conception of charisma. In particular, by carefully documenting her analysis with examples from the careers of noted charismatic political heads of state, the complexity of the generation of the charismatic phenomenon is further underscored. Sociological Views of Charisma From the sociological perspective, discussion of the charismatic phenomenon has primarily centred on two major themes: 1. the situational circumstances fostering charisma and 2. the existence and distribution of charisma within institutional structures. Prior to reviewing these two major themes, it should be underscored that several scholars have also studied the routinization of charisma in a wide variety of settings including social movement organizations (Trice 6 Beyer, 1984; Zald & Ash, 1966), in liberal arts colleges (Clark, 1970), and in Utopian communities (Kanter, 1968). For example. T r i c e and Beyer (1984), using data obtained primarily from participant observations over a 20 year period, compared the routinization of charisma in two social movement organizations: Alcoholics Anonymous and the National Council on Alcoholism. While the organizations differed markedly in their degree of - 26 -routinization, both had created an administrative structure to implement their respective missions and, in that sense, had routinized the charisma of their founders. In addition. Alcoholics Anonymous had developed elaborate rites and ceremonials to diffuse the charisma of its founder throughout the organization and strong written and oral traditions to perpetuate the charismatic's mission. Kanter (1968), in her study of nineteenth century American Utopian communities, referred to the dispersion of charisma throughout the community as "institutionalized awe". In Kanter's (1968, p.514) analysis, institutionalized awe "consists of ideological systems and structural arrangements which order and give meaning to the individual's life and attach this order and meaning to the social system. These not only satisfy the individual's 'need for meaning,' ...but also provide a sense of Tightness, certainty, and conviction.. .that promotes a moral evaluative commitment and surrender to collective authority". Since investigations into the routinization of charisma are not directly relevant to the present discussion of charisma, the remainder of this section will address the two major themes occurring in the sociological literature. Situational Circumstances Fostering Charisma In accordance with Weber's conception, some scholars have reported that charismatic leaders have typically emerged during a period of social upheaval which causes distress and agitation among a group of people (e.g., Barnes, 1978; Tucker, 1970). For example, Barnes (1978), in a historical study of charismatic religious leaders, found they existed during - 27 -an era of radical social change in which a new formulation of religious beliefs was possible. In contrast to this position, others have argued while the roots of charisma are oriented towards social situations, the presence of crisis is not necessarily part of the situation (e.g., Berger, 1963; Clark, 1970; Dow, 1969; Eisenstadt, 1968; Friedland, 1964; Ceertz, 1977; Shils, 1965). For example, based on empirical data of the founding of political and trade unions in Tanganyika, Friedland (1964, p.25) observed: In any social situation there can be found incipient charismatics. Before incipient charismatics can emerge as genuine, the social situation must exist within which their message is relevant and meaningful to people. More specifically, he suggests that: Charisma appears in situations where (a) leaders formulate inchoate sentiments deeply held by masses; (b) the expression of such sentiments is seen as hazardous; (c) success... is registered (Friedland, 1964, p. 18). Therefore, according to Friedland's (1964) investigation, appropriate social circumstances are necessary for potentially charismatic leaders to be recognized and heeded. In an historical analysis of three distinctive liberal arts colleges -Antioch, Reed, and Swarthmore - Clark (1970, p. 241) describes the conditions which "cause high and low probabilities of the occurrence of both intense and concentrated charisma as well as its attenuated and - 28 -dispersed expressions." According to him, the occurrence of charisma in organizations is partially controlled by choice. That i s , individuals who appear strongly charismatic may be judged as too unsettling and therefore inappropriate for the stability and continuity of the existing structure. Hence, they are not selected for positions within the organization (Clark, 1970). Consistent with Weber's formulation, Clark (1970) also argues that charisma can be controlled by followers' denial of the exceptional personal qualities of the charismatic leader. If charisma is not attributed to the leader, then in that context, the leader does not have it (Clark, 1970). Clark (1970) further outlines three conditions which facilitate the occurrence of charisma. F i r s t , in accordance with Weber's view, crisis precipitates charisma. Charismatic leaders, by virtue of their extra-ordinary personal qualities and inspirational sense of mission, can infuse renewed purpose into the organization and mobilize collective effort and resources. The second condition fostering the emergence of charisma is the creation of a new organization (Clark, 1970). In the initial establishment phase, a charismatic leader may be selected for the top administrative position in order to build and shape the character, purpose, and image of the new organization (Clark, 1970). The final situation conducive to the development of charisma entails "a set of evolving conditions in which the charismatic figure picks up - 29 -support, gains in power, sets the direction of change, and extends the leeway for personal influence on policy and events" (Clark, 1970, p.244). Clark (1970, pp.244-245) concludes: Thus new organization is open to charisma, and c r i s i s helps to create i t , but neither is a necessary condition. Potentially charismatic men can enter successful and relatively stable organizations and encourage the conditions that realize their charisma. What is initially required is an opening for leadership, usually manifested by a willingness to improve. The potentially charismatic figure who works without the benefit of new organization or organizational c r i s i s is usually forced to string out his break with tradition. His charisma is shielded by patience. However, that his impact is more evolutionary need not diminish the magnitude of the change or the effect upon others and upon the initiation of a legend. In summary, recent conceptual and empirical work has advanced our thinking regarding the situational circumstances facilitating the expression of charisma. In contrast to the restrictive view of charisma as grounded in c r i s i s , it is argued that charisma can emerge in a wide range of social contexts (e.g., Clark, 1970; Dow, 1969; Eisenstadt, 1968; Shils, 1965; T r i c e & Beyer, 1984). In Geertz's (1977, p.152) words, charisma "does not appear only in extravagant forms and fleeting moments, but in an abiding, if combustible,- aspect of social life that occasionally bursts into open flame." Moreover, consistent with Winner's (1984) view, contemporary sociologists also recognize that a confluence of forces spawns the development of charisma including the social context, the exceptional individual, and the ideas s/he espouses (e.g., Dow, 1969; Geertz, 1977; T r i c e & Beyer, 1984). - 30 -The Existence and Distribution of Charisma in Organizations As discussed earlier, many sociologists have challenged Weber's contention that charisma is a force which must exist outside of institutional structures (e.g., Barnes, 1978; Berger, 1963; Blau & Scott, 1962; Dow, 1969; Etzioni, 1961; Parsons, 1937; Shils, 1965; Tucker, 1970). They argue that within bureaucratic or rational-legal structures the potential for personal charismatic expression exists. By virtue of their personal and exemplary qualities, charismatic leaders can create a further legitimacy for actions extending beyond their stipulated offices (Dow, 1969, p.311). Thus, in Shils' (1965, p. 202) words, "charisma can become an integral element in the process of secular institutionalization." Based on the premise that charisma exists within complex institutional structures, several scholars have sought to explicate its distribution within organizations. While some argue that charisma primarily exists in the top echelons of an organization (e.g., Katz & Kahn, 1978), others, notably Etzioni (1961), suggest that charisma may be functionally required and actually developed in a large variety of organizational positions, not just the top ones. According to Etzioni (1961, p.203), charisma is the "ability of an actor to exercise diffuse and intense influence over the normative orientations of other actors." In his analysis, there are three major forms of distribution of charisma within organizations: at the top only, in all line positions, or in one or more ranks other than the top. T h i s distribution is related to the compliance structure of the organization: coercive organizations have no charismatic organizational positions; utilitarian organizations tend to have charisma concentrated in top positions only; pure normative organizations are likely to have a line concentration and - 31 -social normative have a rank concentration (Etzioni, 1961, pp.208-209). Etzioni (1961) further suggests that there are three central determinants of the distribution of charisma in these organizational structures: (1) the nature of the involvement required (moral versus calculative or coercive), (2) the distribution of means-ends decisions among various organizational positions, and (3) the distribution of control over instrumental and expressive activities among various organizational positions. "Positions in which decisions about ends are made, expressive performances are con-trolled, and moral involvement of subordinates is necessary, requires charismatics. For positions in which decisions about means are made, from which instrumental activities are.controlled and in which moral involvement of subordinates is not necessary, charismatics are not required for effective performance: their presence may even be dysfunctional" (Etzioni, 1961, p.211). Therefore, in contrast to Weber's theorem, Etzioni's (1961) analysis suggests that charisma may originate in a wide range 4 of organizational positions and is compatible with established organizational structures. To date, the discussion of charisma has focused on the political and sociological spheres of inquiry. These perspectives have provided valuable insights into the nature and effects of charisma in large scale political or social movements. Moreover, the sociological view has discussed the existence and distribution of charisma in organizational settings. To fully appreciate, however f the phenomenon of charisma within organizations, the focal point of this dissertation, a psychological focus is necessary. It is to this perspective that we now t u r n . - 32 -Psychological Views of Charisma The psychological perspective has primarily focused on the personality characteristics of charismatic leaders, their behaviours, and their effects on followers within organizational settings. For example, the personality structure and dynamics of charismatic leaders have been described from multiple perspectives (e.g., McClelland, 1975; Mcintosh, 1970; Redl, 1942; Rutan & Rice, 1981; Zaleznik, 1974). From a psychoanalytic viewpoint, charisma designates "the force of the externalized unconscious, that is, unconscious tendencies which slip into awareness in the guise of an external force. The aura of magic springs from the resonance between what is perceived to be the external reality and the unconscious thought which is the real source of the experience" (Mcintosh, 1970, p.902). In his examination of leadership traits, Zaleznik (1974) distinguishes between the inner directed personality of charismatic leaders who establish strong emotional bonds with their followers and the outer directed personality of consensus leaders who negotiate among diverse interest groups to reach satisfactory agreements. Finally, based on his work on the power motive, McClelland (1975) has distinguished between the personalized and socialized faces of charisma. Personalized charismatic leaders, through their over-whelming authority and personal dominance, evoke feelings of obedience and loyal submission in their followers (McClelland, 1975, p.259). In contrast, socialized charismatic leaders, by v i v i d l y expressing meaningful goals and expressing confidence in followers' abilities to attain these goals, strengthen and inspirit their followers. - 33 -Other researchers have conducted in-depth explorations of charismatic leader behaviours in diverse organizational settings and roles. For example. Day (1980) described the development of a radically new social service program by a director of a maternity home. Roberts (1984) studied a school di s t r i c t superintendent who was a catalyst in introducing innovations to the di s t r i c t and was instrumental in developing a supportive climate for change and innovation. Finally, Conger (1985) conducted an exploratory field study of three highly charismatic, three moderately charismatic, and two noncharismatic American business executives. Despite the wide range of organizational contexts and positions examined, these studies revealed high convergence among the charismatic leader behaviours reported. In sum, the distinguishing behaviours of the charismatic leaders were the creation and articulation of a meaningful vision; active campaign-ing for the vision; unconventional or countercultural behaviours and practices; captivating speaking style; the ability to excite others; and high energy and dynamism. Further insights into charismatic leader behaviours are offered by Bennis and Nanus (1985) in their study of 80 chief executive officers and 10 in-depth interviews with successful innovative leaders. According to their results, "the capacity to relate a compelling image of a desired state of affairs" which involves and empowers followers is a fundamental component of charismatic or transformative power (Bennis S Nanus, 1985, p.33 ) . Moreover, through the use of metaphors, symbols, ceremonies and insignia, the chief executive officers transmitted their vision and fostered subordinates' commitment to this vision. In addition, these leaders could communicate their vision to induce the commitment of their multiple - 34 -constituencies. Persistence and consistency in maintaining the organi-zation's course, especially in adverse circumstances, were further competencies of these leaders. Yet they also had the capacity to change the organization and its members when faced with new conditions. An interesting analysis of the relationship between types of leadership and the emotional tone of the organization is proposed by Berlew (1974). He argues that just as leadership theory has advanced beyond the custodial mode that generates neutrality to the management mode that generates satisfaction, it should now move to the charismatic mode that creates excitement. Three components of charismatic leader behaviour are outlined. F i r s t , charismatic leaders develop a common vision related to values shared by organizational members. Second, they discover or create value related opportunities and activities within the framework of the mission and goals of the organization. Focal values include self reliance, community, excellence, service, and citizenship. Finally, charismatic leaders make organizational members feel stronger and more in control of their own destinies, both individually and collectively, by communicating high expectations, rewarding effective performance, encouraging collaboration, creating success experiences, and providing assistance when requested. A more encompassing examination of the development and maintenance of a charismatic association within organizational settings is presented by Oberg (1972). He proposes five sets of conditions for the institutional-ization of charisma in operating large organizations, both public and private, in contemporary Western society. F i r s t , four personal qualities - 35 -that are indispensible for charismatic leaders are described including: (1) the prestige of demonstrated achievement, (2) the ability to empathize with, and communicate their understanding of, the needs of their followers, (3) the ability to look and act the part, and (4) the possession of personality traits which permit them to accept the charismatic attachment of others. The second set of requirements for the development of institutionalized charisma involves organizational members' willingness to follow their conscious or unconscious desire for charismatic attachment and dependency. A third set of conditions relates to the nature of the decisions the leader is called on to make and/or execute. Decisions involving either unclear goals or unclear means or both are the likeliest candidate for charismatic leadership. A fourth set of conditions for the institutionalization of charisma involves four kinds of deliberate myth making or charisma building efforts including: (1) charisma facilitating policies and practices, (2) the use of symbols of prestige and status, (3) the use of ritual, and (4) the use of executive dramaturgy. Finally, the corporate charter, creed, or ideology can contribute to charisma. Oberg (1972) concludes that organizations which are able to develop and sustain the charismatic loyalty and devotion of their members would appear to have a strong advantage in the struggle for organizational s u r v i v a l . Further insights into intraorganizational charisma can be gleaned by reviewing two theories of charismatic leadership: House's (1977) theory of charismatic leadership and Bass' (1985) theory of transformational leadership. These theories represent a major contribution to our understanding of charismatic leadership within an organizational setting. Accordingly, these theories and the accompanying research evidence are examined in detail below. - 36 -House's Theory of Charismatic Leadership An Overview of the Theory Drawing on the political science, sociological, and social psychological literatures. House (1977) has advanced a theory of charismatic leadership within organizations. He describes and advances numerous propositions regarding the personal characteristics of charismatic leaders, their behaviour, their effects on followers, and situational factors associated with the emergence and effectiveness of charismatic leaders. With regard to personal characteristics, charismatic leaders have exceedingly high levels of self confidence, dominance, need for influence, and strong conviction in the moral righteousness of their beliefs (House, 1977, pp.193-194). On the basis of Sashkin's (1977) commentary on his theory. House also acknowledges that charismatic leaders possess intellectual fortitude, integrity of character, and speech fluency (Sashkin, 1977, p.214). The behaviour of charismatic leaders encompasses the articulation of a transcendent goal in order to provide meaning and to generate excitement; the communication of high performance expectations for followers and the exhibition of confidence in their ability to meet such expectations; the conveyance of messages that arouse motives especially relevant to mission accomplishment; the role modelling of values and beliefs; and the engagement in image building to create the impression of competence and success. These behaviours are linked to charismatic leaders' effects on followers including followers' trust in the correctness of the leader's beliefs, similarity of followers' beliefs to those of the leader, unquestioning - 37 -acceptance of the leader, loyalty to and affection for the leader, willing obedience to the leader, identification with and emulation of the leader, emotional involvement of followers in the mission, heightened goals of the followers, and the feeling on the part of the followers that they are able to accomplish or contribute to the accomplishment of the mission (House, 1977, p.191). Finally, stressful circumstances and work roles that are conducive to ideological value orientation are situational factors facilitating followers' receptivity to ideological appeals. Research Evidence To date, three studies have been conducted which bear on House's (1977) theory of charismatic leadership. For example. Smith (1982) tested House's (1977) proposition that charismatic leaders can be differentiated from noncharismatic leaders on the basis of scaled subordinate responses. Using a broad based nomination strategy, 30 charismatic and 30 noncharismatic leaders were identified from a wide range of formal work organizations. Specifically, the researcher asked full-time employed evening students to provide names of leaders they knew personally and considered highly charismatic. In addition, the same students identified a second leader, who held a similar position as the nominated charismatic leader, whom they considered to be effective but not charismatic. The immediate subordinate of each nominated leader in the sample completed a question-naire measuring 18 different constructs derived from the theory. Discriminant analysis of the questionnaire responses indicated that charismatic leaders could be distinguished from noncharismatic leaders on the following dimensions: leader dynamism, self esteem of subordinates. - 3 8 -experienced meaningfulness of work, felt back-up from the leader, self disclosure to the leader, work week length and performance ratings of subordinates. T h i s study demonstrates a high degree of correspondence between the constructs articulated in House's (1977) theory and the above empirically identified dimensions. That i s , in accordance with House's (1977) theoretical proposition, reputedly charismatic leaders do have effects on followers' performance, motivation, t r u s t , and self esteem. Thus, on the basis of Smith's (1982) findings, it can be concluded that the charismatic phenomenon exists and can be reliably measured with respect to intraorganizational leaders. Further empirical evidence pertaining to the effects of charismatic leadership is presented by Yukl and Van Fleet (1982). They conducted four studies investigating the relationship between the extent to which leaders engaged in specific types of leader behaviour and leader effectiveness in combat, simulated combat, and two noncombat situations. Two studies used a questionnaire - correlational methodology and two used content analysis of critical incidents (Yukl & Van Fleet, 1982). Of particular relevance to the present discussion is Yukl and Van Fleet's measure of "inspirational leadership" which was partially derived from House's (1977) charismatic theory (House, 1984). According to their formulation, leader behaviours reflective of inspiration include instilling pride in individuals; using pep talks to build morale; setting a personal example by his/her own behaviour; providing personal encouragement to a subordinate to build his/her confidence; and making cadets feel proud of their unit by complimenting good performance (Yukl & Van Fleet, 1982, p.98). The results indicated that in both combat and noncombat situa-- 39 -tions, inspiration, performance emphasis, role clarification, and criticism-discipline leader behaviours were related to group performance. With regard to inspiration, under combat conditions effective inspirational leader behaviour that appeals to subordinates' values and ideals, promotes identification with the group and its purpose, and builds subordinates' self confidence, motivated subordinates to expend the extra effort needed to succeed against formidable odds (Yukl S Van Fleet, 1982, p.101). In noncombat situations, inspirational leadership increased subordinates' motivation to perform monotonous dri l l routines and "built strong commitment to a military career that involves substantial personal and economic sacrifices" (Yukl 6 Van Fleet, 1982, p.101). Thus the findings of Yukl and Van Fleet (1982) lend support to House's (1977) proposition that inspirational or charismatic leader behaviour can enhance subordinate effort. Moreover, the results of this study are especially significant given the use of multiple methods and situations. More recently. House (1985a) has tested his charismatic theory of leadership via biographical analysis of Canadian and American heads of state. Based on both popular political history literature and prominent political historians' judgements, heads of state were classified as charismatic, noncharismatic, or equivocal. On the basis of these judgements, a sample of unequivocally charismatic and noncharismatic leaders was selected. The behaviour of these leaders and their effects on subordinates were examined utilizing biographies from selected cabinet members. Specifically, the domain of observation for these relations was a particular issue that arose during the term of each leader and cabinet member pairing. A content analysis of the relevant literature ( i . e . . - no -autobiographies, collections of personal papers, diaries, official biographies) pertaining to this issue and to the leader-member relationship was performed to determine the presence or absence of charismatic or noncharismatic effects and behaviours. The preliminary results suggest that in comparison to noncharismatic leaders, charismatic leaders exhibited greater self confidence, higher performance expectations, and more positive consideration. With regard to leader effects, in comparison to followers of noncharismatic leaders, followers of charismatic leaders had higher obedience, acceptance, and trust in their leader; made more positive statements about the mission and no negative statements about the mission; had more positive feelings about the situation; and were more positive and self confident. A particularly intriguing finding is that followers of charismatic leaders exhibited greater positive and negative affect toward the leader than followers of non-charismatic leaders. Several explanations may be posited for this finding. For example, charismatic leaders may polarize followers; commanding allegiance, reverence and loyalty among supporters and generating hatred, animosity, and fear among opponents. An alternative explanation advanced by House (1985a) is that leaders may be selectively charismatic. That i s , drawing on Craen and Cashman's (1975) in-group and out-group distinc-tion, leaders may be charismatic towards "entrusted lieutenants" thereby generating positive affect and noncharismatic towards "outcasts" thereby generating neutral or negative affect. While the data are still being analyzed, it may be tentatively concluded that House's (1977) charismatic theory has received empirical support thus far, with respect to the effects and behaviours of charismatic political heads of state. - 41 -Bass' Theory of Transformational Leadership An Overview of the Theory Drawing on the penetrating analysis of transactional and trans-formational leadership proposed by Burns (1978), Bass (1985) has recently developed a model of transformational leadership in organizational settings. He argues that the current leadership literature has heavily emphasized transactional models of leadership (e.g., path-goal, vertical dyad linkage, and operant conditioning theories of leadership). In these models, leadership is conceived of an exchange process in which subordinates receive rewards for reaching established objectives. Specifically, transactional leaders, by clarifying task requirements contribute to subordinates' confidence that with some degree of effort they can succeed in accomplishing their assignments. Transactional leaders also recognize subordinates' needs and clarify how these needs can be met if they exert the necessary effort. In essence, this process reflects leadership by contingent reinforcement (Avolio & Bass, 1985). According to Bass (1985), transactional leadership results in expected effort and performance on the part of followers. Consistent with Burns' (1978) paradigm, Bass (1985, p.11) posits that transformational leaders motivate followers to strive for transcendental goals and for higher order self actualization needs rather than focusing on their immediate self interests. In comparison to transactional leaders, transformational leaders are more likely to obtain higher levels of effort from subordinates who receive self reinforcement from performing a task rather than from external rewards (Avolio & Bass, 1985, p.7). - 42 -Therefore, through transformational leadership, followers are inspired to exert extraordinary effort. In addition to heightened motivation, followers also form a deep emotional attachment to their leader. Based on both qualitative and quantitative procedures, Bass (1985) sought to determine the behaviours and effects of transformational leaders. Adopting Burns' (1978) definition of transformational leadership, Bass (1985) surveyed 70 male senior executives to describe transformational leaders they had encountered during their career. Based on this survey and on a comprehensive review of the leadership literature, 142 items describing transformational and transactional leaders were generated. Eleven graduate students, using detailed definitions of transactional and transformational leadership, assigned each item to one of three categories: transactional, transformational, or can't say. Items which raters could reliably categorize were selected for inclusion in a questionnaire (The Multifactor Leadership Questionnaire - MLQ) administered to 176 military officers who rated their immediate superiors with respect to how frequently they exhibited each leadership characteristic. Results of a principal component factor analysis with a varimax rotation of the 73 final items yielded five leadership factors which accounted for approximately 90% of the total variance. Identical factors have emerged in an independent factor analysis of 360 managers' questionnaire responses (Avolio 8 Bass, 1985). Two leadership factors, labelled as transactional, were contingent reward and management-by-exception. Contingent reward (accounting for 6.3% of the variance) refers to the leader instructing subordinates what - 43 -to do to attain a desired reward for their efforts. Management-by-exception (accounting for 3% of the variance) entails the leader avoiding providing direction if customary ways are working and permitting subordinates to continue doing their job as long as performance goals are met. Three transformational factors - charisma, intellectual stimulation, and individualized consideration - were identified. Charisma (accounting for 65% of the variance) refers to the leader's ability to effectively articulate a captivating vision; to inspire and encourage higher order effort on the part of followers; and to instill respect, faith, loyalty, and trust in him/herself. Intellectual stimulation (accounting for 6% of the variance) encompasses the leader's ability to suggest creative, novel ideas which result in a discrete leap in followers' conceptualization, comprehension, and discernment of the nature of problems and their solutions (Bass, 1985). Individualized consideration (accounting for 6% of the variance) refers to the leader's developmental and individualistic orientation towards followers. The leader provides examples and assigns tasks to followers on an individual basis to help them significantly alter their abilities and motivation (Bass, 1985). In Bass' (1985) view, leadership is conceived as a five factor profile. While transactional and transformational leadership are conceptually distinct, Bass (1985) argues that these two leadership styles can be exhibited by the same individual to different degrees. According to his model, "transformational leadership will contribute in an incremental way to extra effort, effectiveness, and satisfaction with the leader as well as to _ nn -appraised subordinate performance beyond expectations that are a t t r i -butable to transactional leadership" (Bass, 1985, p.229), Therefore, transformational leadership augments transactional leadership. It is informative to compare and contrast the theoretical formulations proposed by House (1977) and by Bass (1985). As stated by Bass (1985, p.54), House's (1977) theory focuses on the more observable, rational aspects of charisma. In order to capture the full flavour of charisma, Bass (1985) contends that the emotional components need to be emphasized. Specifically, by using v i v i d , colourful, persuasive language, by employing meaningful symbols and imagery, and by generating a sense of excitement and adventure surrounding the mission, charismatic leaders appeal to the feelings, sentiments, and emotions of their followers. Bass (1985) further expands House's (1977) model by recognizing that charisma represents one component of the transformational process. According to Avolio and Bass (1985, p.16), the process of charismatic and transformational leadership differs in that: The transformational leader excites subordinates, but goes further in coaching them to think on their own and to develop new ventures which will further the group's goals while also developing the subordinate in his/her own right. Although the outcomes may be identical in the short term, it is the transformational leader who builds within the subordinate the willingness and motivation to question future systems and rules that the transformational leader never dreamed of in his/her original vision. -- 45 -Research Evidence Bass and his associates (Avolio £ Bass, 1985; Bass, 1985; Waldman, Bass, £ Einstein, 1985) have conducted a series of studies examining the effects of transformational and transactional leadership on individual, group, and organizational effectiveness. As reported in Bass (1985), results from surveys of 256 business managers, 23 education admin-istrators, 45 professionals, and 176 senior army officers consistently revealed that the three transformational factors were more highly correlated with perceived unit effectiveness and subordinate satisfaction with the leader than the two transactional factors. The incremental effect of transformational leadership over trans-actional leadership in predicting individual work effort and performance was recently investigated by Waldman, Bass and Einstein (1985). In this study, 256 managers from a manufacturing firm and 261 officers from the military rated the behaviour of their current immediate supervisors, on the MLQ. An index of individual performance effectiveness and measure of extra effort were developed and completed by the participants. For the manufacturing sample, the participants' performance appraisal score based on evaluations by superiors approximately six months prior to the questionnaire administration constituted the index of performance effectiveness. While the problem of common method variance threatens the validity of this study, the results revealed that transformational leadership factors were more highly correlated with followers' effort and job and work group performance than were transactional leadership factors. The results further indicated that transformational leadership factors predicted unique variance in followers 1 effort and performance beyond that of transactional - 46 -leadership factors. Thus, in accordance with Bass' (1985) model, trans-formational leadership augmented the motivational effects of transactional leadership rather than substituting for it. Avolio, Waldman, Einstein, and Bass (1985) (cited in Avolio & Bass, 1985) have further investigated the effect of transformational leadership on group performance. At the beginning of a school semester, 162 Master of Business Administration students were randomly assigned to nine member teams to participate in a management simulation game. A peer nomination procedure was used to select the leader for each team. At the end of the semester, each group member rated the leader on the MLQ, satisfaction with the leader, and leader effectiveness in managing the group. Performance data consisted of a variety of financial indicators measured over eight full quarters of performance. The findings revealed that teams with leaders having higher ratings of transformational leadership had significantly higher performance, higher levels of satisfaction with their leader, and greater effectiveness as leaders. However, as Avolio and Bass (1985) point out, since the leader ratings occurred at the end of the semester, it is possible that by virtue of their successful performance, team members erroneously attributed transformational qualities to their leaders. Attempts to resolve this issue through post hoc analyses of the relationship between leader ratings and team performance during the initial and latter parts of the semester have led to conflicting results. While a definitive test of Bass' (1985) transformational leadership theory remains to be conducted, the research evidence discussed above provides support for two postulations. F i r s t , transformational leadership - 47 -can significantly contribute to subordinate performance above and beyond transactional leadership. Second, transformational leadership has a positive impact on individual and group performance and satisfaction in a variety of organizational settings. To conclude, the foregoing review of the political science, socio-logical, and psychological literatures has underscored the complex, multifaceted nature of charismatic leadership. As Dow (1969, p.315) succinctly summarizes: There is no single charismatic temperament or personality type, but there is a charismatic phenomenon which can be theoretically and empirically isolated as an independent form of authority. Basically, it involves a distinct social relationship between leader and follower, in which the leader presents a revolutionary idea, a transcendent image or ideal which goes beyond the immediate, the proximate, or the reasonable; while the follower accepts this course of action not because of its rational likelihood of success.. .but because of an affective belief in the extraordinary qualities of the leader. Thus the leader appeals neither to intellectually calculable rules, nor to tradition, but to the revolutionary image and his own exemplary qualities with which the follower may identify. If such identification occurs, that is, if these reciprocal role expectations are met, the relationship is charismatic. T h i s applies in small group dynamics as well as in large-scale social movements. In particular, viewing charisma from a psychological perspective focuses our attention on the personality characteristics of charismatic leaders, their behaviours, and their effects on followers within organizations which are modernized, bureaucratic, and conventional. Moreover, the current theoretical and empirical literature suggests that intraorganizational charismatic leaders can profoundly influence followers' effort, performance and affective responses towards such leaders. Therefore the study of the behaviours and effects of charismatic - 48 -leaders within organizational settings would appear to be a worthwhile endeavour. Theory and Research Related to Structuring and Considerate Leadership Styles T h i s section reviews the literature related to structuring and considerate leadership styles and their effects on subordinate performance on and adjustment to an ambiguous task. Three major areas of leadership research are discussed: (1) the behavioural approach, (2) the contingency approach, and (3) leadership under conditions of stress. The Behavioural Approach The behavioural approach has focused on identifying the behavioural correlates of effective leadership. T h i s approach was based on the assumption that "knowledge of the behaviour patterns which characterize effective leaders would provide a rational basis for the design of programs to instill these behaviour patterns in actual or potential leaders" (Vroom, 1976, p.1530). Thi s section reviews experimental findings concerned with task oriented and socioemotional leadership and field studies associated with initiating structure and consideration. - 49 -Task Oriented and Socioemotional Oriented Leadership Based on observational studies of role differentiation in small experimental discussion groups. Bales and his colleagues (e.g., Borgatta, Bales & Couch, 1954; Bales & Slater, 1955; Bales, 1958) identified two specialized leadership roles: (1) achievement of a specific group task by directing, summarizing, and providing ideas (task oriented) and (2) maintenance or strengthening of the group's social relations by alleviating frustrations and disappointments (socioemotional oriented). Subsequent research has examined the conditions under which task oriented and socioemotional oriented leaders influence group performance and satisfaction (e.g.. Burke, 1967; Custafson, 1968; Custafson & Harrell, 1970). Under conditions where tasks were intrinsically satisfying to group members or members were highly committed to task accomplishment, the task oriented leader was deemed as critical to group success and there was less need for socioemotional leadership (House & Baetz, 1979, p.358). In contrast, where tasks were intrinsically dissatisfying to group members or members had low commitment to the task, the socioemotional leader was necessary to provide social satisfaction (Gustafson, 1968; Verba, 1961). Apparently, the task oriented leader under such conditions was likely to be resented and therefore socioemotional leadership was needed to compensate for this resentment. A study by Bales (1958) bears directly on this issue. He found that for task oriented leaders who permitted members to give feedback and to raise objections, qualifications, questions, and counter suggestions, there was no relationship between task orien-tation and li k i n g . However, for those task oriented leaders who did not allow such feedback, the relationship between task orientation and liking was negative. - 50 -In summary, the above findings suggest the following empirical generalizations: 1. Task oriented leadership is necessary for effective performance in all work groups. 2. Acceptance of task oriented leadership requires that such leaders permit group members to respond by giving feedback, making objections, and questioning the leader. 3. Socioemotional oriented leadership is required in addition to task oriented leadership when groups are not performing satisfying or ego involving tasks (House £ Baetz, 1979, p.359). Initiating Structure and Consideration Initiated in the late 1940s, the program of leadership research at the Ohio State University began by attempting to identify through factor analysis the dimensions needed to characterize differences in leader behaviour. Two dimensions were identified: consideration and initiating structure. Consideration reflects "the extent to which an individual is likely to have job relationships characterized by mutual trust, respect for subordinates' ideas, and consideration of their feelings" while initiating structure reflects "the extent to which an individual is likely to define and structure his role and those of his subordinates toward goal attainment" (Fleishman £ Peters, 1962, pp.43-44). These dimensions are conceptually related to those labelled by Bales (1958) as socioemotional oriented and task oriented. Over 50 published studies have assessed the relationship between leader initiating structure and consideration behaviour and various criteria of leader effectiveness including subordinates' satisfaction with their leader - 51 -and their job, performance, turnover, grievances, absenteeism and morale (House & Baetz, 1979, p.360). While the findings have been mixed, two major conclusions may be drawn. F i r s t , leaders who are high on con-sideration tend to have subordinates with fewer absences (e.g., Fleishman, Harris, & Burtt, 1955), lower turnover and grievance rates (e.g., Fleishman & Harris, 1962) and greater satisfaction with their job and their leader (e.g., Anderson, 1966; Fleishman, 1973; Halpin & Winer, 1957; Lowin et a l . , 1969; Stogdill, 1974) than those low on consideration. In addition, consistent with the findings on socioemotional leadership, a multitude of field studies have found that leader consideration is positively related to the satisfaction of subordinates who work on str e s s f u l , f r u s t r a t i n g , or dissatisfying tasks (e.g., Downey, Sheridan, S Slocum, 1975; House, 1971; House S Dessler, 1974; Stinson & Johnson, 1975). However, there is an absence of association between leader consideration and subordinate performance under dissatisfying task conditions (House & Baetz, 1979). The second major conclusion to be drawn is that correlations between leader initiating structure and subordinate satisfaction and performance show considerable variability with numerous reports of positive, zero, or negative relationships (e.g., Anderson, 1966; Chemers, 1984; Fleishman, Harris, 6 B u r t t , 1955; Cilmore et a l . , 1979; Korman, 1966). The lack of stronger, more consistent results with respect to the correlates of initiating structure has been traced to the particular scales used to measure this dimension. According to Schriesheim, House, and Kerr (1976), the Leader Behaviour Description Questionnaire (LBDQ) forms - 52 -are comprised mainly of items describing a leader who actively commun-icates with subordinates, facilitates information exchange, designs and structures the work of his/her subordinates and relationships among subordinates in their performance of work. In contrast, the Supervisory Behaviour Description Questionnaire (SBDQ) is comprised primarily of items describing a highly production oriented leader who is autocratic and punitive. Schriesheim and his colleagues (1976)reviewed studies pertaining to initiating structure and examined the results obtained from each version of this scale independently. Consequently, apparent inconsistencies in the findings were decreased and three empirical generalizations were drawn. Fi r s t , when measured by the SBDQ, leader initiating structure, while negatively related to satisfaction of the supervisor's subordinates, is generally positively related to performance ratings by superiors of manufacturing first-level supervisors and to ratings of their work group's performance. These findings were further supported with regard to noncommissioned infantry officers and air force officers. A consistent, yet considerably weaker, pattern of relationships was found for non-manufacturing supervisors of clerical workers performing routine tasks. Second, when measured by the revised LBDQ, initiating structure behaviour exhibited by first-level supervisors of nonmanufacturing employees performing routine tasks has a very weak positive relationship with subordinate ' satisfaction. Finally, high occupational level employees engaged in nonroutine, creative, or analytic work consistently react more favourably to leader initiating structure irrespective of the instrument used (Schriesheim et a l . , 1976, pp.301-302). Thus, as House - 53 -and Baetz (1979, p. 362) conclude, the findings from most field studies of leader initiating structure are congruent with those of small group research regarding task oriented leadership. The Contingency Approach There are several contingency theories of leadership which postulate that leader effectiveness is contingent upon the interaction of certain leader attributes with the specific parameters of the group, task, and environment (Chemers £ Rice, 1974). Two contingency theories are especially relevant to the present study; (1) Fiedler's (1967) contingency theory and (2) House's (1971) path-goal theory. Fiedler's Contingency Theory of Leadership Based on extensive research on personality variables and group characteristics, Fiedler (1967) and Fiedler and Chemers (1974) proposed a contingency theory of leadership which postulates that leader motivation structure and situational favourability interact to predict effective and ineffective leaders. The leader's motive hierarchy is measured by the Least Preferred Co-worker (LPC) scale. Leaders who describe their least preferred co-worker in positive terms (high LPC leaders) are assumed to be relationship motivated and those who describe their least preferred co-worker in negative terms (low LPC leaders) are assumed to be task motivated. An extensive body of literature indicates that the relationship - 54 -motivated leader is more attentive and responsive to interpersonal dynamics, more concerned with avoiding conflict and maintaining high morale, and more likely to behave in a participative and considerate manner. In contrast, the task motivated leader is more focused on the task aspects of the situation, more concerned with task success, and is inclined to behave in a s t r u c t u r i n g , directive, and somewhat autocratic manner (e.g., Chemers, 1984; Fiedler, 1967; Fiedler S Chemers, 1974; Rice, 1978). Situational favourability is defined as the extent to which the situation enables the leader to exert influence over subordinate performance (Fiedler, 1967). It is a combination of three variables: (1) the quality of leader-member relations (the degree of trust and support which followers give the leader), (2) task structure (the degree to which goals and procedures for accomplishing the group's task are clearly specified), and (3) position power (the degree to which the leader has the formal authority to reward and punish followers) (Fiedler, 1967). By dichotomizing each of these situation variables, Fiedler (1967) derived eight distinct combinations of leadership situations which defined eight octants. According to the theory, situational favourability is highest when leader-member relations are good, the task is highly structured, and the leader has substantial position power (Octant 1). Alternatively, situational favourability is lowest when leader-member relations are poor, the task is unstructured, and the leader has little position power (Octant 8) (Fiedler, 1967). The theory postulates that in situations of high (Octants 1,2, and 3) or low favourability (Octant 8) the task motivated leader is most effective; in situations of moderate favourability (Octants 4, 5, 6 and 7), the relationship motivated leader is more effective. - 55 -The contingency theory of leadership has generated intense debate (e.g., Ashour, 1973; Graen, O r r i s , & Alvares, 1971; Shiflett, 1973). Controversy has surrounded the basic validity of the contingency model, the conceptual meaning of the LPC measure, and the specification of the situational favourableness dimension (e.g., Ashour, 1973; Chemers & Rice, 1974; Graen, Alvares, O r r i s , S Martella, 1970; Green & Nebecker, 1977; House & Baetz, 1979; Schriesheim S Kerr, 1977; Shiflett, 1973; Singh, 1983). Fiedler (1971a, 1971b, 1973, 1977) has replied to most of the criticisms and the debate over his model continues. Through the years a large number of studies have assessed the hypothesized relationships for various octants and have found both supporting (e.g., Fiedler, 1967, 1972; Fielder, O'Brien, & llgen, 1969; Green & Nebecker, 1977; Hunt, 1967, 1971; llgen & O'Brien, 1974; Rice & Chemers, 1973, 1975) and non-supporting (e.g., Shiflett & Nealey, 1972) evidence for the contingency model. At present, there are only three published complete tests of the theory (House S Baetz, 1979). While the findings of Chemers and Skrzypek (1972) supported the model in seven of the eight octants, studies by Graen, O r r i s , and Alvares (1971) and Vecchio (1977) failed to support the model. However, Fiedler (1978) attributed these failures to methodological manipulations inadequate to test the theory. House and Baetz (1979, p.381) have concluded, given the multitude of partial tests that support the theoretical predictions for various octants, that the theory has some predictive power. More recent investigations of the contingency theory of leader effectiveness have applied meta-analytic techniques to various Fiedler - 56 -based studies. For instance, Strube and Garcia (1981), in their meta-analysis of the contingency model, found all but octants three and seven were fully supported by data from both laboratory and field studies. However, Vecchio (1983) has criticized this study for a variety of biases in selection of studies and interpretation of results. In a more rigorous meta-analysis of 38 Fiedler based studies testing individual octants of the model, Crehan (1984) found that the majority of studies only partially supported the theory. She concluded that the conflicting evidence may not be a function of the model itself but rather the substantial amount of inconsistency in its testing and recommended some ways in which the testing procedure should be standardized in order to accurately assess the mode I'.s validity. House's Path-Goal Theory of Leadership The path-goal theory of leadership focuses on how leader behaviour affects the motivation and satisfaction of subordinates. Based on the theoretical formulations of Evans (1970), the theory has been extended by House and his associates (House, 1971; House & Dessler, 1974; House & Mitchell, 1974). The theory consists of two basic propositions. F i r s t , "leader behaviour is acceptable and satisfying to subordinates to the extent that they see it as either an immediate source of satisfaction or as instrumental to future satisfaction" (Filley, House, & Kerr, 1976, p.254). Second, "leader behaviour will be motivational to the extent that (1) it makes satisfaction of subordinate needs contingent on effective performance and (2) it complements the environment of subordinates by providing the - 57 -coaching, guidance, support, and rewards which are necessary for effective performance and which may otherwise be lacking in subordinates or in their environment" (Filley, House, & Kerr, 1976, p.254). According to House (1971, p. 324), the leader's motivational functions are to increase personal payoffs to subordinates for goal attainment, to make the path to those payoffs easier to travel by clarifying it and reducing barriers, and to increase opportunities for personal satisfaction. Path-goal theory is composed of three interactive components which influence subordinate motivation and satisfaction: leader behaviour, subordinate perceptions, and situational variables. Each of these components is briefly discussed below. Leader behaviour. Although the initial statement of path-goal theory utilized the two dimensions of initiating structure and consideration as representative of the leader's behaviour (House, 1971), the current version of the theory (House & Mitchell, 1974) includes four categories of leader behaviour: supportive, directive, participative, and achievement-oriented. According to the theory, each of these leadership styles will lead to task motivation and subordinate satisfaction under different task structures. Subordinate perceptions. Serving as the intervening variables in path-goal theory, subordinate perceptions consist of effort-to-performance expectancy, performance-to-reward expectancy, and valence. That i s , in choosing between maximal work effort and minimal work effort, subordin-ates assess the probability that a given level of effort will lead to - 58 -successful task completion and attainment of task goals (effort-performance expectancy), the probability that successful task completion will lead to desirable outcomes and undesirable outcomes (performance-reward expectancy), and the desirability of each outcome (valence) (Yukl, 1981, p. 145). The theory proposes that if the leader can increase valence perceptions and clarify and increase expectancy probabilities, greater effort and higher satisfaction will result. Situational variables. Two situational factors are hypothesized to moderate the relationship between the effects of leader behaviour and the satisfaction and motivation of subordinates. These are: (1) the personal characteristics of the subordinate and (2) the work environment and task demands with which the subordinate must cope to accomplish work goals and to satisfy their needs (Filley, House, S Kerr, 1976; Schriesheim & Kerr, 1977). Subordinate characteristics include needs (e.g., needs for achievement, affiliation, and extrinsic rewards), ability to perform the task (e.g., job s k i l l s , knowledge, and experience) and personality traits (e.g., locus of control, self esteem). The environment includes factors not within subordinates' control, but which nevertheless affect their ability to perform effectively and to satisfy their needs (House & Dessler, 1974). Three important contingency factors in the environment are the task structure, the formal authority system of the organization, and the primary work group. In essence, path-goal theory suggests that leader behaviour, modified by the characteristics of subordinates and the work environment. - 59 -influences subordinate perceptions of valence and expectancies, which then can result in higher motivation and satisfaction. Based on this theory, several hypotheses have been advanced. The preponderance of studies have focused on testing two hypotheses derived from the theory. The fir s t hypothesis proposes that the lower the task structure the more positive the relationship between instrumental (task oriented, structuring) leader behaviour and subordinate task satisfaction and expectancies. That i s , when tasks are nonrepetitive, complex, and ambiguous, instrumental leadership would help subordinates to clarify path-goal relationships, thereby increasing their job satisfaction and expectancies. T h i s hypothesis has received mixed support (e.g., Downey, Sheridan, S Slocum, 1975; Greene, 1979; House S Dessler, 1974; Schriesheim & Kerr, 1977; Schriesheim & Murphy, 1976; Schriesheim & Schriesheim, 1980; Sims & Szilagyi, 1975; Stinson £ Johnson, 1975). However, it should be noted that several studies provide support for the postulation that under ambiguous task conditions there is an absence of a relationship between considerate leadership and individual task satisfaction and performance (e.g., Greene, 1979; House, 1971; House & Dessler, 1974). The theory further proposes that, when tasks are highly structured, subordinates are likely to resent their supervisors' attempts to initiate more structure (House & Dessler, 1974). Further task structure may be viewed as being excessively directive and restrictive, resulting in lower satisfaction to the subordinate and lower expectancy that his/her performance leads to rewards (House S Dessler, 1974). Th i s leads to the - 60 -second hypothesis: under highly structured tasks, supportive (socio-emotional oriented, considerate) leadership is hypothesized to be more positively related to subordinate satisfaction and expectancies than under structured tasks. T h i s hypothesis has been supported in several studies (e.g., Downey, Sheridan S Slocum, 1975; House & Dessler, 1974; Sheridan, Downey, S Slocum, 1975; Sims £ Szilagyi, 1975; Stinson £ Johnson, 1975; Szilagyi £ Sims, 1974). The conflicting findings regarding the moderating effects of task structure on the relationship between structuring leader behaviour and satisfaction and expectancies may be due to the inadequacies and differences in operationalizations of the theory's leader behaviour variables (Schriesheim £ Von Glinow, 1977). In an examination of the effects of different operationalizations of path-goal theory's leader behaviour constructs, Schriesheim and Von Glinow (1977) found that the SBDQ structure scale and its derivatives were poorer operationalizations of the theory's constructs than the other scales, the LBDQ and the LBDQ-XII. They concluded that better designed and adequately operational ized research is needed in order to satisfactorily evaluate the merits of the path-goal theory of leadership. Subsequent research utilizing appropriate operationalizations of the theory's independent, dependent, and moderator variables have yielded both supportive (Greene, 1979; Johns, 1978) and nonsupportive (Schriesheim £ Schriesheim, 1980) results. . More recent studies investigating a broader range of task dimensions (task feedback, task variety, opportunity to deal with others)(Schriesheim £ DeNisi, 1981) and of leader behaviours (upward influencing, achievement - 61 -oriented, contingent approval, a r b i t r a r y and punitive)(Fulk S Wendler, 1982) have yielded results strongly supportive of the underlying premises of path-goal theory. Accordingly, Schriesheim and DeNisi (1981, p.595) concluded that "the basic logic of the theory seems to have substantial merit and usefulness in predicting and explaining leadership phenomena." Leadership Under Conditions of Stress Job stress in the form of role conflict and role ambiguity has been linked to job dissatisfaction, job related tension and anxiety, reduced performance effectiveness and a greater likelihood of leaving the organization (e.g., House & Rizzo, 1972; Johnson & Stinson, 1975; Kahn, Wolfe, Quinn, Snoek, & Rosenthal, 1964; Miles, 1976; Rizzo, House, & Lirtzman, 1970). Considerable research has focused on discovering the appropriate leadership styles that might buffer or alleviate the negative effects of job stress. For example, Schriesheim and Murphy (1976) reported that when people feel anxious about their job situation, task direction correlates positively with performance. In a study of role conflict and role ambiguity among 3725 United States Navy personnel, LaRocco and Jones (1978) concluded that leader support, defined as helping subordinates to achieve work goals, was positively correlated with greater job satisfaction, higher self esteem, and a greater tendency to remain in the organization. Finally, Katz (1977) found when affective conflict was present, leader structuring behaviour in university departments was directly related to department effectiveness. These findings were confirmed in a subsequent laboratory experiment in which - 62 -students were hired to perform coding tasks under structuring or considerate leaders and under high or low affective conflict conditions. Under high affective conflict conditions, leader structuring was positively related to task performance. The hypothesis that affective conflict would be positively related to the desire for increased leader consideration was not supported in either the field or laboratory study. Considerable empirical research has found that considerate leader behaviour serves as a source of social satisfaction and support for the employee (e.g.. House, 1981; Katz & Kahn, 1966; Sheridan S Vrendenburgh, 1979; Szilagyi & Sims, 1974). For example. Seers and his colleagues (1983) found under conditions of role ambiguity and role conflict in a large federal government agency that subordinates' satisfaction with their supervisor increased when the supervisor exhibited considerate behaviour. A similar study of 89 middle-lower managerial personnel in a large, heavy equipment manufacturing firm in the Midwest found indi-viduals working with high consideration supervisors and experiencing a high level of role conflict or role ambiguity were more intrinsically satisfied than those with low consideration leaders while it was positive for those with high consideration leaders. As Abdel-Halim (1982) observed, while leader support tended to act as a buffer against deterioration in intrinsic job satisfaction associated with role conflict or role ambiguity, it failed to mitigate the heightened feelings of anxiety associated with role conflict. The organizational socialization literature also suggests that leaders who exhibit considerate behaviour may facilitate newcomers' integration into their new work environment. Hall and Nougaim (1968), for example, found - 6 3 -that the primary concern among young AT&T managers during their fir s t year of employment was getting established within and accepted by the organization. Katz (1978) has also shown that in the initial socialization stage of job longevity, employees appear highly receptive to role features that help establish and solidify feelings of personal acceptance, reassurance, and contribution within their new job environments. Thus during the initial stage of the organizational socialization process, supervisory coaching, psychological support, and positive interpersonal feedback helps newcomers to establish a sense of contribution and worth, to feel more secure and acceptable, and to find their overall niche in the scheme of things (Hall, 1976; Katz, 1980; Van Maanen, 1976). Hence the organizational assimilation process should proceed more smoothly when the leadership climate displays a high degree of consideration (Van Maanen, 1976, p.96). In summary, the literature suggests that structuring leader behaviour can facilitate individuals' task adjustment and performance by defining work role expectations and alleviating anxiety associated with initially stressful job situations. Considerate leader behaviour, on the other hand, has no effect on individuals' task adjustment and performance under ambiguous task conditions. The literature further suggests that structuring leader behaviour has no effect on individuals' interpersonal adjustment. In contrast, considerate leader behaviour enhances the interpersonal adjustment of subordinates as evidenced in their establish-ment of positive relations between superior and subordinate under stressful conditions (e.g., Abdel-Halim, 1982; Seers et a l . , 1983). - 64 -A L I T E R A T U R E REVIEW OF GROUP PRODUCTIVITY Overview According to Schriesheim and his associates (1979, p.107): Substantial progress in research on leadership and management can be facilitated by greater attention to the role of leadership in the context of small groups and a greater appreciation of the interdependence which exists between leaders and members of groups. They persuasively argue that the relationship between leaders and groups is important to understand for several reasons. F i r s t , leaders emerge from groups or are appointed to manage them and therefore require an understanding of group processes in order to be effective. Second, leaders play a critical role in determining the structure, direction, and goals of the groups they manage and in establishing and maintaining productive social relationships within those groups. Finally, leaders are influenced by the groups they lead and to the extent they understand the nature of this influence process, the better equipped they will be to modify their own behaviour accordingly (Schriesheim, Mowday, & Stogdill, 1979, p.107). Despite these compelling reasons for including groups in studies of leadership effectiveness, recent leadership research has predominantly focussed on the leader-subordinate dyad rather than the work group (Hunt, Osborn, & Schriesheim, 1978; Stogdill, 1974). One unfortunate consequence of this focus has been the separation of leader-subordinate relations from their social context, despite well established evidence of the - 65 -importance of group factors in interpersonal behaviour {e.g.. Hare, 1976; Shaw, 1976). While some leadership theories consider the primary work group as an important factor (e.g.. House's (1971) path-goal theory, Graen and Cashman's (1975) vertical dyad linkage theory, and Fiedler's (1967) contingency theory), only a few recent studies have examined group variables (e.g., Greene & Schriesheim, 1980; Podsakoff & Todor, 1985; Schriesheim, 1980). The present study extends the examination of group variables by investigating the role of group productivity norms in the leader-subordinate context. The research literature on groups is vast. A comprehensive review of this literature in the present study would fill a volume in itself. Accordingly, only the group literature that pertains directly to the present topic is reviewed. Specifically, this section initially discusses the potent influence of the work group on organizational newcomers. Next, the concept of group norms and the determinants of conformity to production norms will be explicated. Finally, empirical research related to low and high performance norms in industrial work groups will be described. The Influence of the Work Group on Newcomers'  Adjustment to the Organization Several researchers have observed that the work group has a major influence on newcomers' adaptation to their new organizational environment (Burns, 1955; Chadwick-Jones, 1964; Dubin et a l . , 1976; Evan, 1963; Hall, 1976; Hackman, 1976; Katz, 1980; Louis, 1980b; Van Maanen, 1976, - 66 -1978). For example, ethnographic studies of new medical students (Becker, Ceer, Hughes, & Strauss, 1961) and of Coast Guard recruits (Dornbush, 1955) have viv i d l y described how the work group can act as a defense against the extreme suppression and regimentation accompanying training and as a source of emotional support. Hall (1969) and Schein (1968) note that peer groups cushion the impact of "reality shock" accompanying the newcomer's encounter with the organization and provide role models to help the newcomer manage identity change. Thus, the work group profoundly influences the organizational socialization process by providing individuals with social support and by helping them navigate a path through the boundary passage process (Van Maanen, 1976, p.92). As Van Maanen (1978, p.20) notes, learning during socialization "does not occur in a social vacuum strictly on the basis of official and available versions of job requirements." Newcomers turn to other members of the work group to obtain assistance in interpreting the role demands dictated by the organization, to seek information about appropriate attitudes and behaviours, and to learn the informal networks (e.g., Feldman, 1981; Graen, 1976; Hall, 1969; Louis, 1980b; Schein, 1968). Therefore, according to Feldman (1981, p.314): The work group is a particularly important factor in determining how closely new recruits adjust to group norms and values. The group can filter out information that contradicts dominant values, so that values may be more readily accepted by newcomers. The work group can also exert some control over the amount of information new recruits get, and can advise recruits about the credibility of different sources of information. - 67 -Thus newcomers who have not yet had a chance to develop through experience their own personal maps of the organization, are heavily dependent upon members of their work groups for data about "what leads to what" and "what's good in the organization" (Hackman, 1976, p.1512). A particularly critical role of the group is to act as a normative referent for appropriate types of work behaviour such as levels of output and quality. In their review of the research evidence on leader-group interactions, Schriesheim and his colleagues (1979, p.109) observed that "group norms are often a more potent influence on individual and group performance than are individual, managerial, or organizational factors". The following section examines the nature and determinants of group productivity norms. Croup Norms for Productivity The influence of groups upon the attitudes and performance of its members is well documented in the social psychology literature (e.g., Allen, 1975; Asch, 1952; Cartwright S Zander, 1960; Newcomb, 1958, 1961 ; Sherif, 1936). Specifically, it has been consistently observed that "interaction among persons tends to decrease the variance in their behaviour, and, in the extreme, can produce highly standardized behaviour patterns" (Vroom, 1969, p.223). In accounting for this phenomenon, extensive use is made of the concept of group norms, informal rules that groups adopt to regulate and regularize group members' behaviour (e.g., Feldman, 1984, p. 47; Hackman, 1976; Shaw, 1976; - 68 -Vroom, 1969). Norms are created and enforced by group members. That is, through interaction, group members acquire and transmit information concerning the actions which will be rewarded or punished (Vroom, 1969, p.223). Thus norms establish the basis for social control in the group. In particular they define the kind of behaviour which is necessary for or consistent with the realization of goals adopted by the group (Hare, 1976, p.19). Croup norms regarding productivity have been the primary focus within organizations (e .g . , Lott S Lott, 1965; Vroom, 1969). Codes with regard to appropriate levels of production are manifest "in the outcomes of group activity, specifically, limited within-group variance in individual performance... as well as in such aspects of the group process as the fact that group members attend to the level of performance of others in their group and consistently reward those performing at appropriate levels while punishing those performing at inappropriate levels" (Vroom, 1969, p.223). Determinants of Conformity to Productivity Norms There are several circumstances associated with group members' adherence to performance norms. First, it is postulated that groups characterized by friendliness, cooperation, interpersonal attraction and related indications of group cohesiveness exert strong influence upon their members to behave in accordance with work norms and standards (Shaw, 1976, p.201). Since members value the interpersonal rewards available in highly cohesive groups, they are unwilling to risk losing these rewards by - 69 -violating group norms (Cummings, 1981 , p.252 ) . Some research evidence is available which suggests that highly cohesive groups are usually able to effectively control the behaviour of their members so that behaviour approximates the group norm. For example, in a laboratory study conducted by Schacter and his colleagues ( 1 9 5 1 ) , conditions of high versus low cohesiveness and high versus low productivity norms were created by experimental manipulation. Specifically, in groups composed of three female undergraduates constructing cardboard checkerboards, cohesiveness was manipulated by instructions regarding the congeniality of other group members, while the direction of norms was manipulated by intercepting notes sent from one member to another and substituting prewritten sets of notes designed to influence members to increase or decrease their level of production. The results indicated that the experimentally created norms tended to have the hypothesized effects, that i s , increasing production in the case of high production norms and decreasing production in the case of low production norms. However, the effect of group cohesiveness on conformity was in the same direction for both low and high productivity norms, reaching statistical significance only for the low production norm. In a replication of the Schacter et a l . experiment using male participants, a different task, and an extended production period, Berkowitz (1954) found significantly greater conformity in the high cohesive groups than in the low cohesive groups, irrespective of the direction of the productivity norm. Moreover, the results clearly indicated that the differential conformity of high and low cohesive groups persisted after the induction of norms had ended. - 70 -Similar findings regarding the effects of group cohesiveness on conformance to productivity norms were reported by Seashore (1954) using survey techniques. In a study of 228 work groups in a machinery factory, he found no relationship between cohesiveness and productivity, but, as expected, a fairly strong negative relationship between cohesiveness and the amount of variance in the productivity of group members. In comparison to workers in low cohesive groups, workers in high cohesive groups were more likely to produce at or about the same level as their co-workers. Collectively, these laboratory and field findings suggest that highly cohesive groups are effective in enforcing the performance norms adopted by groups. In addition to group cohesiveness, the research evidence suggests that member acceptance of group-supplied performance norms is a function of four interrelated factors. F i r s t , the more ambiguous the task, the greater the tendency for the member to rely on the group as a source of information, leading to greater conformity to group norms (e.g., Asch, 1952; Blake, Helson, S Mouton, 1957; Kiesler, 1969a, 1969b; Rakestraw S Weiss, 1981). Second, the less competent a member is at the task, or the more competent the group i s , the greater the conformity (e.g., Hackman, 1976; Hare, 1976; Rakestraw & Weiss, 1981; Rosenberg, 1961). T h i r d , holding constant member self confidence, a group member will tend to accept group norms to the extent that s/he perceives the group as being a credible source of information (e.g., Hackman, 1976; Rosenberg, 1961) and/or as having a higher status (e.g., Lefkowitz, Blake, & Mouton, 1955; Raven £ French, 1958). Finally, the greater the unanimity of group members' views, the more an individual will accept information provided by - 71 -the group (e.g., Allen S Levine, 1971). According to Hackman (1976), the influence of these factors on an individual's acceptance of performance goals and standards may be most pronounced early in a member's tenure in a group, before the member has had an opportunity to develop expecta-tions through experience. The following sections discuss the research related to the establishment and maintenance of low and high group productivity norms. Low Productivity Norms Observations of group productivity norms reported in the literature have predominantly focused on organized restriction of output (e.g., Lott & Lott, 1965; Roethlisberger & Dickson, 1939; Roy, 1952; Whyte, 1955). The classic and perhaps f i r s t quantitative demonstration of this phenomenon was provided by Roethlisberger and Dickson (1939) in the course of their experiments at Western Electric. They found systematic, group determined restriction of individual productivity in the manufacture of telephone equipment, a finding that has often been corroborated in the subsequent research literature (e.g., A r g y r i s , 1957; Coch S French, 1948; Dubin, 1958; French & Zander, 1949; Roy, 1952; Viteles, 1953; Whyte, 1955). Coch and French (1948), for example, describe how a group of workers lowered their productivity rate and exerted strong pressure for conformity to this reduced speed following a change to a new work method which was no more difficult than the former one. Similarly, Newcomb (1958) described a case in which the group had established a production - 72 -norm of 50 units per day, but one worker wanted to exceed the norm. Her attempts to do so were so successfully quelled by her peers that her output declined below the 50 unit norm. The subsequent dissolution of the work group resulted in the employee doubling her output within a short time period. The restriction of output and what Dubin (1958) has called "non-formal behaviour" has been consistently observed in industrial groups. Lupton (1976), in his description of shop floor behaviour, noted that work groups promote a sense of job security by controlling the work pace, provide nonwork preoccupations to relieve the strain of monotonous tasks, and through group pressures and sanctions secure from their members adherence to informal group norms. There are many detailed accounts of the various games workers play to hamper management control such as "goldbricking", "quota restriction", and "chiselling" (e.g., Burawoy, 1979; Haraszti, 1978; Roy, 1952). For instance, in the practice of chiselling: Work groups establish and enforce conventions regarding the manner in which individuals ought to report their performance to management. The accepted practice . . . i s that 'windfall' gains that result from exploiting easy performance times not be reported as such. The unreported 'windfall' hours are 'saved' and reported as time spent on difficult tasks, where in fact little or nothing has actually been gained. The 'windfall' time also was used to compensate for the diminished opportunity to earn that was due to unanticipated and uncontrollable workflow inter-ruptions... [Chiselling] is defended by workers on the ground that it irons out fluctuations in their earnings and conceals from management the existence of 'slack' performance standards, thus giving the workers a degree of flexibility in setting a reasonable schedule (to them) and continuing the apparent overall relationship between effort and reward. In order to maintain group influence over the level of individual earnings, it is necessary, as many studies have shown, that - 73 -[chiselling] be supplemented by group influence over the procedure for setting work standards on new jobs and on the procedures for allocating jobs among individuals. In short, work group influence on the reward-effort relationship has been observed to rely for its effectiveness, in many cases, on being comprehensive in its scope (Lupton, 1976, p. 178). High Productivity Norms While low productivity norms have received considerable attention in the literature, it would be fallacious to assume that the social pressures exerted by co-workers consistently lead to the lowering of productivity. Informal norms can induce a higher level of performance rather than restriction of output (e.g., Coch & French, 1948; Lawrence & Smith, 1955; Maier & Hoffman, 1964). For example, as Vroom (1969, p.224) has observed: The Hawthorne experiments in the Relay Assembly room... showed that a small group of workers placed in an isolated experimental environment achieved a high level of productivity and showed a more or less continual increase during the experimental period. While the experimental design did not permit unequivocal interpretation of this finding, the investigators attributed i t , at least in part, to the development of a new set of norms regarding behaviour on the job. These norms were consistent with, rather than antithetical to, the economic objectives of the formal organization of which the group was a part. In the years since the Hawthorne experiments, a long line of research has added to the evidence that peer influence can play a strong role in increasing performance if the group norms are congruent with the productivity goals of the system. For example, Likert (1956), in summarizing some of the findings obtained in field studies by Michigan's Survey Research Centre, accorded a highly significant role to the work group in producing lower absence rates, better interpersonal relations. - 74 -more favourable attitudes toward the job and company, and higher productivity goals. In a study of 40 agencies of a leading life insurance company. Bowers and Seashore (1966) found support, goal emphasis, and work facilitation provided by the peer group were positively correlated with various measures of employee satisfaction and organizational performance. Finally, Ouchi (1981) notes that the strong emphasis on collectivity orientations in Japanese work groups enhances individual feelings of personal worth and importance in task performance. In summary, the literature suggests that the group can and does have a powerful impact on individual work performance. Croup norms enforcing restriction of productivity can impede task performance, while group norms supporting high productivity can enhance task performance. PERFORMANCE Laboratory and field studies have traditionally examined the relationship between leadership style (e.g., initiating structure and consideration) and subordinates' behaviour, particularly their job performance (e.g., Greene, 1977; McCall & Lombardo, 1978). The measures of subordinate performance as an outcome of leader-subordinate inter-actions are diverse. Field studies have used global ratings of overall performance, objective organizational measures such as the amount of productivity, total sales, and salary increase, summary indices of work group or subunit performance (e.g., total absenteeism, number of units produced by the group, dollar value of the group's output), performance - 75 -evaluations by peers or superiors with respect to a variety of dimensions (e.g., quality and quantity of work, dependability, attendance, ability to get along with others, knowledge of work, initiative on the job), and self perceptions of effort toward quality and quantity of work (e.g., Campbell, 1977; Greene, 1979; Schriesheim, Mowday, S Stogdill, 1979; Sheridan, Downey, & Slocum, 1975; Stogdill, 1972; Szilagyi & Sims, 1974). Measures of subordinate performance in laboratory studies are typically based on individuals' performance on an experimental task (e.g., Hunt, 1971; Katz, 1977; Vecchio, 1979). For example, for an experimental task involving cleaning, f i l i n g , and adjustment of spark plugs according to specification sheets, Lowin, Hrapchak, and Kavanagh (1969) measured both quantitative ( i . e . , the number of plugs processed in 1 hour) and qualitative ( i . e . , the number of plugs whose settings deviated from the specifications) task performance. Similarly, Gilmore and his associates (1979), for an experimental task involving the coding of interactions in videotaped leaderless group discussions using Bales (1950) category system, measured two dimensions of task performance: (1) the total number of interactions coded for all 12 Bales categories (quantity) and (2) the absolute difference between the participant's coding and the consensus of three judges who determined the correct codes (quality). Thus, in laboratory studies of leadership, both qualitative and quantitative dimensions of task performance are typically measured. In addition to subordinate performance, there are other aspects of subordinate behaviour which have been largely ignored and yet which may have considerable bearing on leader-subordinate relationships (Greene, - 7 6 -1977). Hence we now turn to a discussion of subordinate adjustment to new role demands. ADJUSTMENT According to Super (1957), general adjustment is a synthesis of specific adjustments. McCrath (1976, p.1384), in his discussion of role stress, states that behaviour in organizations is contingent not only upon the task activities undertaken and the behavioural settings in which these activities are performed, but also upon the patterns of interpersonal relations within which those behaviours occur. Th i s implies that organizational newcomers, at a minimum, need to adapt to both the task and interpersonal demands in their new role. Task Adjustment Discussions by Lofquist and Dawis (1969), Bhagat (1983), and Graen (1976) are directly related to the task adjustment component. Lofquist and Dawis (1969) developed a theory of work adjustment which is based on the concept of correspondence between an individual and his/her environment. According to this theory, correspondence between an individual and his/her environment implies conditions which can be described as a "harmonious relationship between the individual and the environment" (Lofquist & Dawis, 1969, p.45). This harmonious relationship is characterized as being reciprocally suitable, that i s , the individual is - 77 -judged suitable by his/her environment and the environment is judged suitable by the individual. A basic assumption made by Lofquist and Dawis (1969) is that each individual seeks to achieve and maintain correspondence: when an individual enters a work environment for the first time, his/her behaviour is directed towards fulfilling its requirements. S/he also experiences the rewards of the work environment. If s/he finds a correspondent relationship between him/herself and the environment, s/he seeks to maintain i t . If s/he does not, s/he seeks to establish correspondence, or failing in this, to leave the work environment. Thus Lofquist and Dawis suggest that in monitoring newcomers' work adjustment we should focus on their satisfaction with their role and the organization's satisfaction with their performance. A similar conceptualization of task adjustment has been formulated by Bhagat (1983) in his model of the effects of stressful life events on individual effectiveness within the work setting. He postulates that the experience of stressful life events leads to reduced levels of job involvement, job performance effectiveness, job satisfaction and other work related behaviour outcomes. In contrast to Lofquist and Dawis' (1969) and Bhagat's (1983) focus on outcome-oriented' variables, George Graen (1976) has discussed task adjustment in terms of process oriented variables. According to his role making model, the jobs which new employees fill are partial or incomplete programs which must be completed over time by the organization's - 78 -participants. Moreover, the definition of the job is subject to negotiation and develops through the modification and accommodation of the new employee's role expectations and those of other organizational participants. The participants who hold expectations concerning the newcomer's behaviour and play a part in defining his/her new role are those with a vested interest in his/her performance. Thus the newcomer, his/her immediate supervisor, and peers are involved in the role definition process. These individuals constitute what has been termed the role set (Katz S Kahn, 1966). Thi s model postulates that role ambiguity and role conflict are crucial variables which may impede the role definition process. From the newcomer's perspective, role ambiguity is a lack of knowledge concerning the expectations held by the other members of the role set. More specifically, role ambiguity refers to the extent to which the newcomer accurately perceives the expectations held by those in his/her role set. Role conflict refers to the extent to which the expectations of the various members of the role set are divergent. Thus Craen's (1976) model views task adjustment as the process by which a newcomer's role is defined and focusses on the effects of role ambiguity and role conflict on this process. Interpersonal Adjustment According to McCrath (1976, p.1385), "a person's behaviour on a given task, at a given time and place, is affected by his continuing relations with others involved in that task, with others present in that - 79 -place, and indeed with others (e.g., his superior) not necessarily present at all but 'relevant' to the setting and/or the task". Thus a crucial component of the newcomer's adjustment to his/her new organizational environment is the establishment of interpersonal relationships with his/her supervisor or co-workers. As Katz (1980) has pointed out, newcomers absorb the subtleties of organizational reality - and in particular their own role identities - through symbolic interactions with other individuals; peers as well as superiors. Van Maanen (1978, p.20) states that "any person crossing organizational boundaries is looking for clues on how to proceed. Thus colleagues, superiors, subordinates, clients, and other work associates can and most often do support, guide, hinder, confuse, or push the individual who is learning a new role." Experienced members are seen as a potentially rich source of assistance to newcomers in acquiring context specific interpretation schemes (e.g., Louis, 1980b; Van Maanen, 1977), in learning norms, values, assumptions, and behaviours (e.g., Feldman, 1976, 1981 ), in providing solutions for work problems (e.g., Dornbush, 1955), and in providing psychological support and positive interpersonal v feedback (e.g.. House, 1981). Thi s assistance helps newcomers to establish a sense of contribution and worth, to feel more secure and acceptable, and to find their overall niche in the scheme of things (Hall, 1976; Katz, 1980; Van Maanen, 1976). Thus, by serving as primary associates and informal socializing agents, insiders have a major influence on the newcomer's adjustment to his/her new organizational role (e.g., Dubin et a l . , 1976; Hall, 1976; Louis, 1980b; Van Maanen, 1976). Empirical evidence further suggests that newcomers' initial perceptions and behaviours are significantly shaped by their conceptual schemes of the - 80 -work environment acquired through social interaction. Evan (1963), for example, demonstrated that unstructured interaction time with peers can assist newcomers to rapidly acquire the necessary and appropriate behaviour and attitudes towards their new jobs. Similarly, Feldman and Brett (1983) found that getting help and seeking out information and reassurance from others in the organization were the most favoured coping strategies of new employees. In summary, the distinction between these various components of adjustment recognizes the multifaceted nature of this concept and provides the basis for more accurate and precise formulation of the experimental hypotheses. DEFINITION OF TERMS Based on the preceding literature review, the following terms are defined for the purposes of the present study. The charismatic leadership style is complex. It encompasses several dimensions including the leader's ability to v i v i d l y articulate an ideological goal, to communicate high performance expectations and express confidence in followers' abilities to meet these expectations, and to empathize and communicate his/her understanding of his/her follower's needs (e.g., Bass, 1985; House, 1977; Oberg, 1972). The charismatic leader's effects on followers include inspiring heightened goals and involvement in the task and generating commitment to the leader (e.g.. House, 1977). - 81 -The considerate leadership style is characterized by leader behaviour that emphasizes concern, understanding, and warmth towards his/her group members, strong consideration of their feelings and needs, and participative, two-way communication (Chemers, 1984; Chemers & Rice, 1974; Fleishman & Peters, 1962). The structuring leadership style is characterized by leader behaviour that emphasizes task assignments within the group, goal definition, and establishment of work procedures and standards (Chemers, 1984; Chemers & Rice, 1974; Fleishman £ Peters, 1962). The considerate and structuring dimensions are conceptually related to the behaviour categories of employee centred and job centred developed at the University of Michigan (Katz £ Kahn, 1966), to the concepts labelled by Bales (1958) as "task facilitative" and "socioemotional", and to the behaviours described by House and Mitchell (1974) as "instrumental" and "supportive". Croup productivity norms refer to general standards or rules of conduct adhered to by the group which are intended to regulate and regularize group members' level of effort expended on a task (Festinger, Schachter, £ Back, 1950; Hackman, 1976; Homans, 1950; Kiesler, 1969a, 1969b; Thibaut £ Kelley, 1959; Vroom, 1969). Adjustment is the degree to which the individual copes with and adapts to the task and social demands of his/her environment (Matarazzo, 1972). It has two components: task and interpersonal. Task adjustment - 82 -refers to the individual's capacity to deal with his/her new task demands and the associated outcomes ( i . e . , job satisfaction, job related tension, role conflict, and role ambiguity) (Graen, 1976; Lofquist & Dawis, 1969; McGrath, 1976). Interpersonal adjustment refers to the individual's quality of interpersonal relations with his/her superior and co-workers (e.g., Feldman, 1976, 1981 ; Katz, 1980; Louis, 1980b). EXPERIMENTAL HYPOTHESES A brief recapitulation of the major themes from the research literature on leadership style and group productivity reviewed earlier are presented below, followed by the hypotheses derived from this literature that are tested in this study. It should be noted that some of the hypotheses are posed in a "null hypothesis" format. This is, in some cases, no differences between treatment groups are expected. While the statement of certain hypotheses in a null format might be considered inappropriate, it needs to be recognized that the hypotheses are formulated in this manner for conceptual as opposed to statistical purposes. More specifically, I have chosen to present these hypotheses following the logic of the literature rather than to state them in statistical terms. Furthermore, it needs to be underscored that since this dissertation was conceived during 1983, the hypotheses were formulated on the basis of the relevant literature available up to and including 1983. Therefore, - 82a -while the preceding literature review has incorporated theoretical and empirical work conducted since 1983, this work is excluded from consideration in the development and justification of the hypotheses. In particular, recent work by Willner (1984), by House (1985a), and by Bass and his associates (e.g., Avolio & Bass, 1985; Bass, 1985; Waldman et a l . , 1985) is omitted in the framing of the following experimental hypotheses. The relationship between these works and the empirical research in this study is discussed in Chapter IV. - 83 -MAIN E F F E C T S Leadership Style Task Performance and Task Adjustment Sociological and psychological treatments of charisma recognize that charismatic leaders obtain their effects by vi v i d l y articulating a transcendent goal which clarifies or specifies a mission for followers (e.g., Dow, 1969; House, 1977; Weber, 1947). Moreover, by simultaneously communicating high performance expectations and expressing confidence in followers' abilities to meet these expectations, charismatic leaders enhance followers' motivation and performance in support of the transcendent goal (e.g.. House, 1977; Smith, 1982). Accordingly, charismatic leaders should facilitate followers' task performance and adjustment. Early research on leader role emergence in small experimental discussion groups found that task oriented leadership was necessary for effective performance in work groups (Bales S Slater, 1955). The Ohio State leadership studies further indicated that employees engaged in nonroutine, creative, or analytic work reacted more favourably to leader initiating structure irrespective of the leadership questionnaire used (e.g., Schriesheim et a l . , 1976, p.302). Fiedler's (1967) contingency theory proposes that task oriented leadership is more effective when there is very little task structure. Similarly, path-goal theory asserts that the more ambiguous the task, the more positive the relationship between leader structuring behaviour and subordinate task satisfaction and expectancies. For subordinates with unclear role perceptions, structuring leadership - 84 -would help to clarify path-goal relationships, thereby increasing job satisfaction and expectancies (e.g., House S Dessler, 1974). Finally, in stressful or unfavourable situations, structuring leader behaviour is associated with high subordinate task performance and satisfaction (e.g., Chemers & Skrzypek, 1972; Fiedler, 1967; Fleishman, Harris, & Burtt, 1955; Schriesheim & Murphy, 1976). Therefore, structuring leaders should facilitate subordinates' task performance and adjustment. The leadership literature does not directly address the relative impact of s tructuring and charismatic leadership on individuals' task performance and adjustment. Therefore, as an exploratory hypothesis, it is suggested that there will be no differences between structuring and charismatic leaders on these outcome variables. T u r n i n g to considerate leadership, the Ohio State studies have generally found an absence of association between considerate leader behaviour and subordinate performance under dissatisfying task conditions (House & Baetz, 1979). Similarly, path-goal formulations suggest that under ambiguous task conditions, considerate leaders have no effect on individuals' task performance and satisfaction (e.g., Downey, Sheridan, S Slocum, 1975; Greene, 1979; House, 1971; House & Dessler, 1974; Stinson & Johnson, 1975). Therefore, considerate leaders should have no influence on individuals' task performance and adjustment. The body of leadership literature reviewed and discussed in this chapter suggests the following hypotheses. - 85 -Hypothesis 1. Individuals working under a charismatic leader will have higher task performance than will individuals working under a considerate leader. Hypothesis 2. Individuals working under a charismatic leader will report higher task adjustment than will individuals working under a considerate leader. Hypothesis 3. Individuals working under a structuring leader will have higher task performance than will individuals working under a considerate leader. Hypothesis 4. Individuals working under a structuring leader will report higher task adjustment than will individuals working under a considerate leader. Hypothesis 5. Individuals working under a charismatic leader will have the same level of task performance as individuals working under a structuring leader. 1 Hypothesis 6. Individuals working under a charismatic leader will report the same level of task adjustment as individuals working under a structuring leader. 1 As noted at the beginning of this section, these experimental hypotheses were formulated prior to the recent literature on transformational leadership presented by Bass and his associates (e.g., Bass, 1985; Waldman et a l . , 1985). - 86 -Adjustment to the Leader The theoretical literature suggests that a fundamental aspect of charisma is the extraordinary, intensely personal relationship between a charismatic leader and his/her followers. According to several scholars (e.g., Bensman & Civant, 1975; Dow, 1969; House, 1977; Tucker, 1970), the initial and continuing appeal of charismatic leadership is based on emotional rather than rational grounds in that the follower is inspired to give unquestioned obedience, loyalty, commitment, affection, and devotion to the leader. Therefore, by striv i n g to establish a strong affective bond with subordinates, charismatic leaders should facilitate subordinates' adjustment to the leader. The theoretical and empirical literature indicates that considerate leadership appears to have its primary effects on subordinates' social or psychological maintenance and intrinsic satisfaction (House S Dessler, 1974). For example. Bales and Slater (1955) observed that socioemotional oriented leadership provided social satisfaction for group members which resulted in the reduction of frustration and stress and an increase in two-way communication between superior and subordinates. The Ohio State leadership studies reported that considerate leadership was positively associated with subordinates' satisfaction with their leader (e.g., Bass, 1981, p.382). In addition, under conditions of role stress, several researchers have reported that considerate leader behaviour serves as a source of social satisfaction and support for the employee (e.g.. House, 1981; Seers et a l . , 1983; Sheridan & Vrendenburgh, 1979). Therefore, considerate leaders should facilitate individuals' adjustment to the leader. - 87 -The leadership literature does not directly address the relative effects of considerate and charismatic leaders on individuals' adjustment to the leader. Therefore, as an exploratory hypothesis, it is suggested that there will be no difference between considerate and charismatic leaders on this outcome variable. T u r n i n g to structuring leadership. Smith (1982) has postulated that the levels of liking , attraction, and affection for an instrumental leader would be moderate. Subordinates are attracted to such leaders as a specific source of extrinsic rewards they value and as a facilitator of task accomplishment, rather than as a source of emotional support. Empirical evidence lends support to this postulation. Bales (1958) found no relationship between task orientation and liking for leaders who made it possible for group members to give feedback and to raise objections, qualifications and questions. Therefore, individuals with structuring leaders should have neither positive nor negative adjustment to the leader. Hypothesis 7. Individuals working under a charismatic leader will report higher adjustment to the leader than will individuals working under a structuring leader. Hypothesis 8. Individuals working under a considerate leader will report higher adjustment to the leader than will individuals working under a structuring leader. -- 88 -Hypothesis 9. Individuals working under a charismatic leader will report the same level of adjustment to the leader as individuals working under a 2 considerate leader. Croup Productivity Task Performance As the preceding literature review suggested, in the high productivity condition, the group's role in advocating task accomplishment enhances individual task performance. In contrast, in the low productivity condition, the group's exertion of quota restricting pressures impedes individual task performance. Th i s leads to the following hypothesis. Hypothesis 10. Individuals in high productivity groups will have higher task performance than individuals in low productivity groups. Adjustment to the Task and to the Croup The previously reviewed literature does not directly address the influence of group productivity norms on individuals' adjustment to the task and to the group. However, the small group literature provides some guidance in formulating hypotheses for these two outcome measures. For example, Berkowitz (1954) reported that participants liked their partners (confederates) better when the latter were supposedly proficient on a task than when they were supposedly poor. Similarly, Again, it should be noted that this experimental hypothesis was framed prior to the work of Willner (1984) and Bass (1985). - 89 -Zander (1968) found that members of successful groups are more likely to experience satisfaction, form a favourable impression of themselves and other members of the group, and wish to continue to pursue the activity on which the group was successful. Thus to the extent individuals in high productivity groups perceive their co-workers as successfully accomplishing the task at hand, they should be more satisfied with their co-workers and with the task. In contrast, to the extent individuals in low productivity groups perceive their co-workers as exerting minimal effort on the task, they should be less satisfied with their co-workers and with the task. In addition, laboratory studies examining the effects of social infor-mation cues about a task on individuals' assessment of task characteristics are also relevant to the framing of hypotheses for the task adjustment measures. These studies have consistently shown that in comparison to negative social cues, positive social cues provided by confederate co-workers result in higher task satisfaction and more favourable perceptions of task characteristics (e.g.. G r i f f i n , 1983; White & Mitchell, 1979). Collectively, the research evidence suggests the following hypotheses. Hypothesis 11. Individuals in high productivity groups will report higher task adjustment than individuals in low productivity groups. Hypothesis 12. Individuals in high productivity groups will report higher adjustment to the group than individuals in low productivity groups. - 90 -INTERACTION E F F E C T S Leader Charismatic Behaviour and Croup Productivity The theoretical literature consistently suggests that charismatic leaders, by force of their personal qualities, are capable of having profound and extraordinary effects on followers (House, 1977, p.189). As House and Baetz (1979) note, these effects include commanding loyalty and devotion to the leader, inspiring followers to accept and execute the will of the leader, and heightening followers' confidence in their abilities to reach a goal. Given the potency of the charismatic leader, the effects of high or low group productivity norms should be nullified. Leader Structuring Behaviour and Group Productivity Individuals exposed to a structuring leader and in a high productivity group will experience role clar i t y ; group-sent roles and pressures to conform to these roles will be consistent with leader-sent roles. According to role theory (Kahn et a l . , 1964), to the extent that required role information is communicated clearly and consistently to a focal person, it will tend to induce in him/her an experience of certainty with respect to his/her role requirements, effective movement toward goals, and satis-factory job performance. In essence, both the leader and the group structure the individual's reality in mutually reinforcing ways, leading to high task performance, task adjustment, adjustment to the leader, and adjustment to the group. . - 91 -Individuals exposed to a structuring leader and in a low productivity group will experience role conflict; group-sent roles and pressures to conform to these roles will diverge from leader-sent roles. Role theory states that when the behaviours expected of an individual are inconsistent s/he will experience stress, become dissatisfied, and perform less effectively than if the expectations imposed on him/her did not conflict (e.g., Graen, 1976; Kahn et a l . , 1964; Rizzo, House, & Lirtzman, 1970). Moreover, individuals experiencing role conflict tend to reduce their trust, liking and respect for the role senders from whom the conflict stems. Empirical evidence supports these contentions; role conflict has shown to be directly related to psychological withdrawal from the group, job induced tension, and propensity to leave the organization and inversely related to job satisfaction, job involvement, performance, organizational commitment, and attitudes toward role senders (e.g.. Brief & Aldag, 1976; Greene & Organ, 1973; House & Rizzo, 1972; Johnson & Graen, 1973; Kahn et a l . , 1964; Miles, 1976; Van Sell, Brief, & Schuler, 1981; Yukl & Kanuk, 1981). Therefore individuals working under a structuring leader and in a low productivity group will have low task performance, task adjustment, and adjustment to the leader and to the group. Leader Consideration Behaviour and  Croup Productivity Individuals exposed to a considerate leader and in a high productivity group will experience role clarity; the considerate leader, by providing socioemotional support, and the high productivity group, by encouraging - 92 -conformance with high role expectations, act in a complimentary manner. Accordingly, individuals working under a considerate leader and in a high productivity group will have high task performance, task adjustment, and adjustment to the leader and to the group. Individuals exposed to a considerate leader and in a low productivity group will experience role ambiguity; that i s , they will lack the necessary information for effective performance in their new role. The considerate leader, by providing interpersonal concern, and the low productivity group, by advocating conformance with low performance norms, will fail to clarify individuals' role priorities. According to role theory, ambiguity should increase the probability that a person will be dissatisfied with his/her role, will experience anxiety, will distort reality, and will thus perform less effectively (Kahn et a l . , 1964; Rizzo et a l . , 1970). This argument has received empirical support. The evidence indicates that role ambiguity tends to be associated with lower job satisfaction, involvement, and performance, increased tension and anxiety, and unfavourable attitudes towards role senders (e.g., Beehr et a l . , 1976; Greene & Organ, 1973; Johnson & Stinson, 1975; Lyons, 1971; Rizzo et a l . , 1970; Van Sell et a l . , 1981). Therefore individuals working under a considerate leader and in a low productivity group will have low task performance, task adjustment, and adjustment to the leader and to the group. Collectively, the theoretical literature and empirical findings lead to the following experimental hypothesis. - 93 -Hypothesis 13. It is expected that leadership style will interact with group productivity such that: (a) Individuals exposed to a structuring leader and in a high productivity group will have i) higher task performance ii) higher task adjustment iii) higher adjustment to the leader and iv) higher adjustment to the group than individuals exposed to a structuring leader and in a low productivity group. (b) Individuals exposed to a considerate leader and in a high productivity group will have i) higher task performance ii) higher task adjustment iii) higher adjustment to the leader and iv) higher adjustment to the group than individuals exposed to a considerate leader and in a low productivity group. (c) Individuals exposed to a charismatic leader and in a high productivity group will have the same level of i) task performance ii) task adjustment iii) adjustment to the leader and iv) adjustment to the group as individuals exposed to a charismatic leader and in a low productivity group. - 94 -C H A P T E R II M E T H O D T h i s c h a p t e r c o n s i s t s o f t w o s e c t i o n s : ( 1 ) e x p e r i m e n t a l d e s i g n a n d ( 2 ) e x p e r i m e n t a l p r o c e d u r e . I n t h e f i r s t s e c t i o n , a n o t e o n l a b o r a t o r y s t u d i e s i s f o l l o w e d b y a d e s c r i p t i o n o f t h e e x p e r i m e n t a l d e s i g n a n d o f t h e e x p e r i m e n t a l t a s k . T h e o p e r a t i o n a l i z a t i o n o f t h e i n d e p e n d e n t , d e p e n a e n t , a n d i n d i v i d u a l d i f f e r e n c e v a r i a b l e s i s s u b s e q u e n t l y d i s c u s s e d . D e t a i l s o n t h e e x p e r i m e n t a l p r o c e d u r e a r e d e a l t w i t h i n t h e s e c o n d s e c t i o n . I n i t i a l l y , t h e s e l e c t i o n a n d t r a i n i n g o f t h e e x p e r i m e n t a l c o n f e d e r a t e s a n d v a l i d i t y c h e c k s o n t h e i r p e r f o r m a n c e a r e d i s c u s s e d . S u b s e q u e n t l y , t h e e x p e r i m e n t a l p a r t i c i p a n t s , s e t t i n g , a n d p r o c e d u r e a r e d e s c r i b e d . F i n a l l y , t h e i s s u e o f d e m a n d c h a r a c t e r i s t i c s i s a d d r e s s e d . E X P E R I M E N T A L D E S I G N A N o t e o n L a b o r a t o r y S t u d i e s W h i l e t h e v a s t m a j o r i t y o f l e a d e r s h i p s t u d i e s a r e p e r f o r m e d i n f i e l d s e t t i n g s ( H u n t , O s b o r n , S S c h r i e s h e i m , 1 9 7 8 ) , m a n y r e s e a r c h e r s a r g u e t h a t t h e r e s u l t s a r e i n c o n c l u s i v e a n d c o n f u s i n g ( e . g . , G i l m o r e , B e e h r , & R i c h t e r , 1 9 7 9 ; S a s h k i n S G a r l a n d , 1 9 7 9 ) . H e n c e t h e r e h a v e b e e n r e p e a t e d c a l l s i n t h e l i t e r a t u r e f o r l e a d e r s h i p s c h o l a r s t o r e t u r n t o t r u e - 95 -experimental designs conducted in the laboratory in order to sharpen our understanding of the conditions under which leader behaviour actually causes subordinate outcomes (e.g., Hollander, 1979; Sashkin & Garland, 1979). Many writers have examined the strengths and apparent weaknesses of the laboratory research method (e.g., Berkowitz & Donnerstein, 1982; Festinger, 1971; Fromkin & Streufert, 1976; Seashore, 1971; Weick, 1979). Thi s method allows testing of causal hypotheses, precision and clarity in the manipulation and measurement of variables, and control of sources of variation that may be significantly related to the dependent variables of interest (Sashkin & Garland, 1979, p. 67). Properties suggested as weaknesses of the laboratory research method include artificiality of the experiment, unrealistic homogeneity of the sample, demand characteristics of the experiment, and limited external validity of the study (e.g., Adair, 1984; Campbell et a l . , 1970; Cronbach, 1957; Dipboye & Flanagan, 1979; Orne, 1969; Sashkin & Garland, 1979; Weick, 1979). The validity of the experimental method for research on human behaviour has been defended on several grounds. The criticism of artificiality in laboratory experiments has been labelled a false issue (e.g., Campbell et a l . , 1970; Fromkin S Streufert, 1976; Sashkin & Garland, 1979; Weick, 1977, 1979). For example, according to Weick (1977, p. 124): One of the ironies of laboratory experimentation is that presumed liabilities turn out to be conceptual assets for organizational researchers. To illustrate, research participants are apprehensive about being evaluated, but so are ambitious - S6 -employees. Laboratory tasks require limited sk i l l s , ignoring the 'rest' of what the person brings to the laboratory, but the same holds true with a division of labour and partial inclusion. Relationships between experimenter and respondent involve asymmetrical power, but the same holds true for superiors and subordinates. Participants seldom know why they are doing the things they do in laboratories but employees often operate under similar conditions of ignorance and faith. Participants in laboratory groups seldom know one another intimately, but the same is true in organizations where personnel transfers are common, where temporary problem solving units are the rule, and where impression management is abundant. People parti c i -pate in experiments for a variety of reasons, but the decision to participate in an organization is similarly over-determined. Finally, people are suspicious of what happens to them in laboratories but so are employees suspicious as they become altered to the reality of hidden agendas and internal politics. Therefore, in Weick's (1977, 1979) view, while participants need to adjust to unfamiliar laboratory tasks and settings, there is reason to believe that their response is similar to those in other work situations. Laboratory researchers have typically utilized participants who are most readily available - college students. Although this may create a problem of generalizability, Sashkin and Garland (1979, p. 69) have observed that college students "may be more heterogeneous with respect to characteristics associated with leadership than would be a group of managers selected from within one company or industry, which is the kind of sample most often used in field studies of leadership." They further argue that "since large numbers of college students, particularly those in business schools, will one day occupy managerial positions in a wide variety of organizations, there may be greater justification for generalizing from this population to 'managers in general' than from a sample of managers studied within one 'real' organization" (Sashkin & Garland, 1979, p. 69). - 97 -Discussions in the literature regarding subject effects have failed to confirm the assumption that individuals in a laboratory setting adopt either a good subject role, seeking to confirm the experimenter's hypothesis, or a subversive subject role, attempting to sabotage the experiment (e.g., Berkowitz & Donnerstein, 1982; Kruglanski, 1975; Silverman, 1977; Weber £ Cook, 1972). Instead, their concern with how they will be judged (evaluation apprehension) is more important than their concern about fulfilling the experimenter's expectancies or confirming his/her hypothesis (Berkowitz & Donnerstein, 1982; Weber & Cook, 1972). Although occasion-ally the participants' desire to "look good" may induce them to respond in accordance with the experimenter's expectancies, it appears more frequently that their desire to appear competent and "normal" induces them to respond naturally without being influenced by the experimenter's expectancies (Weber & Cook, 1972). With regard to the external validity of an experiment, many scholars have argued that the generalizability of findings should be relatively narrow, restricted to a small population of participants (e.g., Dipboye S Flanagan, 1979; Sashkin & Garland, 1979; Weber S Cook, 1972). While some scholars have seriously questioned the common belief that field settings provide for more generalizability of research findings than laboratory settings do (e.g., Berkowitz & Donnerstein, 1982; Dipboye S Flanagan, 1979), it is v e r y possible that an effect observed to be significant in the laboratory would not be sustained under conditions in which a multitude of other variables were allowed to operate simultaneously and over an extended period of time (Sashkin S Garland, 1979). - 98 -In summary, the foregoing literature review suggests that while the external validity of laboratory experimentation may be questionable, other apparent sources of weakness of this methodology have failed to be confirmed. Moreover, the laboratory research method permits the testing of causal hypotheses, precision and clarity in the manipulation and measurement of variables, and control of extraneous sources of variation. Therefore, laboratory experimentation has a justifiable place in the study of organizational leadership. Experimental Design This laboratory experiment examined the effects of three different leadership styles and two levels of group productivity on individuals' adjustment and performance. A 3 x 2 factorial design ( K i r k , 1982) with equal cell sizes was used (see Table 1). In order to maintain as much control as possible over the experimental conditions, each of the groups consisted of four individuals: the leader, two co-workers, and the participant. Depending upon the particular leadership treatment being implemented, the formally designated leader demonstrated a s t r u c t u r i n g , considerate, or charismatic style of leadership. To accomplish this, the leader placed in charge of the group was an experimental confederate completely instructed in her role. (See the Confederate section on p.125 for a complete explanation). For the group productivity treatment, two co-workers either encouraged the participant to do the task (high productivity condition) or - 99 -Table 1 Experimental Design Croup Productivity Treatments High Productivity Low Productivity Charismatic y i * y 2 Leadership Structuring y 3 Treatments Considerate p 5 y 6 *average values for adjustment and performance for the participants assigned to the various treatment cells. - 100 -discouraged the participant from doing the task (low productivity condition). Since substantial control was required to accomplish this treatment, the co-workers were also experimental confederates. (See the Confederate section on p. 125 for a complete explanation.) Thus, confederates participated in all of the treatment replications. Only the fourth group member was an experimental participant in each treatment. By allowing the group memberships to vary only in the fourth individual, more exact replications within and between the various experimental conditions were attained. Finally, it should be noted that in designing the experiment the author was mindful of the ethical issues involved. Ethical approval for this experiment was granted by the University of Briti s h Columbia's (B.C.) Committee for Research on Human Subjects, prior to its implementation. Experimental Task In designing the experimental task, there were several essential criteria that had to be satisfied. F i r s t , to simulate adjustment to a completely new work situation, in addition to the experimental setting and the organizational members ( i . e . , the leader and the co-workers), the task needed to be ambiguous. Moreover, to test the hypotheses of the present study in terms of "the relative influence of leadership style and group productivity norms on individuals' adjustment to and performance in a new work environment, an unstructured task was required. Second, the task had to be realistic and have high face validity in order to accurately and - 101 -meaningfully portray events encountered in a business environment. T h i r d , in order to permit the participant to develop clear perceptions of the leadership style and co-worker productivity, the task needed to be conducive to considerable interaction among the entire experimental group - the leader, the co-workers, and the participant. In this way, the leader was able to display to the participant that her style was the same towards all members of the work group ( i . e . , the co-workers and the participant). Fourth, to control for participants' prior task experience, the task needed to be unfamiliar to them. Finally, to ensure the participants would think about the present work situation, the task needed to be both lengthy and involving. Based on the above cr i t e r i a , an in-basket exercise developed by Zenger Miller and Associates (1975) was selected as the experimental task (see Appendix A ) . The in-basket exercise is an elaborate, realistic situational test (e.g.. Crooks, 1977; Frederikson, 1962; Meyer, 1970). The situation is relatively unstructured, with a multitude of possible solutions and of means to those solutions (e.g., Frederikson, 1962; Lopez, 1966). In the present study, the in-basket exercise requires the participant to think of him/herself as Rex/Rhonda Andrews, General Manager of Marketing for the Cogen Products Marketing Division. In his/her assumed role, the participant must act on separate in-basket items, including letters, reports, and memoranda, by r e f e r r i n g , delegating, making decisions, requesting further information and, in general, exercising good administrative judgement. S/he has to record everything s/he does in writing and provide a brief statement of the reason(s) for the actions undertaken. A f t e r an item is completed, the participant is requested to place it in an out-basket. - 1 0 2 -The in-basket exercise was divided into two sections. The f i r s t section consisted of 15 memos with a time limit for completion of 45 minutes. The second section constituted an optional task which comprised a further 5 memos with a time limit for completion of 15 minutes. These time limits were established so that each participant would work on the exercise an equal amount of time and so that none of the participants would be able to complete all of the memos within the specified time periods. T h i s was done to ensure that satisfaction ratings would be derived from the leader-subordinate and/or co-workers-subordinate interaction(s) rather than from completion of the task. Operational Definitions of the Independent Variables Leadership Style The values, verbal and nonverbal behaviours, interaction style, and paralinguistic cues for each leadership style are delineated in Table 2. Each style is described below. Charismatic Style The operationalization of charismatic leadership was primarily derived from the psychological literature on intraorganizational charisma (e.g., Berlew, 1974; House, 1977; Oberg, 1972; Yukl & VanFleet, 1982). Based on this literature, the domain of charismatic behaviours was selectively sampled; hence a composite picture of a charismatic leader emerged. Specifically, the charismatic leadership style was operationalized as high charisma: the leader articulated an ideological goal ( i . e . , goal in the Table 2 Operational Definition of Leadership Styles Leadership Style Dimensions Orientation Verbal Behaviours Nonverbal 1. Charismatic Concerned with inspiring heightened goals and involvement in the task. Generates followers' commitment and loyalty. Articulates an ideological goal. Guides participants in altering their behaviours, ideas, and values in order to perform the task. Communi-cates high performance expectations and exhibits confidence in participants' ability to meet such expectations. Alternates between pacing and sitting on the edge of her desk, leans towards participants, maintains direct eye contact, has a relaxed posture, animated facial expressions. 2. Considerate Concerned with the social and emotional tensions and needs of the participants. Engages in participative two-way conversation. Expresses concern for the personal welfare of the participants. Reassures and relaxes the participants. Emphasizes the comfort, well-being and satis-faction of the participants. Sits on the edge of her desk, leans towards participants, maintains direct eye contact, has a relaxed posture, friendly facial expressions (smiling). 3. Structuring Concerned with accom-plishing the task. Results oriented. Emphasizes the meeting of dead- Sits on the edge of her lines and quantity of work to be accomplished. Provides detailed direction and assist-ance to participants to get the task completed. Schedules the work to be done. Maintains definite standards of performance. desk, has periodic direct eye contact, neutral facial expressions. Table 2 continued Operational Definition of Leadership Styles Leadership Dimensions Style Interaction Style Paralinguistic Cues 1. Charismatic Projects a powerful, dynamic, confident image. Ability to emphathize with, and communicate her understanding of the needs of her followers (generates the feeling "she understands me"). High level of intonation: engaging voice tone. captivating. 2. Considerate Friendly, approachable, responsive to others, appreciative, willingness to listen. Considerable intonation: friendly voice tone. warm. 3. Structuring Neutral: neither warm nor cold. Some intonation: businesslike, factual voice tone. - 1 0 5 -sense of an overarching purpose for which to s t r i v e ) , communicated high performance expectations and exhibited confidence in subordinates' ability to meet these expectations, and empathized with the needs of his/her subordinates. The highly charismatic leader also projected a powerful, confident, and dynamic presence (e.g.. House, 1977; Oberg, 1972; Sashkin, 1977; T r i c e & Beyer, 1984). The operationalization of nonverbal behaviours and paralinguistic cues ( i . e . , volume, speech rate, intonation) was based on the communication literature (Edinger & Patterson, 1983; Friedman, Prince, Riggio, & DiMatteo, 1980; Friedman & Riggio, 1981). According to the research conducted by Friedman and his colleagues, nonverbal emotional expressive-ness ( i . e . , the use of facial expressions, voice, gestures, and body movement to transmit emotion) is an essential characteristic of charisma, the ability to move, inspire, or captivate others. Paralinguistically, the charismatic leader was trained to have a captivating, engaging voice tone. To capture the dynamism and energy of charisma, nonverbal ly the charismatic leader alternated between pacing and sitting on the edge of her desk, leaned toward the participant, maintained direct eye contact, and had a relaxed posture and animated facial expressions (Edinger & Patterson, 1983; Friedman et a l . , 1980, 1981). It should be underscored that charismatic leadership is a qualitatively different phenomenon than a high structure-high consideration (in Blake and Mouton's (1964) terms, a 9,9 style) style of leadership. While charisma contains some elements of consideration, that i s , an ability to empathize and communicate with subordinates, it also encompasses several - 1 0 6 -other leader behaviours such as articulating a compelling vision of a desired future state of affairs, building a confident and dynamic image and so on. Moreover, charisma has minimal overlap with structuring leader behaviours. The structuring leader focusses st r i c t l y on the task, organizing and defining the way work is to be done. In contrast, the charismatic leader does not explicitly define the task, rather s/he provides an overarching goal to strive for and expresses confidence and excitement in the subordinate's ability to reach that goal. Considerate Style The operational ization of considerate leadership was based on definitions of consideration (Bass, 1981; Stogdill, 1971), items in the SBDQ (Fleishman, 1957), LBDQ (Hemphill S Coons, 1957), and LBDQ-XII (Stogdill, 1963), and prior operationalizations of considerate leader behaviour in experiments using confederates as leaders (e.g., Katz, 1977; Lowin, Hrapchak, & Kavanagh, 1969; Tjosvold, 1984). The considerate style was operational ized as high consideration: the leader exhibited concern for the personal welfare of the participant, engaged in participative two-way conversations, emphasized the comfort, well-being, and satisfaction of the participant, and was friendly and approachable. Rather than actually structuring the task for the participant, the considerate leader focussed on the interpersonal aspects of the task by expressing reassurance and support, making the participant feel at ease, and stressing the importance of high morale. Thus the considerate leader focussed on the social and emotional tensions and needs of the participant in the work situation. « - 1 0 7 -The operationalization of paralinguistic cues and nonverbal behaviours was based on the communication and counselling literatures (e.g., Edinger S Patterson, 1 9 8 3 ; LaCrosse, 1 9 7 5 ; Mehrabian, 1 9 7 2 ; Truax & Carkhuff, 1 9 6 7 ) . Paralinguistically, the considerate leader was trained to have a warm voice tone. Nonverbally, the considerate leader sat on the edge of her desk, leaned toward participants, maintained direct eye contact, and had a relaxed posture and friendly facial expressions (i . e . , smiling, positive head nods). Structuring Style The operationalization of structuring leadership was based on definitions of initiating structure (Bass, 1 9 8 1 ; Stogdill, 1 9 7 4 ) , items in the SBDQ (with the exception of punitive and autocratic items) (Fleishman, 1 9 5 7 ) , LBDQ (Hemphill S Coons, 1 9 5 7 ) and LBDQ-XII (Stogdill, 1 9 6 3 ) , and prior operationalizations of structuring leader behaviour in experiments using confederates as leaders (e.g., Katz, 1 9 7 7 ; Lowin, Hrapchak & Kavanagh, 1 9 6 9 ; Tjosvold, 1 9 8 4 ) . The structuring style was operationalized as high structure: the leader explained the nature of the task, decided in detail what should be done and how it should be done, emphasized the quantity of work to be accomplished within the specified time period, and maintained definite standards of work performance. The structuring leader also answered any task related questions. Thus the structuring leader focussed on defining and organizing the way work was to be done. A notable limitation of the leadership literature is the failure to specify the interaction style of structuring leaders. There are few linkages between the personality traits of structuring leaders and the behaviours - 108 -they display. Thus it is unclear whether structuring leaders are cool and aloof, warm and friendly, and so on. In the present study, structuring leaders acted in neutral fashion towards people: they were neither warm nor cold. Neutrality in the structuring leader's interaction style was achieved both verbally and nonverbally. Verbally, the leader factually provided background information on the project, read the instructions aloud, explained the task, and, in general, acted in a businesslike manner. Paralinguistically, the structuring leader was trained to have a moderate level of speech intonation. Nonverbally, the structuring leader sat on the edge of her desk, maintained intermittent eye contact, and had neutral facial expressions ( i . e . , absence of smiling and positive head nods)(Edinger 8 Patterson, 1983; LaCrosse, 1975). To ensure that any differences that were detected could be attributed to the manipulation of the leadership style variables in accordance with the operational definitions described above, attempts were made to minimize any sources of extraneous variation. One possible source of variation was the leader's attire. Accordingly, for all leadership styles, the leaders wore identical attire. Specifically, since the participants expected to encounter a professional business manager, the leaders wore dark coloured conserva-tive business suits. Croup Productivity The values, verbal and nonverbal behaviours, interaction style, and paralinguistic cues for the group productivity conditions are outlined in Table 3. The operationalization of group productivity was based on prior - 109 -Table 3 Operational Definition of Group Productivity Dimensions 1. Orientation Group Productivity 2. Behaviours a) verbal b) nonverbal 3. Interaction Style High Productivity Low Productivity 1. To encourage the participant to do the exercise. Advocate high task productivity. 2. a) Expresses consider-able interest in and enthusiasm about the exercise. Has a serious attitude about the project. b) Displays intense involvement in the exercise: reads in a concentrated manner; writes steadily. Friendly, approachable, interested in the participant. 1. To discourage the participant from doing the exercise. Advocate low task productivity. 2. a) Expresses dis-interest in the exercise. Has a casual attitude about the project. c) Displays boredom and lack of involve-ment in the exercise: leans back in his/her chair, looks out of the window, sighs, doodles, plays with his/her pen. Friendly, approachable, interested in the participant. 4. Paralinguistic 4. Warm voice tone. Cues 4. Warm voice tone. - 110 -experimental operationalizations in which positive or negative task performance cues were provided by confederate co-workers (e.g.. White, Mitchell, & Bell, 1977; White & Mitchell, 1979). In essence, the co-workers served as models for appropriate work behaviour by exhibiting either highly productive or minimally productive behaviour. The co-workers were seated in the same room as the participant and worked on their own tasks which involved analyzing business cases. In both conditions the co-workers appeared friendly towards and interested in the participant. The operationalization of each condition is described below. High Productivity Condition In the high productivity condition, the co-workers advocated high task productivity to the participant. That i s , following White and his colleagues (1977, 1979), the co-workers made positive comments about the task itself and completed the task activity. Verbally, they expressed interest in and enthusiasm about the nature and relevance of the task. To further facilitate the participant's involvement in the task, they made positive and encouraging statements about doing the exercise. In essence, the co-workers attempted to build the participant's motivation to do the task. Nonverbally, the co-workers exhibited intense involvement in the task by reading in a concentrated manner and writing steadily. Moreover, when the leader asked the participant and each co-worker to perform an optional task, the co-workers further demonstrated their keen interest by willingly complying with the leader's request. - 111 -Low Productivity Condition In the low productivity condition, the co-workers advocated low task productivity to the participant. That i s , following White and his associates (1977, 1979), the. co-workers made negative comments about the task and did not complete the task activity. Verbally, they expressed disinterest in and boredom with the task and made negative remarks about the usefulness and relevance of the exercise. In essence, the co-workers attempted to reduce the participant's motivation to do the exercise. Nonverbally, the co-workers displayed a lack of involvement in and commitment to the task by leaning back in their chairs, looking out of the window, sighing occasionally, and playing with their pens. Moreover, their disinterest in the task was further reflected in their refusal to perform an optional task when the leader requested this of them. As discussed in Chapter I, there are several group characteristics which enhance member acceptance of group supplied performance norms such as having high status, acting as credible, competent sources of information, and sharing identical views of work (Hackman, 1976). The co-worker roles were developed accordingly: (a) to enhance their status, co-workers portrayed third and fourth year commerce students; (b) to appear credible, the co-workers claimed to have already done the in-basket exercise; and (c) to ensure unanimous views, both co-workers espoused either high task productivity or low task productivity, depending on the experimental treatment being implemented. There was opposite sex pairing of co-workers ( i . e . , one co-worker was male, the other was female). One co-worker acted as a fourth year - 112 -commerce student, reported to the secretary's office 10 minutes prior to the participant's scheduled arrival time, initiated conversation with the participant, and left f i r s t after the optional task. (See the Experimental Procedure section on p. 137 for complete details.) The other co-worker portrayed a thi r d year commerce student and reported to the secretary's office 5 minutes after the participant's scheduled arrival time. (See the Experimental Procedure section on p. 137 for complete details.) The assignment of these roles to the male and female co-workers was counterbalanced within each experimental condition. The co-workers wore typical commerce student attire: the male co-workers wore jeans or corduroy slacks and casual shirts and carried either a briefcase or a knapsack; the female co-workers wore slacks or skirts and sweaters. Experimental Scripts To maximize the consistency in portrayal of a leader role or a co-worker role, the leader and co-worker comments were made according to prescribed scripts (see Appendix B). For the leader roles and for the co-worker roles, the scripts are identical in content (except for the manipulation of the independent variables) and are approximately the same length. Manipulation Checks As discussed above, leaders and co-workers behaved in a pre-specified manner. To determine whether the participants correctly perceived the intended treatment, a number of post-experimental manipulation checks were made. - 113 -For the leadership style manipulation checks, 31 statements, 10 pertaining to the considerate style, 10 pertaining to the structuring style, and 11 referring to the charismatic style, were rated on 5-point Likert type scales (see Appendix C ) . The group productivity manipulation checks consisted of three items on the post-experimental questionnaire: two bipolar items on the group atmosphere scale (enthusiastic - unenthusiastic; productive - nonproductive) and one item on the role conflict scale (I received incompatible requests from my manager and the other students). Operational Definitions of the Dependent Variables Task Performance Task performance was measured in two ways. F i r s t , since self perceived task competence is an important outcome of successful adjustment to a new job (e.g., Feldman, 1977; Fisher, 1983), participants rated their individual performance on the in-basket exercise using a modification of Mott's (1972) 8-item measure of unit performance (see Appendix D). Supportive validity and reliability data for this measure are reported by Mott (1972). In addition, studies which have modified this measure to have participants provide self-ratings of their individual performance have reported a reliability coefficient alpha of .84 for this measure (e.g., Fulk & Wendler, 1982; C. Schriesheim, 1979; J. Schriesheim, 1980). Task performance was also measured by scoring the in-basket exercise according to dimensions derived from the empirical literature (e.g., Bentz, 1981; Frederiksen, 1962; Hemphill e t a l . , 1962; Lopez, 1966; Meyer, 1970; Thornton & Byham, 1982) as well as discussions with experts - 114 -in the area of assessment centres (Bentz, personal communication, January 9, 1984; Dunnette, personal communication, January 9, 1984). To measure the participant's quantitative performance on the in-basket exercise, the number of in-basket items attempted within the 45 minute time period and within the 15 minute time period was summated. The overall quality of the participant's performance on the in-basket exercise was also measured. Thi s measure was a subjective judgement based on raters' impressions of the managerial skills demonstrated in handling the in-basket items (Meyer, 1970). It was not solely concerned with the number of items addressed by the participants; an individual could handle all of the in-basket items and still receive a low score for quality if the items were disposed of in "inappropriate" ways. By the same token, one might ignore or otherwise fail to handle many items, and this would also result in a poor quality score. Therefore, a high quality score reflects the ability to deal with the in-basket items in a qualitatively excellent way and a general willingness to tackle at least the majority of the items. Quality of the in-basket performance was rated on a 5-point Likert type scale. A rating of one designated poor managerial performance whereas a rating of five designated excellent managerial performance. The scale included five dimensions of the quality of managerial task performance: 1. The establishment of priorities; 2. The relation to other items or background information in handling the items; - 115 -3. The systematic organization of the items by definitely scheduling work and setting up a calendar; 4. The demonstration of administrative skills such as appropriate use of delegation, obtaining more information, discussing with others, or handling directly; and 5. The provision of a rationale for the decisions. Three rating points on the scale were operationally defined. A rating of one indicated that the participant did not establish priorities, ignored background information in tackling the items, made indefinite plans (e.g., see me; let's d i s c u s s ) , primarily handled all items him/herself with minimal use of delegation or consultation with others, and provided no rationale for decisions. A rating of three indicated that the participant frequently established priorities, had some awareness of the interrelationships among items, definitely scheduled important meetings, demonstrated good managerial skills such as asking for more information, and provided a rationale for major decisions. A rating of five indicated that the participant consistently established priorities, interrelated items and background information when handling the items, systematically organized his/her work by using the calendar provided and by definitely scheduling meetings, demonstrated excellent administrative skills such as appropriately delegating matters to others, and provided detailed reasons for decisions. Two managers (1 male, 1 female) with seven years and ten years of business experience respectively, were trained to rate the quality of in-basket performance. The entire training procedure required approximately 20 hours. Further, since the rating of the in-basket exercise was conducted over a 3 week period, prior to each rating session, the judges were retested on pilot data in an attempt to maintain the consistency and reliability of their ratings. - 1 1 6 -The judges' training for rating the quality of in-basket performance consisted of three components: (1) doing the in-basket exercise, (2) discussing the quality rating scale, and (3) practicing rating the quality of performance on pilot data until a 90% accuracy criterion level was met. After reading an entire in-basket for a participant, the judges independently rated the overall quality of managerial performance on the exercise. A f t e r the judges had rated a randomly ordered set of 20 exercises, they compared their ratings and discussed all judgements on which they had not yet reached unanimous agreement, thus arriv i n g at a master judgement for each in-basket. To determine the interrater reliability on the quality rating scale, the average reliability was computed (Ebel, 1951). Consistent with previous research (e.g., Bentz, 1984; Meyer, 1970), the average interrater reliability on the quality ratings was .87. It should be noted that in Ebel's method of computing the interrater reliability, the between raters' variance is excluded from the e r r o r term. Th i s precludes the possibility of generalizing these results to other situations. Further, Ebel's formula gives the reliability of the average rating of the two judges. A chi-square test (Lawlis S Lu, 1972) was then conducted on the interrater agreement for the quality of performance scale. Interrater agreement was defined as identical ratings between the two judges. The interrater agreement was significant, x 2 U ) = 7.80, £ < .01, indicating that the observed agreement was greater than the agreement which could be expected on the basis of chance. To determine whether interrater - 117 -agreement was high, moderate, or low, the T index for agreement (Tinsley & Weiss, 1975) was computed. A fairly high agreement between the two judges was obtained, T = .78. In addition to quantitative and qualitative measures of task performance, the number of courses of action taken by the participants on the in-basket exercise was measured. Specifically, for each in-basket exercise, the judges counted the number of different courses of action proposed. A f t e r the judges had computed the total number of actions taken for a randomly ordered set of 20 exercises, they compared their computations and discussed all judgements on which they had not yet reached unanimous agreement, thus a r r i v i n g at a master judgement for each in-basket. According to Bentz (1981), individuals who take many courses of action are productive and tend to see a variety of implications in the items. Therefore, this measure seems to incorporate both quantitative and qualitative aspects of performance. In summary, there were four measures of task performance: (1) self-rated performance, (2) the number of in-basket items attempted, (3) the quality of in-basket performance, and (4) the number of courses of action suggested. Task Adjustment The measurement of task adjustment incorporated both outcome oriented and process oriented variables. With regard to the former. - 118 -Lofquist and Dawis 1 (1969) research suggests that in monitoring an individual's work adjustment we should focus on his/her satisfaction with his/her role. Accordingly, based on Scarpello and Campbell's (1983) empirical evidence that global measures of job satisfaction are not equivalent to the summation of many facet responses, both general and specific job satisfaction measures were used in the present study. Two items designed to tap overall job satisfaction were adapted from Hackman and Lawler's (1971) questionnaire on employees' reactions to work (see Appendix E ) . To measure the degree to which participants were satisfied with particular aspects of their task, the Job Descriptive Index (JDI) (Smith, Kendall, 6 Hulin, 1969) satisfaction with work scale was used (see Appendix F ) . In this scale, the respondent is asked to check whether each item in a series of short statements and adjectives applies to the task. The participants marks Y if it does, N if it does not, and ? if s/he cannot decide. Work satisfaction is computed by summing the weights assigned to the response associated with each item (Smith et a l . , 1969). According to Smith and her colleagues (1969), the corrected split-half internal consistency for the satisfaction with work scale is .84. In addition, the psychometric properties, reliability, and validity of the JDI have been extensively examined, generally with highly favourable results (e.g., Dunham, Smith, S Blackburn, 1977; Cillet £ Schwab, 1975; Imparato, 1972). With regard to-process oriented variables, Graen's (1976) role making model suggests that role ambiguity and role conflict are crucial variables in the role definition process. Thus, this study adapted Rizzo, House, and Lirtzman's (1970) scales of role ambiguity and role conflict (see Appendices - 119 -C and H, respectively). Role ambiguity is defined in terms of the lack of clarity of role expectations and demands and lack of predictability of behavioural outcomes. Role conflict is defined as the perception of incompatible or incongruent demands placed on the role incumbent. Participants were asked to indicate on a 7-point true-false response scale, the extent to which the items were descriptive of their task situations. Rizzo and his colleagues (1970) have reported reliability estimates ranging from 0.78 to 0.84 for these instruments. Studies of the psychometric properties (Schuler, Aldag, & Brief, 1977) and item response character-istics (House, Schuler, & Levanoni, 1983) of the role ambiguity and role conflict scales have yielded supportive results. To assess the frequency with which the participants were bothered by feelings of role ambiguity and role conflict, work overload, inadequate performance feedback, and other stressful task conditions, Kahn, Wolfe, Quinn, and Snoek's (1964) index of job related tension was adapted (see Appendix I). A recent study by MacKinnon (1978) has empirically supported the stability of the factor structure of this index. In addition, Cronbach's alpha reliability coefficients ranging from .73 to .84 have been reported for this index (e.g., Jamal, 1984). In summary, there were five measures of task adjustment: (1) general task satisfaction, (2) specific task satisfaction, (3) role ambiguity, (4) role conflict, and (5) work related tension. - 120 -Interpersonal Adjustment As discussed in Chapter I, a crucial component of the newcomer's adjustment to his/her new social reality is establishing interpersonal relationships with his/her superior and co-workers. The success of this adjustment may be indicated by three different measures: 1. The quality of the participant's relationship with the leader as measured by a series of Likert scales adapted from Tjosvold (1984) (see Appendix J ) . 2. The quality of the participant's relationship with the co-workers was measured by the group atmosphere scale (Fiedler, 1967) (see Appendix K). According to Fiedler (1962), this scale indicates the perceived pleasantness or stressfulness of the group interaction. The items consisted of a series of bipolar adjectives using an 8-point semantic differential type format. Supportive reliability data (corrected split-half reliability is .90) and validity data are presented by Fiedler (1967). In addition, as reported by Martin and Hunt (1980), the group atmosphere items loaded on the same factor as a series of attraction to group items developed by Seashore (1954). 3. The degree to which the participant is personally committed to and motivated by the leader and/or the co-workers as measured by his/her willingness to perform a subsequent task. It should also be noted that two other measures were utilized to further examine the participants' adjustment to their new work situation: (1) the co-workers unobtrusively recorded the participants' reactions to the task (see Appendix L) and (2) the participants noted their impressions of the Management Trai n i n g Project prior to completing the post experimental questionnaire (see Appendix M). As part of the contract between the author-and her committee members, these process measures were not content analyzed for this dissertation. - 121 -Operational Definitions of the Individual Difference Variables "It is generally agreed that behaviours and attitudes tend to be a function of the interaction between the person and his or her environmental situation. As a result, employee reactions are most likely influenced by both their psychological and personality characteristics as well as by their definitions of and interactions with the overall work setting" (Katz, 1980, p.119). Therefore, as Jones (1983) and Katz (1980) argue, it is important to study how different kinds of individuals negotiate their way along the job longevity continuum. Moreover, the leadership literature suggests that follower characteristics need to be examined since individual differences moderate the relationship between leadership style and subordinate work attitudes and performance (e.g., Chemers, 1984; House & Dessler, 1974). Accordingly, this study measured several indi -vidual characteristics that may influence the individual's adjustment to his/her new work situation including need for achievement, tolerance for ambiguity, and need for affiliation. Need for Achievement In examining the predictors of managerial success, Hall (1976) concluded that there is some evidence to suggest that individuals with a higher need for achievement are likely to be more successful in their managerial careers. As Katz (1980, p. 119) notes, "one possible explanation ... is that such individuals are more successful because they are more adept at handling their many socialization encounters as they cycle up the managerial ladder. Rather than simply waiting for others to define for them the many aspects of their new environments, managers with the aforementioned characteristic ... may be more active in seeking - 122 -such definitions or in defining their own sense of reality, including goals and expectations. 1 1 T h i s contention has received mixed empirical support. For example, Abdel-Halim (1981) found that managers with high need for achievement who worked on enriched high scope jobs seemed to be unaffected by high levels of role ambiguity. In contrast, Johnson and Stinson (1975) reported that military officers with high need for achievement were more dissatisfied when they perceived their task assignment to be relatively ambiguous or when they received conflicting demands from their role set. Need for achievement was measured by Jackson's (1974) Personality Research Form (PRF) need for achievement scale using forms A and B combined (see Appendix N). T h i s scale consists of 40 dichotomous items measuring the motivation of the individual to accomplish difficult tasks, to maintain high standards, and to put forth effort to attain excellence. Jackson (1974) and Mayes and Ganster (1983) present extensive empirical evidence demonstrating highly acceptable reliability (corrected odd-even reliability is .86) and convergent and discriminant validity for this scale. Tolerance for Ambiguity Several researchers consider tolerance for ambiguity as essential in managing the stress imposed by an uncertain environment (e.g., Lorsch & Morse, 1974; Sarbin & Allen, 1968). For example, Kahn et al. (1964) found a significant relationship between role ambiguity and job related tension for individuals classified as high in need for cognition. No relationship was found for individuals classified as low in need for cognition. In addition, Lyons (1971), in exploring the moderating effects - 123 -of need for clarity and propensity to leave, voluntary turnover, job tension, and job satisfaction among community hospital nurses, found significantly greater relationships between role clarity and turnover for individuals classified as high in need for clarity than for individuals classified as low in need for cla r i t y . Tolerance for ambiguity was measured by the revised Rydell-Rosen Ambiguity Tolerance scale (Kirton, 1981) {see Appendix 0 ) . Tolerance for ambiguity is defined as an individual's relative preference for undefined, unstructured, and ambiguous settings. T h i s 11-item scale has reasonably high internal consistency (KR-20 = .71), replicates closely on fairly large general population samples, and produces expected and consistent negative relationships with some other concepts which might be expected from theorizing including dogmatism, inflexibility, and conservatism (Kirton, 1981). Need for Affiliation A number of researchers have presented evidence to show that individuals high in need for affiliation are concerned about establishing and maintaining relationships with others (e.g., Jackson, 1974; McClelland, 1975). For example, such individuals tend to interact more often with work associates and therefore be influenced by them, to remain physically closer to others, to laugh more, and to engage in more reciprocal dialogue than individuals with weaker affiliative needs (e.g., Lansing & Heyns, 1959; Thomas & G r i f f i n , 1983). Furthermore, the performance and satisfaction of individuals with high affiliative needs is enhanced by superiors who demonstrate warmth and concern for others (e.g., French, - 124 -1955; House S Dessler, 1974; McKeachie et a l . , 1966). Finally, as House (1977, p. 203) has observed, "when task demands require affiliative behaviour, as in the case of tasks requiring cohesiveness, team work, and peer support, the arousal of the affiliative motive becomes highly relevant to performance and satisfaction." Need for affiliation was measured by Jackson's (1974) PRF need for affiliation scale using forms A and B combined (see Appendix P). T h i s scale consists of 40 dichotomous items measuring the motivation of the individual to seek and maintain social relationships and to readily accept people. As reported by Jackson (1974), the corrected odd-even reliability of this scale is .88. In addition, this scale has acceptable convergent and discriminant validity (Jackson, 1974). - 125 -EXPERIMENTAL PROCEDURE Confederates Selection of Leaders To accurately and realistically portray the leadership styles and group conditions of interest, professional actors and actresses were hired as experimental confederates. Recruitment posters were placed in various locations in the University of B.C.'s Theatre Department and 30 actors and actresses auditioned for the confederate positions. The actors and actresses were auditioned in pairs by the author and her supervisor. To assess the natural leadership styles of the actors and actresses, their improvisation s k i l l s , and their competence and versatility in acting, an ambiguous exercise was designed. They were asked to enact the following situation: One of you will assume the role of a manager in any organization of your choosing. The other will be a new employee. The role of the manager is to imagine a task you want the employee to perform. You can invent a task or draw on your own personal experiences. How would you go about getting the employee to do the task? They were then asked to use the same job assignment, but this time to inspire the employee, to put everything on the line to get this person to do the task. On the basis of these auditions, six actresses were initially chosen to portray the various leadership styles. To maximally control for extraneous - 126 -differences across actresses, two actresses were subsequently selected to portray all three leadership styles. The rationale for selecting these two actresses was their adeptness in portraying all three leadership styles credibly and consistently, their extensive professional acting experience, and their identical ages (27 years old) and similar physical characteristics. The choice of actresses rather than actors for the leader roles was based purely on skill and performance. That i s , those chosen to portray the leader roles were selected on the basis of their ability to fulfill the skill and performance requirements. Moreover, according to Bass (1981, p. 499), "the preponderance of available evidence is that no consistently clear pattern of differences can be discerned in the supervisory style of female as compared to male leaders, although individual studies have been able on occasion to find some positive indications, but not necessarily in the same directions." Specifically, once legitimized as a leader, women actually do not behave differently from men (Bass, 1981). Most often, only modest, if any effects of leader gender on subordinates' perceptions of the leader have been reported (e.g., Bartol & Wortman, 1975; Day & Stogdill, 1972; Donnell & Hall, 1980; Eskilson £ Wiley, 1976; Osborn & Vicars, 1976; Rice, Instone, & Adams, 1984). While there is mixed empirical evidence regarding differences between male and female subordinates' attitudes and expectations towards female leaders (e.g.. Petty & Lee, 1975; Petty S Miles, 1976), a majority of studies have reported minimal differences in subordinates' satisfaction with their task or their supervisors as a function of the gender of the supervisor (e.g., Bartol, 1974; Bartol & Wortman, 1975, 1979; Osborn & - 127 -Vicars, 1976; Petty S Bruning, 1980; Trempe, Rigny, & Haccoun, 1985). Furthermore, results, primarily from laboratory studies, suggest that leader gender generally is not a consistent factor in determining group performance (e.g., Bartol, 1978). The aforementioned studies have primarily examined leader gender within the context of the traditionally researched dimensions of considerate and structuring leader behaviour. With regard to charismatic leadership, several scholars have presented case studies of effective female charismatic leaders (e.g.. Day, 1980; Roberts, 1984; T r i c e £ Beyer, 1984). There-fore, as T r i c e and Beyer (1984, p. 57) have concluded, "there is no evidence that being female is an insurmountable barrier to being a successful charismatic leader." Selection of Co-workers, Secretary, and Interviewer To realistically reflect the workplace and the Commerce student population, six actors and six actresses were selected to assume the co-worker roles. These confederates were either graduates from or advanced students in the University of B.C.'s Theatre Department and all had extensive stage experience. The co-worker's ages ranged from 19 to 28 (M = 21.1 years, SD = 2.47). For the secretary role, an individual was needed to greet the participant and request him/her to complete demographic and individual difference measures and to greet and direct the co-workers to the leader's office. An additional role responsibility was to accurately time the experimental procedure. . Due to the extensive time demands of the - 128 -secretary role, two actresses were chosen to enact this role. They were 29 and 38 years old, respectively. For the interviewer role, an individual was required to act as a Research Associate from the University of B.C.'s Faculty of Commerce. The interviewer was responsible for administering the experimental questionnaire and for skillfully and tactfully questioning the participants regarding their reactions to the task, the leader, and the co-workers. A University of B.C. Commerce graduate with considerable theatre training and stage experience was selected to portray the interviewer role. He was 27 years old. T r a i n i n g Leader roles. For the leader roles, the confederates received 30 hours of training including an in-depth description of their roles and a demonstration of the behaviours, emotional states, body language, facial expressions, and paralinguistic cues to be emitted by viewing videotapes of actual managers portraying the different leadership styles (Kotter, 1979, 1980). The actresses also engaged in extensive practice reading of the scripts and in role plays which were videotaped. They received extensive feedback on their taped performances. In addition to scheduled training time, the actresses rehearsed together to maximize their similarities in portraying the leader roles. [Copies of the videotapes are housed in the ; Business School L ibrar ies of the Univers ity of Western Ontario and U.B.C.] Co-worker roles. The confederates received 20 hours of training for their roles as th i r d and fourth year commerce students. T h i s training included studying the commerce programme and its options, reviewing - 129 -course descriptions and outlines, familiarization with the various commerce faculty members through the faculty brochure, attending third and fourth year commerce courses and skimming the textbooks, becoming aware of opportunities in industry, reading the commerce student newspaper. The Cavalier, and, in general, learning about the commerce student culture. The confederates also received a detailed description of their roles, rehearsed their scripts thoroughly, and performed role plays which were videotaped and subsequently reviewed. Particular attention in training the confederates was paid to appropriately handling any inputs to the conversation by the experimental participant. The confederates were instructed to acknowledge any comment by the participant and then to proceed with the scr i p t . The co-workers were also responsible for recording the participants' reactions during the experiment. Hence they were also trained to unobtrusively observe the participant and accurately record their observations on a questionnaire. To conceal the questionnaire from the participant's view, it was embedded in the package of case materials placed in each co-worker's in-basket. Secretary role. For the secretary role, the actresses had 5 hours of training including a description of their role, practice reading of the sc r i p t , and rehearsing responses to possible questions. They also became very familiar with the organization of the office and its filing system. The actresses wore appropriate secretarial attire ( i . e . , a tailored dress or a s k i r t and blouse). - 130 -Interviewer role. For the interviewer role, the actor had 5 hours of training which involved a description of his role, practice reading of the script, and rehearsing answers to anticipated questions. He wore slacks and a sweater. None of the confederates were aware of the purpose and hypotheses of the study. This was verified when they were asked after the study was completed to describe their interpretations of the study, its purpose, and its research questions. Interpretations varied widely and included one held by some confederates that the author might be studying them. The actual purpose of the study was not given as an interpretation by any of the confederates at this debriefing. Validity Checks on the Confederates' Performances After training, there were several checks on the validity of the confederates' enactment of their respective roles. First, the confederates role played their scripts until judged by the author as credible, consistent, and providing performances that clearly distinguished the operational definitions of charismatic, considerate, and structuring leadership styles. ""Second, a convenience sample of 20 judges viewed preliminary videotapes of each actress enacting the three leadership styles. They rated each actress on the leadership style manipulation checks (see Appendix C) and provided written and verbal feedback to the author about their impressions of the actresses' performances including verbal and nonverbal behaviours and paralinguistic cues. The judges' ratings and - 131 -impressions confirmed the operational definitions of the charismatic, s t r u c t u r i n g , and considerate leadership styles. T h i r d , to v e r i f y that the two actresses were accurately portraying the different leadership styles and were highly similar in these portrayals, 203 judges, naive to the study's purpose, rated videotapes of each actress enacting the three styles. The judges consisted of six classes of f i r s t year Masters in Business Administration (MBA) students at the University of B.C. Each class rated one videotape op 31 5-point Likert scales measuring considerate, s t r u c t u r i n g , and charismatic leader behaviours (see Appendix C ) . To test for differences between actresses in their enactment of the various leadership styles. Student's _t tests were computed for each actress on the aggregated leadership style manipulation checks. A summary of these results is presented in Table Q1 (see Appendix Q). Inspection of this table reveals that none of the t tests were significant (p's > .05). These results suggest that the actresses were very similar in their portrayal of each of the three leadership styles. That i s , the behaviour of each actress on the charismatic leadership style was perceived to be the same, as was their perceived behaviour on each of the other two leadership styles. Therefore, in subsequent analyses of the pilot data, the actress variable was aggregated for each leadership style. Fourth, to test for differences between the three leadership styles, a series of Student's jt tests were computed (see Table Q2 in Appendix Q). With the exception of the contrast between the considerate and charismatic styles on the considerate manipulation check, examination of Table Q2 indicates highly significant differences between the s t r u c t u r i n g , - 132 -considerate, and charismatic styles on the aggregated leadership style manipulation checks {pjs < .001). These results suggest that there were clear differences in the students' perceptions of the three leadership styles. S u r p r i s i n g l y , there was a lack of clear differentiation between the considerate and charismatic styles on the considerate manipulation check (£ = .08). Inspection of the pilot data revealed that on the considerate manipulation check the average rating for the considerate leader was 3.43 on a 5-point scale, while the average rating for the charismatic leader was 3.59. Consistent with the operational definition of charisma, the charismatic leader was expected to demonstrate some considerate behaviour given her ability to empathize with and communicate her understanding of followers' needs. However, the considerate leader was expected to be rated substantially higher on this manipulation check. Anecdotally the MBA students reported that the conversation between the considerate leader and the student on the videotape appeared contrived. Therefore the actresses practiced conversing in a more genuine and credible manner and redid their considerate leader roles in front of the video camera. The new videotape was rated by two second year MBA classes (N^ = 38). Subsequent statistical analyses revealed that while there were no significant differences between the two actresses' portrayal of the considerate style, t{33) = 1 .04, £ > .05, there were highly significant differences between the considerate and charismatic styles on the considerate manipulation check, t(33) = 7.61, p < .00005. - 133 -Fifth, to ensure standardization within and between leaders' and co-workers' enactment of their roles and to ensure the leaders and co-workers maintained their required roles throughout the study, the means and standard deviations on the manipulation checks were reviewed on a periodic basis by the author during the experiment. The means and standard deviations on these checks remained consistent throughout the experiment. Sixth, at the beginning of each week for the duration of the experiment, the confederates reviewed the videotapes of their previous performances. As a final check on the confederates' competence in fulfilling their roles, a pilot test of the study was conducted. Eighteen undergraduate commerce students, three in each of the six experimental conditions, completed the experimental task. Visual inspection of the means and standard deviations on the manipulation checks, as well as post-experimental interviews with students, revealed that the students accurately perceived differences between the three leadership styles and between the group productivity conditions. In addition to providing opportunities for the confederates to practice their roles, the pilot test also served to refine the timing of the experimental procedure, to determine the optimal length of the in-basket exercise, and to assess demand characteristics. Specifically, with regard to the timing of the procedure, the entrance and exits of the co-workers were altered to enhance the realism of the project. That is, one co-worker - 134 -reported to the secretary's office 10 minutes prior to the participant's scheduled time, while the other co-worker reported 5 minutes after the participant's a r r i v a l . The co-workers also departed from the workroom at staggered times (see the Procedure section on p.137 for complete details). To determine the optimal length of the in-basket exercise, the number of memos was varied within the 45 minute time period. In order to prevent fatigue and feelings that the task was impossible to accomplish, 15 memos were selected from a complete set of 30 memos for the exercise (see the Experimental Task section on p. 100 for complete details). Finally, with regard to demand characteristics, the students in the pilot study were suspicious of some of the post-experimental questions about the co-workers (e.g.. To what extent did the other students seem to be working on their task?). These questions were subsequently omitted in the study, and the manipulation checks for high and low group productivity were embedded in the role conflict scale and the group atmosphere scale (see the Manipulation Check section on p.112 for a complete explanation). Participants One hundred and forty-four commerce and business administration undergraduates at the University of B.C. participated in the present study. There were 24 participants in each of the six experimental 3 conditions. The proportion of males and females was approximately the 3 Two participants were dropped from the experiment due to their prior exposure to the in-basket exercise as determined through post-experimental questionning. - 135 -same for all conditions. The majority of participants (91.7%) were in their second year of the commerce programme, while the remainder (8.3%) were in t h i r d year commerce. The participants, on average, had one year of part-time work experience and five months of full-time work experience. There were 89 male and 55 female participants and their average age was approximately 20 years (M = 20.56 years, SD = 1.51). The participants were recruited from Commerce 220 (Management of Organizational Behaviour) and received an additional 3% in their course grade for their participation in the study (see Appendix R for a description of the recruitment of experimental participants). These students were considered to be an appropriate sample for the following reasons: (1) the majority had not been exposed to an in-basket exercise, the experimental task and (2) they had not yet received extensive instruction in organizational theory, especially with regard to leadership. The participants signed up for time slots which were randomized for the six experimental conditions. To maintain consistency and accuracy in confederate performances, only two of the six experimental treatments occurred per day. It should be further noted that the actresses played each leader role (charismatic, considerate, or structuring) an approximately equal number of times. In addition, the actors and actresses played each co-worker role (high productivity or low productivity) approximately the same number of times. - 136 -Experimental Setting The study was conducted in the Canada Employment Centre interview offices located in Brock Hall on the University of B.C. campus. The offices were physically separate from the main operation of the Canada Employment Centre and were self-contained. That i s , one central entrance led into four separate offices used for the study. The offices were designed to resemble the temporary business offices of the Mackenzie Institute (see Appendix S for pictures of the offices). One room, without windows, was the secretary's office and had a desk, chairs, filing cabinet, electric typewriter, telephone, coffee machine, i n - and out-baskets with correspondence, and pictures and a 1983 calendar on the wall. Another room, with windows, was the manager's office and had a large, impressive oak desk and chair, desk blotter, executive diary, teak i n - and out-baskets, pen set, policy manuals, and appropriate artwork on the walls. The remaining two offices had windows and served as the workrooms. They contained two tables, three chairs, and three sets of i n - and out-baskets. The seating arrangements were identical in both workrooms. To facilitate interactions between the co-workers and the participant, one co-worker sat at a table by him/herself facing the window; and the participant sat adjacent to one co-worker and facing the other co-worker. To further establish a realistic work environment, signs on the walls and doors of the offices were used to identify the study with the Mackenzie Institute. As well, questionnaires and other materials were fictitiously identified and presented as a part of the normal work routine of the project. - 137 -Experimental Procedure Commerce and Business Administration undergraduates at the University of B.C. were asked to participate in the Management Training Project (see Appendix R). The students were informed by the Course Co-ordinator of Commerce 220 that the University's Faculty of Commerce was co-operating with the Mackenzie Institute, a Vancouver based management consulting firm, in an important project designed to ascertain the practical business skills of Commerce students. Specifically, the purpose of the project was to ensure that the business administration curriculum and teaching methods were responsive to industry's and government's future management needs. The students were required to work with a manager and fellow students i n , a business simulation for 2 hours and would receive, in addition to an educational experience, a bonus of 3% in their course grade for participating. The sign-up book for the experiment was then circulated in the classes and any questions were answered by the Course Co-ordinator. Similar techniques for creating a realistic work situation in the context of a research study have been used previously (e.g., Cilmore, Beehr, S Richter, 1979; Katz, 1977; Lowin et a l . , 1969). As a reminder, the participants were telephoned the evening prior to their participation in the experiment and were asked to report to the temporary campus offices of the Mackenzie Institute located in Brock Hall. The participants were tested individually. - 1 3 8 -The experiment was conducted in nine stages (see Appendix B for the complete experimental s c r i p t s ) . F i r s t , one co-worker reported to the secretary's office 1 0 minutes before the participant's scheduled time and proceeded to complete a general questionnaire not subsequently used in the study. When the participant a r r i v e d , s/he was initially greeted by the secretary and was asked to sign-in and to complete the demographic data form for the project's personnel records (see Appendix T ) . Meanwhile, the secretary escorted the co-worker to the leader's office. Upon her return, the secretary asked the participant to complete the consent form and the individual difference measures (see Appendices N, 0 , and P). While the participant was filling out these questionnaires, the other co-worker reported to the secretary's office and stated s/he was in third year commerce and was here for the second part of the project. The secretary asked him/her to sign-in and then directed him/her to the leader's office. Five minutes later the leader took the co-workers into the appropriate workroom. Thus the confederates were treated like the experimental participant - they filled out forms and questionnaires and were subject to the same work procedure as the participant. When the participant had completed the questionnaire, the secretary informed the leader of the participant's a r r i v a l . To begin the second stage, the leader greeted the participant in the secretary's office and subsequently escorted the participant into her office. The leader introduced the participant to the experimental task by stating that the Mackenzie Institute was attempting to ascertain the practical business skills of commerce students in selected colleges and - 139 -universities across Canada and had designed several exercises for this purpose. The exercise the participant would be working on today was an in-basket. The participant was informed by the leader that s/he would be working with two other commerce students who were in third and fourth year and that they had done the in-basket exercise the previous week and were currently working on other tasks. The leader then escorted the participant into the office where the co-workers were working and introduced them to the participant. A f t e r seating the participant and giving him/her the exercise, the leader checked on the co-workers' progress and then exited. The participant then began the in-basket. In stage three, five minutes after the leader had departed, the group productivity manipulation was initiated. Up until that point, there had not been any direct interaction between the co-workers and the participant. If the participant was assigned to the high productivity condition, then one co-worker started a conversation stating how s/he had already done the in-basket exercise and had found it very interesting. For the low productivity condition, one co-worker remarked how s/he had done the in-basket exercise and had found it boring. The co-workers then completed unobtrusively the f i r s t part of the process measure (see Appendix L ) . In stage four, approximately 20 minutes later, the leader re-entered the workroom to check on the participant's and co-worker's progress on the task and then exited. - 140 -In stage five the group productivity manipulation was continued. Five minutes after the leader had departed (stage four) the highly productive co-workers inquired about the participant's task progress and affirmed the relevance of the exercise. In the low productivity condition, one co-worker asked the participant how much time remained and expressed complete boredom with the task. The co-workers then completed unobtru-sively the second part of the process measure. In stage six, twenty-five minutes later, that is 85 minutes into the study, the leader re-entered the workroom to collect the exercises. She then asked the participant and co-workers to perform an optional task which took 15 minutes. Regardless of their decision to comply or not comply, they were to report to the secretary's office and fill out a questionnaire. The leader subsequently left so the participant could make the decision independent of any possible further leader influence through physical presence in the room. In stage seven, for the highly productive condition, the co-workers remained to do the optional task. One co-worker left after 12 minutes had elapsed while the other co-worker departed after 15 minutes. In the low productivity condition, one co-worker refused to do the optional task, stating that s/he had more important things to do, and left immediately. The other co-worker left 5 minutes later after clearly not attempting to do the optional task. D u r i n g this stage, the participant decided whether or not to do the optional task. - 141 -In stage eight, if the participant completed the task before the time limit expired, then s/he reported to the secretary's office. S/he was subsequently introduced to the interviewer, a Research Associate from the Faculty of Commerce, who escorted the participant back into the workroom and administered the post-experimental questionnaire. If the participant was still working on the task after 15 minutes had elapsed, then the interviewer entered the workroom and administered the questionnaire. The rationale for asking the participant to complete the measures was that since the Faculty of Commerce would be participating in subsequent phases of the project, it wanted to evaluate its success to date (see Appendix B for the interviewer's s c r i p t ) . The participant then completed the questionnaire. The participant was interviewed by the Research Associate to determine his/her reaction to the study and to assess potential demand artifacts (Adair, 1984). Specifically, the participant was asked a series of open ended questions about his/her perceptions of the task, the manager, and the co-workers (see Appendix U). The participant was then asked not to discuss the study with other students and was thanked for participating. Table 4 summarizes the experimental procedure and its duration in minutes. - 142 -Table 4 Experimental Procedure Actual Duration Procedure in Minutes 20 S T A G E 1 - secretary greets the participant and asks him/her to complete the demographic data form and individual difference measures. 15 S T A G E 2 - participant is introduced to the leader, co-workers, and experimental task; leader exits; participant begins the in-basket exercise. 20 S T A G E 3 - group productivity manipulation begins. S T A G E 4 - leader returns and checks the par-ticipant's progress on the task; leader exits. 25 S T A G E 5 - group productivity manipulation continues. S T A G E 6 - leader returns and asks the par-ticipant and co-workers to perform an optional task; leader exits. 1-15 S T A G E 7 - co-workers decide to do/not to do the optional task. Participant decides to do/not to do the optional task. 15 S T A G E 8 - participant completes the post-experimental measures. 10 S T A G E 9 - participant is questioned by the interviewer. - 143 -Administration of the Dependent Measures The dependent measures were administered in the following order. F i r s t , to ensure the individual difference measures were not biased by the experimental manipulations, they were administered in a counterbalanced order during stage one of the experimental procedure. Second, during the experiment, the co-workers completed process measures on the participants' reaction to their situation. T h i r d , at the conclusion of the experiment, measures of task adjustment and performance and measures of interpersonal adjustment were counterbalanced in administration to minimize the likelihood of order effects. Demand Characteristics Several steps were taken to minimize demand characteristics in this study. F i r s t , the experimental setting was elaborately designed and the experimental confederates were well versed in their roles. Second, detailed and plausible cover stories were developed including rationales for the purpose of the study, for the experimental task, and for the post-experimental questionnaire. T h i r d , as recommended by Orne (1969) and Rosenberg (1969), to increase the participants' candidness in answering the post-experimental questionnaire, an interviewer who was not associated with the experimental treatments administered the dependent measures. Fourth, at the end of each experimental session, participants were requested not to discuss their experience with other students in order to ensure that Commerce students' practical business skills would be accurately assessed. Post experimental questioning revealed that while - 144 -participants frequently asked their friends about the project, they had not been told what to expect. There were several indications that the experimental situation had been successfully concealed. During the post-experimental interview, participants were asked what their perceptions were about the purpose of the project, what was expected from them, and how they were supposed to respond. None of the participants indicated suspicion of the project's purpose or of the confederates. In fact, many participants wanted to know if they should return for part two of the Management Training Project as their co-workers had done. They also inquired when the project results would be available. Initially, only the final 18 participants, three from each of the six experimental conditions, were debriefed. During debriefing the participants expressed surprise when they learned that they had been involved in an experiment and that the purpose of the project was different. Collectively, these results suggest that the participants were unaware of the true purpose of the experiment at the time they were studied. A t the conclusion of the study, all participants received a written summary of the purpose, experimental hypotheses, and findings of the study. Specifically, with regard to the in-basket exercise, the participants received a detailed explanation of its purpose, its scoring system, and the results in terms of the qualitative and quantitative performance for different experimental groups in the study. The - 145 -participants were further informed that the study would be extended at other universities since the author will be teaching at another Canadian university next year and plans to do further research on this subject in this setting. The participants were given the author's telephone number if they wished to ask questions or offer comments. - 146 -CHAPTER III RESULTS Experimental analyses were conducted on the following dependent measures: (a) task performance ( i . e . , number of in-basket items attempted, number of courses of action suggested, quality of in-basket performance, and self-rated performance), (b) task adjustment ( i . e . , role conflict, role ambiguity, specific satisfaction, general satisfaction, and job related tension), and (c) interpersonal adjustment ( i . e . , adjustment to the leader and adjustment to the group). The statistical analyses examined the data in reference to the hypotheses outlined in Chapter I. In addition, the individual difference measures ( i . e . , tolerance for ambiguity, need for achievement, and need for affiliation) were statistically analyzed. Prior to presenting the results, an overview of the statistical analyses is presented. Overview of Statistical Analyses Initially the data were examined for accuracy of input, missing values, and outliers and for the appropriateness of assumptions to be used in the following analyses ( i . e . , normality, multicollinearity). Next, to explore and concisely describe the pattern of interrelationships among the leadership style manipulation checks and among the dependent measures, factor analyses were conducted using the principal factor method with 2 multiple R as communality estimates and the varimax method of rotation - 147 -(e.g.. C h i l d , 1970; Green, 1978; Harman, 1967; Rummel, 1970). The reliabilities of the multiple item scales were subsequently evaluated by computing Cronbach's alpha coefficient of reliability (Nunnally, 1978). In addition, intercorrelations among the dependent variables were calculated in order to determine the extent to which the variables were related. To ensure homogeneity of treatment groups and to eliminate competing explanations for the results, several preliminary analyses were subsequently conducted. Student's t^  tests were computed to evaluate the similarity of the two actresses' portrayal of the three leadership styles. Student's t^  tests were also calculated to test for differences between: (a) the fidelity of the performances of the three leadership styles across the two actresses and (b) the high and low group productivity conditions. The continuous data consisted of four measures of task performance, five measures of task adjustment, and two measures of interpersonal adjustment. Scale scores were created by summing the items for each dependent measure. To test the hypotheses outlined in Chapter I, these data were initially submitted to 3( Leadership Style) x 2(Group Productivity) univariate analyses of variance (ANOVAs). Due to the large number of F_ ratios computed in these analyses, an experimentwise alpha level of .05 was used ( K i r k , 1982). Specifically, the experimentwise error rate considers the probability of making one or more Type 1 errors in the set of comparisons under scrutiny (Keppel, 1973). Since the probability of committing T ype 1 errors increases as the number of statistical tests increases, the significance level used to evaluate the comparisons was divided evenly among the 11 comparisons actually being made. Therefore, - 1 4 8 -the significance level was .004 per comparison. In addition, the propor-tion of variance accounted for by each effect was calculated using the omega squared index (Vaughn S Corballis, 1969). All significant effects were subjected to post hoc analyses using the Newman-Keuls sequential range statistic ( K i r k , 1982). To reduce the possibility of spurious findings, an alpha level of .01 was selected. To allow for simultaneous testing of the multiple dependent measures and to account for the various interrelationships among them, a 3(Leader-ship Style) x 2(Croup Productivity) multivariate analysis of variance (MANOVA) was subsequently performed (e.g.. Green 1978; Marascuilo S Levin, 1983; Tabachnick S Fidell, 1983). Thi s test used the Wilks' lambda multivariate test of significance and employed an alpha level of .05. The multivariate eta squared statistic (e.g.. Green, 1978; Tabachnick & Fidell, 1983; Tatsuoka, 1973) was computed to determine the proportion of variance in the linear combination of dependent variables. Post hoc examination of significant MANOVA effects used discriminant function analysis in order to identify which of the multiple dependent variables contribute mostly to the differences between the groups (e.g., Huberty, 1975; Marascuilo & Levin, 1983; Pedhazur, 1982; Tatsuoka, 1970, 1971). In this analysis, both the discriminant function weights and the dependent variables were standardized. Bartlett's approximate x 2 statistic (Tatsuoka, 1971, p. 208) was used to test the significance of the discriminant functions. In addition, the discriminatory power of these functions was computed. Subsequently, group centroids were plotted on a g r i d defined by the discriminant functions. Confidence contours, representing one and two standard deviations, were drawn around each centroid so that group - 1 4 9 -separation and overlap could be examined visually. Interpretation of these plots was further assisted by computation of Roy-Bose confidence intervals (Marascuilo S Levin, 1 9 8 3 ) for pairwise comparisons of the mean stan-dardized discriminant scores. Subsequently, for each significant discriminant function, the pattern of discriminant weights was examined to determine how much each dependent variable contributed to the differentiation between groups. Only those discriminant weights whose absolute values were no less than approximately one-half of the largest discriminant weight were interpreted (Tatsuoka, 1 9 7 0 ) . Three supplemental analyses were performed. F i r s t , a 3 (Leadership Style) x 2 (Croup Productivity) multivariate analysis of covariance (MANCOVA) was computed. The covariates were tolerance for ambiguity, need for achievement, and need for affiliation. Second, Student _t tests were conducted to test for differences between male and female p a r t i -cipants' perceptions of the three leadership styles. T h i r d , the frequency with which the participants complied with the leader's request to perform the optional task of handling five additional memos was computed. Data Screening Utilizing BMDP's Detailed Data Description program, initially the data were examined for accuracy of input. Specifically, for each dependent and individual difference measure, the univariate descriptive statistics were inspected for out-of-range values and plausible means and dispersions. For the general satisfaction dependent measure, two out-of-range values were detected and were replaced with the correct values. As well, for the - 150 -need for achievement and need for affiliation measures, three out-of-range values were found for each measure and were replaced with the correct values. The next step in data screening was to evaluate the extent and pattern of missing data using BMDP's Description and Estimation of Missing Data program. Only a few data points, randomly scattered over cases and variables with no evident patterning on the basis of group variables, were missing from the data set. Missing values were replaced by the estimated group mean. To check for univariate outliers among the dependent and individual difference measures, the standardized scores for each variable were examined. As recommended by Tabachnick and Fidell (1983), a standar-dized score of ± 3.00 was used as a cut for identifying outlying cases. Inspection of these scores revealed that two dependent variables, number of items (minimum standardized score = -3.43; maximum standardized score = 1.32) and number of courses of action (minimum standardized score = -2.48; maximum standardized score = 3.10), exceeded the standardized score criterion. According to Tabachnick and Fidell (1983, p. 74), when the primary purpose of an analysis is to evaluate group differences, it is important to identify outliers for each group separately. If different treatments applied to groups are effective, they may shift scores around enough so that, with groups pooled, those most sensitive to treatment would appear to be outliers. Clearly, one would not want to throw out treatment effects by throwing out those cases. - 151 -Using BMDP's Histogram and Normal Probability Plot program, histograms and normal probability plots of the number of items and number of courses of action variables were examined separately for each treatment group. No univariate outliers were detected. To identify within group multivariate outliers, the Mahalanobis distance between each case and other cases within its group was calculated using BMDP's Estimation of Missing Data program. Outliers were defined as cases with extreme Mahalanobis distance from their group, evaluated as X 2 with degrees of freedom equal to the number of dependent variables (Tabachnick & Fidell, 1983). At an alpha level equal to .05, no outliers were identified. The assumption of normality was assessed by examining the skewness and kurtosis of the dependent measures and covariates. Examination of the standard error for skewness and kurtosis, using a Z value in excess of ± 2.58 as the criterion for rejection of the assumption of normality, revealed no notable skewness for the majority of variables (Tabachnick & Fidell, 1983). Moderate negative skewness for the dependent measure number of items was detected. Kirk (1982, p. 75) argues, however, that "skewed populations have very little effect on either the level of significance or the power of the F test for the fixed-effects model. ... Considering the robustness of the F test to normality when the nJs are equal, the use o f a transformation for this purpose will rarely be advantageous." Accordingly, this variable remained intact. - 152 -To assess multicollinearity, each dependent variable was regressed on all of the other dependent variables (Lewis-Beck, 1980). According to 2 Lewis-Beck (1980), coefficients of determination (R ) near 1.0 reveal 2 nearly redundant variables. For this sample, R ranged from .08 to .61, with the majority reaching .3. Therefore, multicollinearity was not a problem in this data set. Preliminary Analyses Instrumentation for the Independent Variables The leadership style manipulation checks consisted of 31 statements, 10 pertaining to the considerate style, 10 pertaining to the structuring style, and 11 referring to the charismatic style, which were rated by the participants on 5-point Likert type scales (see Appendix C ) . Using BMDP's Factor Analysis program, the leadership style manipulation check items were submitted to the principal factoring method of factor extraction (see Table Q3 in Appendix Q). Based on the criterion of an eigenvalue greater than one (e.g.. C h i l d , 1970; Green, 1978), two factors were extracted and orthogonally rotated using the varimax method. Loadings were considered significant if they exceeded ± .3 (e.g., Chil d , 1970; Green, 1978). Factor one is composed of considerate items with consistently high positive loadings and structuring items with consistently high negative loadings. This suggests a leadership style factor anchored on one end by considerate leadership and on the other end by structuring leadership. This factor is labelled "consideration-structure" and explains 61.03% of the variance. The second factor, labelled "charisma", has - 1 5 3 -consistently high positive loadings on the charismatic items and accounts for 28.95% of the variance. Therefore, the results of this factor analysis suggest that the charismatic, considerate, and structuring leadership styles are clearly distinguished. Using Cronbach's alpha formula, the internal consistency reliabilities were .96 for the considerate and for the structuring scales and .94 for the charismatic scale. According to Nunnally (1978), the criterion for the estimated reliability of research scales is .7. Therefore, based on this criterion, the reliability coefficients for the leadership style manipulation check scales were highly satisfactory. Instrumentation for the Dependent Variables To investigate the integrity of the multiple item continuously scaled dependent measures, factor analyses were performed on the following measures: self-rated performance, role ambiguity, role conflict, job 4 related tension, adjustment to the leader, and group atmosphere. In addition, to evaluate the a priori dimensionality of the three theoretical constructs - task performance, task adjustment, and interpersonal adjustment - the dependent measures comprising each construct were factor analyzed. Using BMDP's program, all factor analyses were performed _ The items comprising two.measures of task adjustment - general task satisfaction and specific task satisfaction (JDI) - were not factor analyzed. The general task satisfaction measure consisted of only two items. The specific task satisfaction measure response format (Y, N, ?) was not appropriate for factor analysis. Therefore, these measures were excluded from individual factor analysis. However, the aggregated scores on each task satisfaction measure, together with the items comprising the remaining task adjustment measures, were submitted to a factor analysis. - 154 -using the principal method of factor extraction. Selection of the number of factors was based on the criterion of an eigenvalue greater than one (e.g.. Chil d , 1970; Green 1978; Tabachnick S Fidell, 1983). If more than one factor was extracted, varimax rotation was employed. Loadings were considered significant if they exceeded ± .3 (e.g. Child, 1970; Tabachnick & Fidell, 1983). The results of these factors analyses are presented below. Task Performance Factor analysis of the self-rated performance scale revealed one important factor explaining 89.96% of the variance (see Table Q4 in Appendix Q). Thi s factor contained all of the self-rated performance items with relatively high positive loadings. The four measures of task performance - number of items attempted, number of courses of action suggested, quality of performance, and self-rated performance items - were subjected to a factor analysis to confirm their independent existence in this sample. As shown in Table Q5 (see Appendix Q), factor one consists of the self-rated performance items with relatively high positive loadings, accounting for 62.79% of the variance. T h i s factor is labelled "subjective task performance". Measures loading on factor two include number of items attempted, number of courses of action suggested, and quality of performance with high positive loadings. T h i s factor explains 30.56% of the variance and is labelled "objective task performance". - 155 -The composition of the second factor is s u r p r i s i n g . As discussed in Chapter II, it was expected that self-rated performance, number of items, number of courses of action, and quality of performance would comprise independent operationalizations of the task performance construct. The results of this factor analysis, however, clearly suggest that there is a strong interrelationship among the objective measures of task performance. Based on these results, it would appear unnecessary to subsequently analyze each measure separately. One possible solution is to select one objective measure to serve as a proxy measure for task performance. However, based on the argument advanced in Chapter II that each objective measure represents a unique indicator of performance on the in-basket exercise (Bentz, 1981), the author decided to retain all three objective measures of task performance for subsequent analyses, recognizing that the results of such analyses need to be interpreted with caution. Another possible solution to account for the interrelationships among the objective task performance measures is to form a linear combination of the dependent measures in subsequent analyses. T h i s solution was adopted through the execution of a MANOVA followed by a discriminant analysis. It should also be noted that factor two, in addition to the objective measures of task performance, contained one self-rated performance item (i.e. How much did you produce in the exercise?). T h i s item had a low positive loading on this factor (.357). To drop this item on the basis of this evidence, particularly in light of the results presented earlier - 156 -concerning the integrity of this scale, did not seem justified (see Table Q4 in Appendix Q). Accordingly, the self-rated performance scale was retained intact. Task Adjustment Factor analysis of the role ambiguity scale items revealed one factor accounting for 96.26% of the variance (see Table Q6 in Appendix Q). This factor was comprised of all the role ambiguity items with high positive loadings. Table Q7 (see Appendix Q) presents the factor analysis of the role conflict items. The factor extracted, explaining 73.25% of the variance, consisted of six items with moderately high positive loadings and two items with very low loadings (.192 and .231). These two items ( i . e . , I had to go against a rule or directive in order to carry out the exercise; I received incompatible requests from my manager and the other students) were excluded from this scale. A second factor analysis of the remaining items yielded one factor comprised of five items with high positive loadings and one item with a low loading (.284) (see Table Q8 in Appendix Q). Th i s item ( i . e . , I worked with a manager and students who operated quite differently) was subsequently omitted from the scale. The next factor analysis of the retained items yielded one factor with relatively high positive factor loadings accounting for 89.97% of the variance (see Table Q9 in Appendix Q). Factor analysis of the job related tension scale items yielded one - 157 -factor accounting for 71.83% of the variance (see Table Q10 in Appendix Q). This factor was composed of all of the tension items with relatively high positive loadings. The five measures of task adjustment - role ambiguity items, role conflict items, job related tension items, and aggregated general task satisfaction and aggregated specific task satisfaction scales - were subjected to a factor analysis to demonstrate their independence for this sample. Three factors were extracted and rotated using the varimax method (see Table Q11 in Appendix Q). Factor one is composed of the role ambiguity items witrr relatively high positive loadings and one job related tension item ( i . e . . Being unclear on just what the scope and responsibilities of your exercise were) with a moderately low negative loading (-.402). T h i s factor explains 51.23% of the variance and is labelled "role ambiguity". The second factor is comprised of the job related tension items with moderately high positive loadings and one role conflict item ( i . e . , I received an assignment without the manpower to complete it) with a relatively weak positive loading (.348). This factor is labelled "job related tension" and accounts for 20.18% of the variance. Factor three consists of the role conflict items with moderately high positive loadings and the aggregated general task satisfaction and the aggregated specific task satisfaction measures with moderately high negative loadings, and two role ambiguity items (i . e . , I knew that I had divided my time properly; Explanation was clear of what had to be done) with low negative loadings (-.318 and -.309, respectively). T h i s bipolar factor, accounting for 9.26% of the variance, was anchored on one end by role conflict and on the other end by task satisfaction. This factor is labelled "task dissatisfaction - task satisfaction". - 158 -The composition of the third factor is s u r p r i s i n g . As discussed in Chapter II, it was expected that general task satisfaction and specific task satisfaction would comprise independent operationalizations of the task adjustment construct. However, the results of this factor analysis suggest that there is a strong interrelationship among the task satisfaction measures. Accordingly, it would be redundant to examine each measure independently. As discussed earlier in this section, a potential solution is to select one task satisfaction measure to serve as a proxy measure. However, based on Scarpello and Campbell's (1983) empirical evidence that global measures of job satisfaction are not equivalent to the summation of many facet responses, both general and specific task satisfaction measures were retained in subsequent analyses. It must be noted that the results of these analyses need to be interpreted with caution. Another possible solution to account for the interrelationships among the task satisfaction measures is to form a linear combination of the dependent measures. Th i s solution was adopted by conducting a MANOVA followed by a discriminant analysis. To summarize, the results of this factor analysis indicate while the measures of general task satisfaction and specific task satisfaction were interrelated, the remaining task adjustment measures - role ambiguity, role conflict, and job related tension - were clearly distinguished. Therefore, the task adjustment construct consists of four components: role ambiguity, role conflict, job related tension, and task satisfaction. - 159 -The results of this factor analysis further indicate that only four items had relatively weak loadings on other factors ( i . e . , one role conflict item, one job related tension item, and two role ambiguity items). However, to drop these items on the basis of this evidence, especially in view of the results presented earlier regarding the integrity of these scales seems unwarranted (see Tables Q6, Q9, and Q10 in Appendix Q). Accordingly these scales were retained intact for this study. Interpersonal Adjustment Factor analysis of the adjustment to the leader scale items revealed one factor explaining 97.99% of the variance (see Table Q12 in Appendix Q). T h i s factor consisted of all of the adjustment to the leader items with high positive loadings. Table Q13 in Appendix Q presents the factor analysis of the group atmosphere scale items. The single factor extracted, explaining 77.08% of the variance, is composed of all of the group atmosphere items with relatively high positive loadings. The two measures of interpersonal adjustment - adjustment to the leader scale items and group atmosphere scale items - were factor analyzed to demonstrate their independent existence in this sample. Two factors were extracted and rotated using the varimax method (see Table Q14 in Appendix Q). Factor one is composed of the group atmosphere items with relatively high positive loadings. T h i s factor, explaining 49.68% of the variance, is labelled "adjustment to the group". The second factor is comprised of the adjustment to the leader items with high positive - 160 -loadings. Accounting for 34.96% of the variance, this factor is labelled "adjustment to the leader". Therefore, the results of this analysis demonstrate that the interpersonal adjustment construct consists of two independent components: adjustment to the group and adjustment to the leader. Scale characteristics, means, standard deviations, and internal consistency reliabilities (Cronbach's alpha) for the multiple item scales are presented in Table 5. Inspection of table shows that the scale reliabilities reached satisfactory levels, exceeding .70 in all cases (Nunnally, 1978). Pearson correlation coefficients among the 11 dependent variables of the study are presented in Table 6. Examination of this table reveals that the majority of correlations were significant (range = .00 to .76; p_'s < .05). The measures of task performance ( i . e . , number of items, number of courses of action, quality of performance) show the highest inter-correlations, whereas the measures of adjustment to the leader and adjustment to the group tend to be weakly correlated with the other dependent measures. The existence of some substantial intercorrelations among the dependent variables demonstrates the need for the use of a multivariate analytic procedure (MANOVA) that takes into account these linear dependencies. It should be underscored, however, that multi-collinearity was not a problem in this data set. As discussed in the Data Screening section (see p. 152), the assessment of the presence of multi-collinearity via the method proposed by Lewis-Beck (1980) indicated that multicollinearity was not problematic. - 160a -Manipulation Checks Actress. To evaluate the similarity of the two actresses' portrayal of the three leadership styles. Student's t. tests were computed for each actress on the aggregated leadership style manipulation checks. These - 161 -Table 5 Scale Characteristics, Means, Standard Deviations and Reliabilities for the Dependent and Individual Difference Measures Scale No. of Items M SD Internal Consistency Reliability Number of Items 1 15. ,28 3 .58 -Number of Actions 1 20. ,88 6 .81 -Quality of Performance 1 2. ,95 1 .03 -Self-Rated Performance 7 22. ,13 4 .35 .83 Role Ambiguity 6 24. ,62 7 .05 .90 Role Conflict 5 18, .99 5 .48 .76 Specific Satisfaction 18 33. ,33 8 .36 .94 General Satisfaction 2 9, .01 3 .05 .72 b Tension 13 36. .59 7 .82 .83 Adjustment to Leader 5 24. .57 6 .45 .94 Group Atmosphere 10 55. ,37 10 .00 .86 Need for Achievement 40 28. .44 4 .39 .98° Need for Affiliation 40 30. ,46 4 .79 .98° Tolerance for Ambiguity 11 6. ,41 2 .35 .94 Note: N = 144. Internal consistency reliabilities were obtained by Cronbach's alpha formula. For two items, alpha is equal to Guttman's split-half coefficient. c F o r items in dichotomous form, alpha is equivalent to the Kuder-Richardson reliability coefficient. Table 6 Pearson Correlation Coefficients for the Dependent Measures Dependent Measure 1 2 3 1 n t e 4 r c o r 5 r e 1 a t i 6 o n s a 7 8 9 10 11 1. Number of Items 1.000 2. Number of Actions .691 1.000 3. Quality of Performance .745 .760 1.000 4. Self-Rated Performance .248 .219 .315 1.000 5. Role Ambiguity .123 .142 .323 .379 1.000 6. Role Conflict -.216 -.278 -.396 -.403 -.529 1.000 7. Specific Satisfaction .208 .323 .375 .291 .407 -.565 1.000 8. Ceneral Satisfaction .162 .265 .349 .365 .480 -.611 .667 1.000 9. Tension -.051 -.020 -.063 -.557 -.305 .402 -.386 -.344 1.000 10. Adjustment to Leader -.009 .226 .091 -.100 -.179 -.179 .344 .234 .000 1.000 11. Adjustment to Group .029 -.024 .011 .048 .178 -.134 .172 .208 .019 -.145 1.000 Note: N = 144. Correlations > 0.214 and > 0.164 are significant at the .01 and .05 levels, respectively. - 163 -results are presented in Table Q15 (see Appendix Q). Examination of this table reveals that none of the t tests were significant (p's > .05). These results suggest that the actresses were very similar in their portrayal of the three leadership styles. Therefore, for the analyses reported below, the actress variable was aggregated for each leadership style. Leadership style. The means and standard deviations for the leader-ship style manipulation checks are presented in Table Q16 (see Appendix Q). To test for significant differences between the three leadership styles, a series of Student's _t tests were computed (see Table Q17 in Appendix Q). Inspection of Table Q17 indicates highly significant differences between the s t r u c t u r i n g , considerate, and charismatic styles on the aggregated leadership style manipulation checks (p's < .0001). These results suggest that there were clear differences in the participants' perceptions of the three leadership styles. Croup productivity. The strength of the group productivity mani-pulation was assessed by examining three items on the post- experimental questionnaire. Two bipolar items on the group atmosphere scale (enthusiastic-unenthusiastic; productive-nonproductive) and one item on the role conflict scale (I received incompatible requests from my manager and the other students) served as manipulation checks. As shown in Table Q18 (see Appendix Q), there were highly significant differences in participants' perceptions of the high and low group productivity norms (p's < .005). This suggests that the group productivity manipulation was successfully achieved. - 164 -Univariate ANOVAs The group means and standard deviations and the cell means and standard deviations for the dependent measures are presented in Tables 7 and 8, respectively. Table 9 summarizes the results from 11 3(Leadership Style) x 2(Croup Productivity) ANOVAs. Based on the Cochran and the Bartlett-Box tests, the hypothesis of the homogeneity of variance was accepted for these analyses (p_'s > .01). Inspection of Table 9 indicates highly significant effects for leadership style, group productivity, and the leadership style by group productivity interaction on role conflict, specific satisfaction, and general satisfaction (p_'s < .0001). Leadership style also had significant effects on the number of courses of action, the quality of task performance, role ambiguity, and adjustment to the leader, while group productivity had a significant effect on participants' adjustment to the group (p_'s < .0001). Omega squares, also reported in Table 9, indicate that the main effect for leadership style accounted for the major portion of the variance. The leadership style, group productivity, and leadership style by group productivity interaction effects were not significant for three dependent variables: 1. The number of in-basket items attempted. 2. The participants' self-rated perceptions of their task performance. 3. The participants' job related tension. Table 7 Croup Means and Standard Deviations for the Dependent Measures Leadership Style Croup Productivity Structuring Considerate Charismatic Low High. Dependent Measures M SD M SD M SD M SD M SD Task Performance Number of Items 15. 35 3.65 14.77 3.98 15 .71 2.83 14.69 3. 79 15.86 3. 19 Courses of Action 19. 06 5.72 19.49 6.06 24 .06 7.26 19.72 6. 22 22.03 6. 47 Quality of Performance 2. 94 .95 2.46 .91 3 .46 .96 2.85 92 3.06 . 96 Self-Rated Performance 23. 02 4.64 20.88 4.14 22 .49 3.89 21.79 4. 52 22.47 3. 93 sk Adjustment 1 Role Ambiguity 28. 77 4.77 16.43 3.58 28 .65 3.01 23.86 3. 54 25.37 4. 03 Role Conflict 19. 42 3.60 22.06 4.43 15 .48 3.21 20.39 3. 42 17.58 4. 07 Specific Satisfaction 30. 83 6.07 29.73 6.34 39 .44 5.58 30.24 5. 81 36.43 6. 18 General Satisfaction 8. 60 1.88 7.38 2.81 11 .06 2.13 8.21 2. 46 9.82 2. 09 Tension 36. 10 7.98 37.89 8.05 35 .77 6.87 38.08 7. 41 35.09 7. 85 :erpersonal Adjustment Adjustment to Leader 16. 49 2.88 26.77 2.55 30 .44 2.34 24.64 2; 49 24.49 2. 69 Adjustment to Group 58. 18 9.58 53.21 7.75 54 .73 7.15 49.99 8. 55 60.75 7. 76 Table 8 Cell Means and Standard Deviations for the Dependent Measures Dependent Measures Structuring Low Productivity High Productivity Considerate Low Productivity High Productivity Charismatic Low Productivity High Productivity M SD M SD M SD M SD M SD M SD Task Performance Number of Items 14.71 4.36 16.00 2.93 14.54 3.66 15.00 4.30 14.83 3.34 16.58 2.32 Courses of Action 18.71 6.15 19.42 5.28 18.17 5.32 20.83 6.79 22.29 7.19 25.83 7.33 Quality of Performance 2.67 1.01 3.21 .88 2.33 .76 2.58 1.06 3.54 1.10 3.38 .82 Self-Rated Performance 21.63 5.19 24.4? 4.09 20.813 4.18 20.92 4.09 22.92 4.20 22.08 3.59 Task Adjustment Role Ambiguity 26.63 4.16 30.92 5.39 16.33 3.62 16.54 3.54 28.63 2.86 28.67 3.16 Role Conflict 24.50 3.31 14.33 3.89 21.50 4.02 22.63 4.84 15.17 2.94 15.79 3.48 Specific Satisfaction 24.42 6.01 37.25 6.12 27.21 5.80 32.25 6.88 39.08 5.63 39.79 5.53 Ceneral Satisfaction 6.33 1.86 10.88 1.89 7.38 3.16 7.38 2.46 10.92 2.36 11.21 1.91 Tension 39.21 7.93 33.00 8.03 38.83 8.14 36.96 7.96 36.21 6.16 35.33 7.57 Interpersonal Adjustment Adjustment to Leader 16.67 2.63 16.33 3.13 26.54 2.69 27.00 2.41 30.71 2.18 30.17 2.51 Adjustment to Croup 53.46 10.29 62.92 8.87 48.67 8.23 57.75 7.27 47.88 7.14 61.58 7.15 Table 9 Analyses of Variance Summary Table for the Dependent Measures Dependent Measure Source of Variation df MS F <!J2(%) Task Performance Number of Items Leadership Style (A) 2 10.75 0.84 -0.21 Croup Productivity (B) 1 49.00 3.86 2.00 A x B ' 2 5.14 0.14 -0.85 Residual 138 12.68 99.06 Courses of Action Leadership Style (A) 2 368.06 8.98* 10.17 Croup Productivity (B) 1 191.36 4.67 2.55 A x B 2 25.25 0.61 -0.55 Residual 138 40.96 87.83 Quality of Performance Leadership Style (A) 2 12.01 13.35* 14.92 c Croup Productivity (B) 1 1 .56 1 .73 0.52 A x B 2 1.52 1.69 0.97 Residual 138 0.89 83.59 Self-Rated Performance Leadership Style (A) 2 60.13 3.32 3.19 Croup Productivity (B) 1 16.67 0.92 -0.05 A x B 2 42.63 2.35 1 .90 Residual 138 18.07 94.96 Table 9 (continued) Analyses of Variance Summary Table for the Dependent Measures Dependent Measure Source of Variation df MS F oi 2(%) Task Adjustment Role Ambiguity Leadership Style (A) 2 2409.36 160.33* 69.33 Group Productivity (B) 1 82.51 5.49 3.11 A x B 2 69.52 4.62 4.89 Residual 138 15.02 22.67 Role Conflict Leadership Style (A) 2 526.76 36.56* 33.53 Group Productivity (B) 1 283.36 19.67* 11 .77 A x B 2 488.42 33.90* 31.82 Residual 138 14.41 22.88 Specific Satisfaction Leadership Style (A) 2 1356.02 37.51* 34.12 Group Productivity (B) 1 1381.36 38.21* 20.99 A x B 2 453.01 12.53* 14.05 Residual 138 36.14 30.84 General Satisfaction Leadership Style (A) 2 169.22 31.43* 30.15 Group Productivity (B) 1 93.44 5.38* 10.46 A x B 2 77.55 14.40* 15.97 Residual 138 5.38 43.42 Table 9 (continued) Analyses of Variance Summary Table for the Dependent Measures Dependent Measure Source of Variation df MS F co2(%) Tension Leadership Style (A) 2 62.69 1.06 0.09 Croup Productivity (B) 1 321.01 5.46 3.09 A x B 2 96.44 1 .64 0.90 Residual 138 58.72 95.92 Interpersonal Adjustment Adjustment to Leader Leadership Style (A) 2 ^ 2505.51 368.25* 83.89 Croup Productivity (B) 1 .69 0.10 -0.64 A x B 2 3.34 0.49 -0.73 Residual 138 6.80 17.48 Adjustment to Croup Leadership Style (A) 2 312.52 4.61 4.86 Croup Productivity (B) 1 4160.25 61.31* 30.11 A x B 2 79.18 1.16 0.24 Residual 138 67.85 64.79 *£ < .0001. - 170 -Inspection of these nonsignificant differences indicated that the participants attempted approximately 15 out of 20 memos and perceived their task performance as adequate (M = 22.13; 3.16 on a 5-point scale). The participants also reported occasional work related tension (M = 36.59; 2.81 on a 5-point scale). Post Hoc Analyses Both the significant leadership style main effect and leadership style by group productivity interaction were subjected to post hoc analyses using the Newman-Keuls test (see Tables Q19 and Q20 in Appendix Q for summaries of the post hoc analyses). Post hoc analyses of the significant group productivity main effect used the group means (see Table 7). Since a significant interaction implies that the effect of one variable changes at different levels of the second variable, the interpretation of significant main effects needs to be tempered according to the nature of this inter-action (e.g., Keppel, 1973; Kirk, 1982). Therefore, with an interaction, the meaning of the main effects must be interpreted with caution. Leadership Style Post hoc analyses indicated that individuals exposed to a charismatic leader had significantly higher quality of task performance when compared to a considerate leader (p < .01). There was an absence of statistically significant differences between charismatic and structuring leaders and between structuring and considerate leaders for individuals' quality of task performance ( £ > .01). Analyses further indicated that individuals with a charismatic leader generated significantly more courses of action than - 171 -individuals with structuring (£ < .01) and considerate leaders (p < .01). There were no significant differences between structuring and considerate leaders (£ > .01). T u r n i n g to task adjustment, post hoc analyses revealed that indi -viduals with charismatic leaders had higher specific and general task satisfaction and lower role conflict than those with structuring (p_ < .01) or considerate leaders (£ < .01). There were no significant differences between structuring and considerate leaders for specific and general task satisfaction (£'s > .01). However, individuals with considerate leaders had higher role conflict than individuals with structuring leaders (£ < .01). For role ambiguity, individuals working under considerate leaders experienced significantly greater role ambiguity than individuals working under structuring (£ < .01) and charismatic leaders (£ < .01). There were no significant differences between structuring and charismatic leaders {£ > .01). Finally, individuals with charismatic leaders had higher adjustment to the leader than individuals with structuring (£ < .01) and considerate leaders (£ < .01). Individuals with considerate leaders had significantly higher adjustment to the leader than those with structuring leaders (£ < .01). The leadership' style effect was not statistically significant for the following dependent measures: number of items, self-rated performance, job related tension, and adjustment to the group. - 172 -Croup Productivity As shown in Table 7, individuals in the low productivity group, as compared to the high productivity group, had lower satisfaction with specific aspects of the task (M = 30.24; M = 36.43), lower general satis-faction with the task (M = 8.21; M = 9.82), lower adjustment to the group (M = 49.99; M = 60.75), and higher role conflict (M = 20.39; M = 17.58). The group productivity effect was not statistically significant for the following dependent measures: task performance ( i . e . , number of items, number of courses of action, quality of performance, and self-rated performance), role ambiguity, job related tension, and adjustment to the leader. Leadership Style x Croup Productivity Graphs of the significant interaction effects are presented in Figures 1, 2, and 3. Examination of Figures 1 and 2 show that individuals with a structuring leader and in a low productivity group experienced substan-tially higher role conflict (M = 24.50) and lower general satisfaction with the task (M = 6.33) than individuals with a structuring leader and in a high productivity group (M = 14.33; M = 10.88). Results of Newman-Keuls post hoc tests indicated that these differences were significant (p's < .01). Individuals with a considerate leader and in both high and low group productivity conditions experienced relatively high levels of role conflict (M = 22.63; M = 21.50) and low levels of general satisfaction (M = 7.38; M = 7.38), however there were no significant differences between the two groups (p's > .01). In contrast, with a charismatic leader under both high and low group productivity, individuals had relatively low role conflict (M = 15.79; M = 15.17) and high general satisfaction (M = 11.21; M - 173 -Figure 1 Interaction Effect for Leadership Style and Croup Productivity on Role Conflict 30 Legend: • High Productivity • Low Productivity 25 J 20 15 J i o t r Structure Considerate Leadership Style Charisma - 174 -Figure 2 Interaction Effect for Leadership Style and Croup Productivity on General Satisfaction Legend: • High Productivity 12 • Low Productivity o rO A3 S_ <u c Q J C J 3 C o e ro CD i i 4 io J 8 J 7 J 6 J Structure Considerate Leadership Style Charisma - 175 -Figure 3 Interaction Effect for Leadership Style and Croup Productivity on Specific Satisfaction Structure Considerate Charisma Leadership Style - 176 -= 10.92). Again, the post hoc tests revealed these differences were not significant (p_'s > .01). Consistent with the results for general satisfaction, Figure 3 reveals that individuals with a structuring leader and in a low productivity group have considerably lower satisfaction with specific aspects of the task (M = 24.42) than those individuals with a structuring leader and in a high productivity group (M = 37.25). The post hoc test indicated that this difference was significant ( £ < .01). Individuals with a considerate leader and in a low productivity group have lower specific task satisfaction (M = 27.21) than in a high productivity group (M = 32.25) and this difference is significant ( £ < .01). In contrast, individuals with a charismatic leader and in both high and low group productivity conditions experienced very high satisfaction with the exercise (M = 39.79; M = 39.08) and this difference is not significant between group conditions (£ > .01). There was a lack of statistically significant interaction effects for measures of task performance ( i . e . , number of items, number of courses of action, quality of performance, and self-rated performance), interpersonal adjustment ( i . e . , adjustment to the leader and to the group), and two measures of task adjustment ( i . e . , role ambiguity and job related tension). MANOVA To take into account the intercorrelations among the dependent variables and to compensate for increased Type 1 errors associated with - 177 -multiple univariate testing of the same data, a 3 x 2 factorial MANOVA was performed on the task performance, task adjustment, and interpersonal adjustment dependent measures. Independent variables were leadership style ( s t r u c t u r i n g , considerate, and charismatic) and group productivity (high and low). Initially, the assumption of homogeneity of variance-covariance matrices was assessed by Box's M test, a multivariate analog of Bartlett's test (Green, 1978; Tabachnick & Fidell, 1983). The results indicated that this assumption was satisfactorily met, approximate F(330, 28475) = 1.181, p > .01. Based on Wilks' lambda criterion, the dependent variables were s i g n i -ficantly affected by leadership style, approximate F (22, 256) = 50.13, p < .0005, group productivity, approximate F(11 , 128) = 10.90, p < .0005, and their interaction, approximate F_(22, 256) = 5.34, p < .0005 (see Table 10). Inspection of Table 10 further reveals a very strong association 2 between leadership style and the dependent variables, n = .96. The 2 association was less substantial for group productivity (ri = .48) and for 2 the interaction (ri = .53) and the dependent variables. According to several statisticians (e.g.. Hair et a l . , 1979; Huberty, 1975; Marascuilo £ Levin, 1983; Pedhazur, 1982; Tatsuoka, 1970, 1971), the appropriate technique for investigating significant MANOVA effects post hoc is discriminant analysis. By applying discriminant analysis, the dependent variables which maximally differentiate between the groups can be identified. Since the interaction effects were relatively weak, the main effects were also subjected to post hoc testing. However, the interpre-tation of tests of main effects must be qualified by the interaction. Table 10 Multivariate Analysis of Variance Summary Table for the Dependent Measures Source of Variation Wilks' A Approximate F Hypothesis df 'Error df Significance Level n 2 ( % > Leadership Style (A) .04 50.13 22 256 .0005 96.00 Croup Productivity (B) .52 10.90 11 128 .0005 48.00 A x B .47 5.34 22 256 .0005 53.00 Note: N = 144. - 179 -Discriminant Analysis Leadership Style For the leadership effect, two discriminant functions were calculated, with a combined x 2(22) = 454.07, £ < .001 (see Table 11). This result indicates that the firs t discriminant function is highly significant. After removal of the f i r s t function, the remaining discriminant function was also significant, x 2 (10) = 180.06, £ < .001, and thus represented an additional dimension for separating the leadership styles. The two discriminant functions accounted for 96% of the between group variability in discriminating among styles. Thus the discriminant functions have substantial discriminating power for differentiating the leadership styles. Figure 4 shows the centroids for the three leadership styles plotted on the two discriminant functions. Confidence contours representing one and two standard deviations are drawn around each centroid to highlight the separations between the leadership styles. Inspection of Figure 4 indicates that the firs t discriminant function differentiates the structuring leadership style from the considerate and charismatic styles. The considerate and charismatic styles are not differentiated on this function. The second discriminant function clearly distinguishes all three leadership styles from each other. Interpretation of Figure 4 is further assisted by the computation of Roy-Bose confidence intervals for pairwise comparisons of the centroids (see Table-12). The confidence intervals are consistent with the graphical interpretation. For discriminant function one, structuring leaders are - 180 -Table 11 Standardized Discriminant Function Coefficients for Leadership Style Dependent Variable Task Performance Number of Items Courses of Action Quality of Performance Self-Rated Performance Task Adjustment Role Ambiguity Role Conflict Specific Satisfaction General Satisfaction Tension Interpersonal Adjustment Adjustment to Leader Adjustment to Group Discriminant 1 Function 2 ,008 ,129 .094 .098 .345 .063 .513 .275 .493 .046 .017 .011 .004 .760 .261 .226 .165 .243 .944 .157 .203 .043 a T h e role ambiguity scale is reverse scored: a low score is associated with role ambiguity while a high score is associated with role cl a r i t y . - 181 -Figure 4 Typology of the Centroids in Discriminant Space for Leadership Style -3 : Dark contours represent one standard deviation; light contours represent two standard deviations. 1 = structuring style- 2 = considerate style; 3 = charismatic style. - 182 -Table 12 Roy-Bose Confidence Intervals for Pairwise Comparisions of the Mean Standardized Discriminant Scores 99% Confidence Interval Leadership S t y l e 3 Discriminant Function 1 L 1 vs. L 2 (4.20, 6.35)* L 1 vs. L 3 (4.73, 6.88)* L 2 vs. L 3 (-0.55, 1.61) Discriminant Function 2 L 1 vs. L 2 (0.88, 3.04)* L 1 vs. L 3 (1.07, 3.22)* L 2 vs. L 3 (3.03, 5.18)* Group Productivity Discriminant Function 1 G 1 vs. C 2 (1 .00, 2.79)* Leadership Style x Group Productivity Discriminant Function 1 L 1 G 1 vs. L 1 C 2 (1.04, 4.68)* L 2 G 1 vs. L 2 C 2 (-1.37, 2.27) L 3 G 1 vs. L 3 G 2 (-1 .00, 2.64) * £ < .01 . L 1 = structuring style; L, = considerate style; L 3 = charismatic style. G. = high group productivity; G ? = low group productivity. - 183 -significantly differentiated from considerate (£ < .01) and charismatic leaders (£ < .01), while the latter leaders are not significantly distinguished (£ > .01). All three leadership styles are significantly separated from each other on discriminant function two (£'s > .01). Inspection of the standardized discriminant function coefficients in Table 11 indicates that the dependent variables adjustment to the leader (-.944) and role ambiguity (.493) maximally separate the leadership styles on the fi r s t discriminant function. The second discriminant function is chiefly defined by role ambiguity (-.760) and quality of performance (-.513) variables. Group Productivity For the group productivity effect, one discriminant function was computed, with x 2 ( H ) = 90.21, £ < .001 (see Table 13). This function accounted for 48% of the between group variability in discriminating among high and low group productivity conditions. As shown in Figure 5, discriminant function one differentiates high group productivity from low group productivity. Examination of Table 12 reveals that this differentiation is significant (£ < .01). Inspection of the standardized discriminant function weights in Table 13 indicate that adjustment to the group (.765) and specific satisfaction with the task (.446) maximally discriminate between high and low group productivity conditions. - 184 -Table 13 Standardized Discriminant Function Coefficients for Croup Productivity Dependent Variable Discriminant Function 1 a Task Performance Number of Items .171 Courses of Action .302 Quality of Performance -.256 Self-Rated Performance -.320 Task Adjustment Role Ambiguity .074 Role Conflict -.290 Specific Satisfaction .446 General Satisfaction .053 Tension -.224 Interpersonal Adjustment Adjustment to Leader -.220 Adjustment to Group .765 a T o aid in the interpretation of discriminant function 1, the weights were multiplied by -1. - 185 -Figure 5 Typology of the Centroids in Discriminant Space for Group Productivity -1 Note: Dark contours represent one standard deviation; light contours represent two standard deviations. 1 = high group productivity; 2 = low group productivity. - 186 -Leadership Style x Croup Productivity For the leadership style and group productivity interaction, two discriminant functions were calculated with a combined x 2 (55) = 101 .58, £ < .001, indicating that the fir s t discriminant function is highly significant. With the f i r s t discriminant function removed, the second discriminant function did not reach statistical significance, x 2 (40) = 16.10, £ > .05. T h i s finding suggests that the second discriminant function does not significantly separate the groups. Therefore, only the f i r s t discriminant function was retained and interpreted (see Table 14). T h i s discriminant function accounted for 45% of the between group variability. Figure 6 indicates that discriminant function one clearly differentiates between the structuring style and high group productivity and the structuring style and low group productivity interaction at one standard deviation. In contrast, the considerate - high productivity and consider-ate-low productivity groups appear to overlap considerably. The charismatic - high productivity and charismatic - low productivity groups also show extensive overlap. Examination of Roy-Bose confidence intervals corroborates these findings (see Table 12). The structuring-high productivity and s t r u c t u r -ing - low productivity groups were significantly separated ( £ < .01). In contrast, the considerate leader combined with low and high group productivity as well as the charismatic leader with the respective group conditions were not significantly differentiated (£ rs > .01). - 187 -Table 14 Standardized Discriminant Function Coefficients for Leadership Style by Group Productivity Dependent Variable Discriminant Function 1 Task Performance Number of Items Courses of Action Quality of Performance Self-Rated Performance Task Adjustment Role Ambiguity Role Conflict Specific Satisfaction General Satisfaction Tension -.154 -.701 .537 -.063 .058 -.786 .369 .305 .339 Interpersonal Adjustment Adjustment to Leader Adjustment to Group -.169 -.222 - 188 -Figure 6 Typology of the Centroids in Discriminant Space for Leadership Style by Croup Productivity Note: Dark contours represent one standard deviation; light contours represent two standard deviations. 1 = structuring style, high productivity; 2 = structuring style, low productivity; 3 = con-siderate style, high productivity; M = considerate style, low productivity; 5 = charismatic style, high productivity, 6 = charismatic style, low productivity. - 189 -Inspection of the discriminant coefficients for the leadership style and group productivity interaction in Table 14 suggests that the primary dependent variables distinguishing between groups are role conflict (-.786), number of courses of action (-.701), and quality of in-basket performance (.537). To conclude, it is interesting to note the similarities and differences between the conclusions just drawn on the basis of multivariate analysis and those based on the univariate ANOVAs. In terms of similarities, role conflict and ambiguity, specific satisfaction, adjustment to the leader and to the group, number of courses of action, and quality of task perform-ance were important dependent measures in both univariate and multi-variate analyses. In terms of differences, general satisfaction, which was among those dependent variables showing the largest differences univariately, was found to contribute negligibly to group differentiation in terms of the linear combination of dependent variables. The similarities and differences between the univariate and multivariate analyses are further explored in the following chapter. Supplemental Analyses Individual Differences A 3(Leadership Style) x 2(Croup Productivity) MANCOVA was performed on 11 dependent variables. In this analysis, the linear combination of the dependent variables was statistically adjusted for the effects of three covariates: tolerance for ambiguity, need for achievement, and need for affiliation. The covariates were judged to be 190 -adequately reliable for covariance analysis (see Table 5). The assumption of homogeneity of regression slopes was satisfactorily met for this analysis {p_ > .01). In addition, the assumption of homogeneity of variance-covariance matrices was assessed by Box's M test (Green, 1978; Tabachnick & Fidell, 1983). Results showed that this assumption was satisfactorily met, approximate F(525, 28154) = 1 .057, £ > .01. Examination of Table Q21 (see Appendix Q) reveals that the dependent variables were significantly related to leadership style, approximate F(22, 250) = 48.76, £ < .0005, to group productivity, approximate F(11 , 125) = 9.80, £ < .0005, and to the leadership style by group productivity interaction, approximate F(22, 250) = 5.30, £ < .0005, according to Wilks' multivariate test of significance. The multivariate test for the covariates failed to reach statistical significance, approximate £(33, 368) = 1.19, £ > .20. Participant Gender To determine if there were differences between male and female participants' perceptions of the three leadership styles. Student's jt tests were computed for participants' ratings of the leaders on the leadership style manipulation checks (see Table Q22 in Appendix Q). Inspection of Table Q22 reveals no significant gender effects for the structuring, considerate, and charismatic leadership styles (£'s > .01). This finding suggests that male and female participants had consistent perceptions of the three leadership styles. - 191 -Performance of the Optional Task The frequency with which participants complied with the leader's request to perform the optional task is shown in Table Q23 in Appendix Q. This table indicates that the vast majority of participants (93.8%) agreed to perform the optional task, while the remainder of participants did not agree (6.2%). There was minimal variation between the three leadership styles for participants' compliance to do the optional task. Similarly, Table Q24 in Appendix Q reveals that in the high group productivity condition, 71 participants agreed to do the optional task, while in the low group productivity condition 64 participants agreed to do the optional task. In total, only 9 participants refused to do the optional task. - 192 -CHAPTER IV DISCUSSION OF RESULTS This chapter is organized into three sections. The f i r s t section discusses the results relevant to the hypotheses outlined in Chapter I based on the univariate analyses. The pattern of similarities and differences between the univariate and multivariate analyses is highlighted in the second section. Interpretation of the results of secondary analyses is presented in the third section. T E S T S OF HYPOTHESES BASED ON UNIVARIATE ANALYSES An Overview The findings in this dissertation were numerous and complex. In order to present and discuss the results as clearly and coherently as possible, a brief overview of the major univariate findings is presented below. The univariate analyses of variance indicated that individuals who worked under charismatic leaders had significantly higher task performance in terms of the number of courses of action suggested on the in-basket exercise, greater task satisfaction, lower role conflict, and higher adjustment to the leader when compared to individuals working under - 193 -structuring and considerate leaders. In addition, the qualitative task performance of individuals with charismatic leaders surpassed that of individuals with considerate leaders. A sur p r i s i n g finding was while individuals with structuring leaders experienced role clarity and reduced role conflict, their task performance and task satisfaction were equivalent to those with considerate leaders. The results further indicated that individuals with considerate and charismatic leaders had higher adjustment to the leader than individuals with structuring leaders. Contrary to expectation, individuals with charismatic leaders had higher adjustment to the leader than individuals with considerate leaders. The group productivity data indicated that individuals in high productivity groups reported significantly higher task satisfaction, lower role conflict, and higher adjustment to the group than did individuals in low productivity groups. An unexpected finding was the lack of s i g n i f i -cant effects for group productivity norms on individual task performance. Interactions between leadership style and group productivity revealed that charismatic leaders, irrespective of high or low group productivity norms, produced high individual task performance, task adjustment, and adjustment to the leader and to the group. In contrast to the charismatic leader, the structuring leader's impact on individuals' task adjustment was modified by group productivity norms. Individuals who worked with a structuring leader and in a high productivity group reported higher specific task satisfaction, higher general task satisfaction, and lower role - 194 -conflict than did individuals who worked with a structuring leader and in a low productivity group. Despite the significant interaction effect of structuring leadership and group productivity norms on individual task adjustment, there was an absence of statistically significant effects on task performance, role ambiguity, and adjustment to the leader and to the group. Finally, individuals with a considerate leader and in a high productivity group had significantly higher specific task satisfaction than those with a considerate leader and in a low productivity group. The results further indicated that individuals exposed to a considerate leader, irrespective of group productivity norms, had marginal task performance and low task adjustment. However, individuals reported positive adjustment to the leader and to the group. The remainder of this section presents and discusses the univariate results in detail, based on the hypotheses posed in Chapter I. LEADERSHIP S T Y L E Task Performance and Task Adjustment Hypothesis 1. Individuals working under a charismatic leader will have higher task performance than will individuals working under a considerate leader. Hypothesis 2. Individuals working under a charismatic leader will report higher task adjustment than will individuals working under a considerate leader. - 195 -These hypotheses were partially supported by the findings of this study. With regard to task performance, individuals with charismatic leaders as compared to individuals with considerate leaders generated significantly more courses of action on the in-basket exercise (M = 24; M = 19; £ < .01) and had significantly higher quality of task performance (M = 3.46; M = 2.46; £ < .01) (see Table 7). Contrary to prediction, there was an absence of a leadership effect on performance in terms of the number of in-basket items completed (M = 15; £ > .01) and in terms of participants' perceptions of their task performance (M = 21.69; £ > .01) (see Table 7). T u r n i n g to the task adjustment data, while there were no significant 5 effects for job related tension, individuals with charismatic leaders reported significantly higher general task satisfaction (M = 11.06; M = 7.38; £ < .01), higher specific task satisfaction (M = 39.44; M = 29.73; £ < .01), higher role clarity (M = 28.65; M = 16.43; £ < .01), and lower role conflict (M = 15.49; M = 22.06; £ < .01) when compared to those with considerate leaders (see Table 7). The effectiveness of charismatic leadership in facilitating individuals' task performance and task adjustment is consistent with theoretical discussions of charisma. For example. Max Weber (1947), in his sociological treatment of charisma, recognized that charismatic leaders obtained their effect's by v i v i d l y articulating a transcendent goal which 5 -There was a lack of significant main effects and interactions for three dependent measures: number of in-basket items completed, self-rated performance, and job related tension. Explanations for the absence of differences on these measures will be discussed later in this chapter (see pp.227-228). - 196 -clarified or specified a mission and which communicated values that had ideological significance for followers. Moreover, in his theory of charismatic leadership in organizations. House (1977) asserted that by simultaneously communicating high performance expectations for followers and by exhibiting confidence in their ability to meet such expectations, the charismatic leader enhances followers satisfaction, motivation, and performance in support of the articulated goal. Therefore, the charismatic leader is asserted "to clarify followers' goals, cause them to set or accept higher goals, and have greater confidence in their ability to contribute to the attainment of such goals" (House, 1977, p. 191). A consistent view of the inspirational effects of charismatic appeals on followers' motivation is offered by Avolio and Bass (1985). They argue that by espousing a vision which followers' view as being worthy of their effort, the charismatic leader excites followers, and, in turn, elevates their anticipated levels of effort. Similarly, Bennis (1982, p.55) posits that a fundamental component of charismatic power is the "capacity to create and communicate a compelling vision of a desired state of affairs" which energizes, involves, and commits followers to pursue group objectives. McClelland (1975, p.260) has accurately captured these charismatic effects on followers: Whatever the source of the leader's ideas, he cannot inspire his people unless he expresses v i v i d goals which in some sense they want. Of course, the more closely he meets their needs, the less 'persuasive' he has to be; but in no case does it make sense to speak as if his role is to force submission. Rather it is to strengthen and uplift, to make people feel they are the origins, not the pawns, of the socio-political system (deCharms, 1968). His message is not so much: 'Do as I say because I am strong and know best. You are children with no wills of your own and must follow me because I know better'; but rather, 'Here are the goals which are true and right and which we share. Here is - 197 -how we can reach them. You are strong and capable. You can accomplish these goals! 1 His role is to make clear which are the goals the group should achieve, and then to create confidence in its members that they can achieve them. Inspection of the means on the charismatic manipulation checks in Table Q16 (see Appendix Q) provides support for the theoretical explanation of charismatic effects discussed above. Individuals with charismatic leaders described their leaders as exhibiting the following behaviours to a significantly greater extent than individuals with considerate leaders: (a) she vi v i d l y describes the importance of the project (M r. = 4 ' 4 8 ' ^ Cons = 2 ' 3 5 ) ; (b) she enthusiastically expresses work goals (M r. = 4.27; M „ =2.17); -<~nar — C o n s (c) she expects you to strive for high work standards (M„. = 3 ' 7 3 ' Mcons = 2 ' 0 0 ) ; (d) she inspires you to do your very best on the exercise (M r, = / * ' 2 5 ' Mcons = 2 ' 3 1 ) < -(e) she is confident in your ability to perform well on the exercise (M~. = 4.13; Mn = 2.42). — C h a r —Cons Collectively, these results suggest that charismatic leaders, in comparison to considerate leaders, may have strengthened and uplifted their subordinates, thereby leading to higher task performance and adjustment. The lower task performance and adjustment of individuals with considerate leaders relative to those with charismatic leaders supports previous theoretical literature and empirical findings. For example, studies of the Ohio State dimensions of consideration and initiating structure have generally found that considerate leader behaviour is not - 198 -related to task performance (e.g., Stogdill, 1974). Similarly, path-goal formulations suggest that in unstructured task conditions, considerate leaders have no effect on individuals' task performance and satisfaction (e.g., Downey, Sheridan, & Slocum, 1975; Greene, 1979; House, 1971; House & Dessler, 1974; Stinson & Johnson, 1975). Hypothesis 3. Individuals working under a structuring leader will have higher task performance than will individuals working under a considerate leader. Hypothesis 4. Individuals working under a structuring leader will report higher task adjustment than will individuals working under a considerate leader. Contrary to prediction, there were no significant differences on task performance measures for individuals working under structuring and considerate leaders. Across both leadership styles, the participants suggested 19 courses of action on average and had fairly satisfactory quality of performance (M = 2.70 on a 5-point scale) (see Table 7). With regard to task adjustment, while individuals with structuring leaders had significantly higher role clarity (M = 28.77) and lower role conflict (M = 19.42) than individuals with considerate leaders (M = 16.43; M = 22.06; p_ < .01) there were no statistically significant differences for general task satisfaction (M = 8.60; M = 7.38; £ < .01) and specific task satisfaction (M = 30.83; M = 29.73; £ < .01) measures (see Table 7). These results indicate that structuring leaders provided subordinates with role clarity by defining work role expectations and reduced role conflict. They indicate. - 199 -however, that these positive effects did not generalize to other aspects of task adjustment or to task performance. The above findings are inconsistent with previous research suggesting that under ambiguous task conditions or in unfavourable situations, structuring leader behaviour is positively associated with subordinate task performance, task satisfaction, and role clarity (e.g., Chemers S Skrzypek, 1972; Fleishman, Harris, S Burtt, 1955; Cilmore, Beehr, & Richter, 1979; Greene, 1979; Gustofson, 1968; House & Dessler, 1974; Schriesheim & Murphy, 1976). In the present study, the provision of specific task directives by the structuring leader were not sufficient to produce high task satisfaction or performance. Several explanations for these results may be advanced. Perhaps the manipulation of structuring leader behaviour was not strong enough to produce the effects hypothesized in this study. However, statistical analyses of the leadership style manipulation checks indicated highly significant differences between the structuring and considerate styles on these manipulation checks thereby reducing the viability of this explanation (see Table Q17 in Appendix Q for complete details). Alternatively, when the task is sufficiently ambiguous, leader structuring behaviour, such as was provided in this study, may simply not be enough to produce a significant change in performance. Instead, subordinates may need to be activated and energized to meet the challenging nature of the task. Examination of the participants' written comments regarding their impressions of the task and the leader revealed - 200 -several themes supporting this explanation. As expected, approximately 85% of the individuals with considerate leaders mentioned while the leader was friendly and approachable, they were confused about how to do the exercise. S u r p r i s i n g l y , 60% of the individuals with structuring leaders reported while the task was clearly explained to them by the leader and they had the opportunity to ask questions, the exercise was still vague and they were unsure of how to handle it. They also commented that the task seemed like a lot of work and was not very exciting or stimulating. T h i s anecdotal evidence suggests while the task was clearly defined by the structuring leader the participants still felt uncertain as to how to proceed and were not highly motivated to do so. Another explanation for these findings is that in comparison to prior empirical studies, this laboratory experiment may have created more ambiguous circumstances for participants to cope with, thereby impeding their potentially positive task adjustment and performance under structuring leadership. Specifically, previous studies examining the effectiveness of structuring leader behaviour on employees' performance and satisfaction under ambiguous task conditions have typically been conducted in field settings. Therefore, while there was variation in the ambiguity of the employees' jobs, other aspects of their work environment were very familiar to them (e.g., their manager, their peers, the office layout, and so on). In contrast, in the present study, participants were involved in a very ambiguous work situation which included new organi-zational members (i . e . , the leader and the co-workers), a new work setting, and a new task. - 201 -Another interpretation for the present findings is that this study did not examine leadership in isolation; both the leader and the group exerted influence on the participants' behaviour. Therefore, the group productivity norms may have affected the salience of the structuring leader's effect on individuals' task performance and adjustment. This explanation will be further developed in the discussion of interaction effects presented later in this chapter. Hypothesis 5. Individuals working under a charismatic leader will have the same level of task performance as individuals working under a structuring leader. Hypothesis 6. Individuals working under a charismatic leader will report the same level of task adjustment as individuals working under a structuring leader. An unexpected finding was while there was no significant difference for qualitative task performance, individuals with charismatic leaders produced significantly more courses of action on the in-basket exercise (M = 24) than individuals with structuring leaders (M = 19; £ < .01) (see Table 7). Moreover, while perceived role clarity was virtually equivalent for individuals with charismatic and structuring leaders (M = 28.65; M = 28.77; £ > .01), individuals with charismatic leaders reported significantly higher general task satisfaction (M = 11.06; .M = 8.60; £ < .01), higher specific task satisfaction (M = 39.44; M = 30.83; £ < .01), and lower role conflict (M = 15.48; M = 19.42; £ < .01) (see Table 7). These results suggest while both charismatic and structuring leaders provide individuals - 202 -with task c l a r i t y , individuals with charismatic leaders have higher task performance in terms of the number of courses of action suggested, greater task satisfaction, and lower role conflict than individuals with structuring leaders. In light of recent comparative analyses of exchange and charismatic theories of leadership, the results of this study appear highly plausible (Bass, 1985; House, 1985b). For example, Bass (1985) argues that exchange or transactional leaders, by clarifying subordinates' role requirements and by explicating how subordinates' needs and wants will be fulfilled in exchange for subordinates' satisfactory performance, motivate subordinates to exert expected effort. In contrast, charismatic or transformational leaders, by appealing to higher values that express subordinates' fundamental needs, motivate followers to exert effort beyond the ordinary limits. House (1985b) has embellished this theme in his insightful discussion of exchange and charismatic theories of leadership. According to him, the essential difference between exchange and charismatic leader behaviour resides "in the components of the subordinate's motivation that are affected by leader behaviour and by the kinds of leader behaviour that affect components of the subordinate's motivation" (House, 1985b, p.1). Specifically, exchange oriented leaders accept subordinates' goals, values, and desires as given; they focus on subordinates' abilities and cognitions required to meet these goals. Charismatically oriented leaders strive to change subordinates' needs and values and therefore their goals; they focus on the level and kinds of goals to which subordinates aspire. - 203 -Therefore, the foregoing analyses of exchange and charismatic theories of leadership clearly suggest that in accordance with this study's findings, charismatic leadership would result in higher subordinate task performance than structuring leadership. It should be noted that this assertion rests on the assumption that structuring leadership is reflective of exchange or transactional leadership. To gain further insight into the present study's findings, it is instructive to examine the specific components of charismatic leader behaviour which may have influenced these results. For example, charismatic leaders, by imputing a mission and ideological goals which heighten subordinates' perceptions of the meaning and importance of their specific task assignment, may forge a strong affective connection between subordinates and their jobs (e.g.. House, 1977; Smith, 1982). This contention has been supported by a recent field study which found that subordinates of charismatic leaders experienced greater meaningfulness of work than subordinates of noncharismatic leaders (Smith, 1982). Thus a crucial dimension that may influence the degree of individual task adjustment and performance is the articulation of goals in terms of values that have ideological significance for subordinates. As Berlew (1974, p.269) notes: The f i r s t requirement for.. .charismatic leadership is a common or shared vision for what the future could be. To provide meaning and generate excitement, such a common vision must reflect goals or a future state of affairs that is valued by the organizations' members and thus important to them to bring about All inspirational speeches or writings have the common element of some vision or dream of a better existence which will inspire or excite those who share the author's values. - 204 -A related explanation for these results is that charismatic leaders communicate messages that arouse motives especially relevant for task accomplishment (House, 1977). Recent research evidence supports this explanation. For example, in a survey of 70 male executives' reactions to transformational leaders they had encountered during their careers, Bass (1985) found that the executives worked "ridiculous" hours, demonstrated higher quality of performance, exhibited greater innovativeness, and expressed a readiness to extend and develop themselves further. Moreover, results of studies conducted in military, industrial, and educational settings have consistently revealed that subordinates with transformational leaders exert greater effort than subordinates with transactional leaders (e.g., Avolio et a l . , 1985; Bass, 1985; Waldman et a l . , 1985). Similar results are reported by Smith (1982) in a study of subordinate responses to charismatic and noncharismatic leaders. He found that subordinates with charismatic leaders worked longer hours and had higher performance ratings than those with noncharismatic leaders. Smith interpreted these results as reflecting the heightened motivation and effort by subordinates in support of the goal of the charismatic leader. Inspection of the means on the charismatic leadership style manipulation checks lends support to this interpretation (see Table Q16 in Appendix Q). Individuals with charismatic leaders described their leaders as exhibiting the following behaviours to a significantly greater extent than those with structuring leaders: (a) she inspires you to do your very best on the exercise (M r. = 4.25; M c + = 2.54); ^ n a r ' — S t r u c - 205 -(b) she increases your motivation to do the exercise ( M „ . = 3.90; t. T i.T». — C h a r M c + = 2.42); — S t r u c (c) she really makes you want to do the exercise ( M r . = 3.73; M s t r u c = 2.27); " C h a r (d) she encourages you to be creative and productive in doing the exercise ( M _ . = 4.40; M C J . = 2.27). — C h a r ' — S t r u c In addition, approximately 75% of the individuals with charismatic leaders anecdotally expressed enthusiasm about the project and felt they were "contributing to something really important, something that would impact on their future". Thus the charismatic leader may act as a catalyst for subordinates' effective task performance and adjustment. Adjustment to the Leader Hypothesis 7. Individuals working under a charismatic leader will report higher adjustment to the leader than will individuals working under a structuring leader. As predicted, individuals reported significantly greater satisfaction, personal liking , comfortableness, and willingness to work with a charismatic leader (M = 30.44) as compared to individuals with a structuring leader (M = 16.49) (see Table 7). Specifically, individuals with charismatic leaders experienced very high adjustment to the leader ( M = 6.09 on a 7-point scale) while those with structuring leaders experienced moderate adjustment-to the leader (M = 3.30 on a 7-point scale). The positive interpersonal relationship between charismatic leaders and their subordinates supports the theoretical literature on charisma. For example, Weber (1947) recognized that a fundamental aspect of - 206 -charisma was the extraordinary, intensely personal relationship between a charismatic leader and his/her followers. In his elaboration of Weber's conceptualization of the charismatic relationship. Tucker (1970, p.73) observes: Oftentimes, the relationship of the followers to the charismatic leader is that of disciples to a master, and in any event he is revered by them. They do not follow him out of fear or monetary inducement, but out of love, passionate devotion, and enthusiasm. Thus the initial and continuing appeal of charismatic leaders is based on emotional rather than rational grounds in that "the follower is inspired enthusiastically to give unquestioned obedience, loyalty, commitment and devotion to the leader and to the cause that the leader represents" (House, 1977, p.191). Theoretical statements regarding the intense emotional attachment between charismatic leaders and their followers are supported by field research. For example, in-depth case studies of charismatic leaders in diverse organizational settings and roles have described the presence of strong affective bonds, characterized by mutual respect, caring, and trust, between charismatic leaders and their followers (Day, 1980; Roberts, 1984). In addition, Bass (1985) found that strong liking, admiration, respect, loyalty, and trust typified the relationship between senior executives and transformational leaders they had encountered during their career. However, as Martin and her colleagues (1983) have noted, in most modern organizations social norms inhibit the open expression of strong feelings of attraction. Rather, they suggest that employees might indicate a deep emotional attachment to a charismatic leader by describing - 207 -his/her abilities or qualities in superhuman terms, by expressions of unusual warmth or emotion-toned admiration, or by going to extraordinary lengths to display loyalty or productivity. It should be underscored that while the results of the present study indicated that a positive interpersonal relationship was established between the charismatic leader and her followers, this relationship did not appear to have the depth or intensity suggested in the theoretical literature. The observed differences in individuals' interpersonal adjustment to the structuring and to the charismatic leaders are consistent with Smith's (1982) postulations regarding the distinction between charismatic and instrumental leader behaviours. He contends while charismatic leaders strive to establish a strong affective bond with subordinates, instrumental leaders view their relationship with subordinates in more calculative, instrumental terms (Smith, 1982, p.19). Thus it is expected that subordinates of an instrumental leader would have moderate levels of trust in their leader, reflecting their evaluation of the leader's expertise to assist them toward goal attainment. In contrast, it is expected that subordinates of a charismatic leader would have high levels of trust in their leader and would willingly put their fate in their leader's hands. Smith (1982, p.20) further posits that levels of liking, attraction, and affection for the leader would be high and generalized for "charismatic" followers, and moderate and more specifically calculative for "instrumental" followers. That is', subordinates of an instrumental leader would be attracted to their leader as a specific source of valued extrinsic rewards and as a facilitator of goal accomplishment, but would not hold the same generalized high level of affection and attraction toward their leader as would subordinates of a charismatic leader (Smith, 1982, pp.20-21). - 208 -Empirical support for these predictions is provided by Smith (1982) in his study of subordinates' responses to charismatic and noncharismatic leaders. He found that in comparison to subordinates of noncharismatic leaders, subordinates of charismatic leaders had higher levels of trust and acceptance by the leader. Consistent results are reported by House (1985a) in his biographical analysis of charismatic and noncharismatic heads of state. He discovered that followers of charismatic leaders had higher obedience, acceptance, and trust in their leader than did followers of noncharismatic leaders. Therefore, the findings of this study, that individuals with charismatic leaders have higher adjustment to the leader than individuals with structuring leaders, are congruent with results from studies conducted in organizational and political contexts. Hypothesis 8. Individuals working under a considerate leader will report higher adjustment to the leader than will individuals working under a structuring leader. The present study's findings supported this hypothesis: individuals working under a considerate leader had significantly higher adjustment to the leader (M = 26.77) as compared to those working under a structuring leader (M = 16.49; £ < .01) (see Table 7). This finding is essentially in agreement with those obtained by earlier researchers which suggest that considerate leaders appear to have their primary effect in terms of social or psychological maintenance. For example. Bales and Slater (1955) observed that socioemotional oriented leaders provided social satisfaction for group members which resulted in reduction of group frustration and stress. In an exhaustive review of the leadership literature, Stogdill - 209 -(1974, p.404) noted that the relationship between leadership styles and satisfaction indicates that "person oriented patterns of leadership tend to enhance employee satisfaction". Bass (1981, p.382), in his update of Stogdill's (1974) Handbook of Leadership, reached the same conclusion: "generally, supervisory consideration seems to be associated with subordinate satisfaction with their supervisors". In addition, under conditions of role stress, several researchers have reported that considerate leader behaviour serves as a source of social satisfaction and support for the employee (e.g.. House, 1981; Seers et a l . , 1982; Sheridan & Vrendenburgh, 1979). Finally, the present study's findings are also in agreement with laboratory studies using confederates as leaders (e.g., Tjosvold, 1984; Weed, Mitchell, S Moffitt, 1976). For example. Weed, Mitchell, and Moffitt (1976) found that subordinates with leaders high in human relations orientation and low in task orientation were satisfied and enjoyed working with such leaders. A study by Bales (1958) is especially relevant to the present study's finding that individuals reported neither positive nor negative adjustment to the structuring leader. He hypothesized that the ratio of verbal interaction received to that initiated would distinguish between task oriented leaders who were well liked and those who were not well liked in small experimental discussion groups. The results indicated that there was no relationship between task orientation and liking for leaders who made it possible for group ' members to give feedback and to raise objections, qualifications, and questions. Therefore, in accordance with the present study's finding, when the structuring leader is responsive to participants' input, there is an absence of association between structuring leadership and participants' interpersonal adjustment to the leader. - 210 -Hypothesis 9. Individuals working under a charismatic leader will report the same level of adjustment to the leader as individuals working under a considerate leader. Contrary to prediction, the results indicated that individuals with charismatic leaders had significantly higher adjustment to their leader (M = 30.44) than individuals with considerate leaders (M = 26.77; £ < .01) (see Table 7). T h i s finding seems su r p r i s i n g given the considerate leader's exclusive focus on establishing a strong emotional bond with subordinates by conveying warmth, acceptance, support and reassurance. However, recent theoretical literature sheds light on this result. In an analysis of charismatic political leadership, Willner (1984) observes while the charismatic relationship is not different from considerate leadership since with each style an emotional bond is established between leaders and followers, it is the quality and intensity of this emotional bond between charismatic leaders and followers that are the significant factors. As discussed earlier, charismatic leaders are thought to generate devotion, awe, and reverence in their followers. T hus it is suggested that charismatic leaders establish a qualitatively different interpersonal relationship with their subordinates as compared to other leadership styles. A related explanation for this result is that under ambiguous conditions individuals may need more than rapport with their leader. Charismatic leaders not only emphasize the importance of the individual, they also have an ability to make individual followers feel powerful, efficacious, and important (McClelland, 1975). For example, one source of John F. Kennedy's charisma may have been this ability; students felt -211 -stronger and more powerful after watching a movie of his inaugural address as President ("Ask not what your country can do for you...") (McClelland, 1975, p.259). Thus emotional attachment and ensuing transfer of feelings of power may be aspects of charismatic leadership that enhance the interpersonal perception of the leader. Moreover, followers' positive emotional attachment to the leader may in turn result in high levels of commitment to the task, motivation, and goal attainment. Thus the charismatic leader can be seen as an object of identification after which followers model their values, goals, and behaviour ( F r i e d r i c h , 1961). Another plausible explanation for more favourable subordinate perceptions of charismatic leaders as compared to considerate leaders is while both leaders are likeable, only charismatic leaders meet subordinates' expectations regarding task clarification. This explanation is consistent with the idiosyncrasy credit concept advanced by Hollander (1964). In attempting to explain the emergence of leadership and the determinants of leader effectiveness within groups, Hollander (1964) asserts that group members' judgements of an emergent leader will be positive to the extent that the leader conforms to expectations and contributes toward the group's goal. Therefore, the more positively disposed impressions of charismatic leaders may be due to their ability to meet followers' expectations regarding their work situation more adequately than considerate leaders. In summary, the foregoing discussion suggests that individuals with charismatic leaders had higher task performance in terms of the number of courses of action suggested, greater task satisfaction, lower role conflict. - 212 -and higher adjustment to the leader in comparison to individuals with structuring and considerate leaders. In addition, the qualitative task performance of individuals with charismatic leaders surpassed that of individuals with considerate leaders. These findings were attributed to several interrelated facets of charismatic leader behaviour: establishing a strong affective connection with subordinates and making them feel relatively powerful and efficacious; imputing a mission and ideological goals which heighten subordinates' work motivation, effectiveness, and per-ceptions of the meaning and importance of their task; and simultaneously communicating high expectations of, and confidence in, subordinates' ability to meet these expectations. As predicted, individuals with considerate and charismatic leaders had higher adjustment to the leader than individuals with structuring leaders. A s u r p r i s i n g result was while individuals with structuring leaders experienced role clarity and lower role conflict, their task performance and task satisfaction was equivalent to those with considerate leaders. It was suggested that the work situation was sufficiently ambiguous that the role clarification behaviour of structuring leaders was simply not enough to produce significant changes in participants' task performance and task satisfaction. This is borne out when one compares charismatic and structuring leaders where role clarity is high under both leaders yet performance and adjustment are different, suggesting once again that something more than clearly defining task expectations for subordinates is needed to produce high task performance and adjustment. It was further suggested that the presence of group members espousing high or low productivity norms may have influenced the salience of the structuring - 213 -leader's effects on the participants' adjustment to and performance on the task. A statistically significant interaction effect was found between leadership style and group productivity. Therefore, interpretation of these results needs to be tempered by consideration of this interaction effect. GROUP PRODUCTIVITY Task Performance Hypothesis 10. Individuals in high productivity groups will have higher task performance than individuals in low productivity groups. An unexpected finding was the absence of a statistically significant group productivity effect for task performance. On average, individuals in the high and low productivity groups suggested 21 courses of action and had satisfactory qualitative performance (M = 2.95) on the experimental task (see Table 7). Several possible interpretations for these findings can be offered. Perhaps the manipulation of high and low productivity norms was not sufficiently strong to cause changes in individual task performance. However, the author felt that increasing the potency of the group productivity manipulation might be perceived by the participants as lacking in authenticity and thereby create demand characteristics. That is, if the co-workers had engaged in excessively disruptive activities during the experimental session (e.g., talking continuously; leaving the office for coffee) the participants' suspicions may have been aroused about - 214 -the realism of their situation. Moreover, the group productivity manipulation checks indicated highly significant differences between high and low group productivity conditions (see Table Q18 in Appendix Q). Therefore, other explanations seem more plausible. Examination of the group dynamics literature sheds some light on this finding. According to some researchers (e.g., Cartwright & Zander, 1960; Thibaut £ Strickland, 1956), the nature of one's membership in a group influences one's tendency to model his/her attitudes and actions after those of other members. Specifically, pressures for conformity to group behaviours or beliefs are highly ineffective in changing member behaviour if he/she is not concerned with maintaining membership in the group that is exerting influence on him/her. Thus, given the participants' temporary membership in a group, the pressures on them to act in accordance with group productivity norms may have been weakened. Thomas and G r i f f i n (1983) reached the same conclusion in their review of the literature dealing with the effects of social and information cues in the workplace on employees' task perceptions, evaluations, and reactions. They asserted that "salient cues may have a greater impact on attitudes and behaviour on an actual job in which continued employment and promotions are contingent on at least partially accurate perceptions of tasks than in the laboratory" (Thomas & G r i f f i n , 1983, p.675). White and Mitchell (1979, p.8J concur with this assertion: "...the effect of the comments of an unknown co-worker in a short work session would intuitively seem to be less important than the comments of a co-worker with whom one works 8 hours a day, 5 days a week, because the ad hoc - 215 -nature of the present groups probably produced far less social pressure to conform than could be expected by a member of a long-term integrated work team." Another closely related explanation for the nonsignificant result may be that conformity to group productivity norms requires more extensive interaction over a longer period of time between the individual and the group. Initially, newcomers may be absorbed in coping with the new task realities in their work situation. Over time, however, the group could exert increasing pressure on newcomers for conformity to high or low productivity standards. T h i s contention is supported by the organizational socialization literature on new employees' integration into their job settings (e.g., Feldman, 1981; Katz, 1980; Van Maanen S Schein, 1979). As Katz (1980, p. 110) has observed, "the new secretary, engineer, or technician faithfully arrives and leaves work at the regularly scheduled hours and seems to perform all task assignments willingly and diligently - at least initially. With increasing awareness of one's job setting, its actual practices, procedures, and norms, the individual is soon freer to decide which tasks will be performed carefully or promptly as well as develop his/her own interpretation of permissible working hours." The argument that conformity to group productivity norms requires more extensive interaction between the individual and the group is further supported by case studies of shop floor behaviour (e.g., Burawoy, 1979; Schrank, 1978). For example, Robert Schrank, in his description of his life in a furniture factory, reported that for the f i r s t few weeks on the job, he busily kept the machine operators supplied with material. It was - 216 -not until his third or fourth week at the factory that he was inculcated with two fundamental lessons of working from his peers; how to work less hard in order to make the task easier and don't do more work than is absolutely necessary (Schrank, 1978). Therefore, as the length of one's membership in a work group increases, the influence of group productivity norms on individual task performance might be more potent. A final interpretation of these findings is that leadership style may affect the salience of group productivity norms on individual task performance. T h i s explanation will be amplified further in the discussion of the interaction effects. Adjustment to the Task and to the Group Hypothesis 11. Individuals in high productivity groups will report higher task adjustment than individuals in low productivity groups. Hypothesis 12. Individuals in high productivity groups will report higher adjustment to the group than individuals in low productivity groups. As predicted, individuals in the high productivity group, as compared to the low productivity group, had higher adjustment to the group (M = 60.75; M = 49.99; £ < .01) (see Table 7). The results further indicated while there was an absence of a statistically significant group productivity effect for role ambiguity, those in the high productivity group had higher general task satisfaction (M = 9.82; M = 8.21; £ < .01), higher specific task satisfaction (M = 36.43; M = 30.24; £ < .01), and lower role conflict (M = 17.58; M = 20.39; £ < .01) than those in the low productivity group (see Table 7). - 217 -These findings are consonant with laboratory studies examining the influence of social information cues about a task on individuals' assessment of task characteristics. Specifically, these studies have consistently shown that in comparison to negative social cues, positive social cues provided by confederate co-workers result in higher task satisfaction and more favourable perceptions of task characteristics (e.g.. G r i f f i n , 1983; Vance & Biddle, 1985; White & Mitchell, 1979). Therefore, the social cues of co-workers are an important determinant of individuals' perceptions of the task environment. The small group research literature sheds some light on the present study's findings. For example, Berkowitz (1954) reported that participants liked their partners (confederates) better when the latter were supposedly proficient on a task than when they were supposedly poor. Similarly, Zander (1968) found that members of successful groups are more likely to experience satisfaction, form a favourable impression of themselves and other members of the group, and wish to continue to pursue the activity on which the group was successful. Thus to the extent the participants in high productivity groups perceived their co-workers as successfully accomplishing the task at hand, the more satisfied they would be with their co-workers and with the task. Another possible interpretation of these findings is that group norms for high productivity were congruent with the task demands faced by individuals, thereby leading to positive adjustment to the task and to the group. In contrast, the group norms for low productivity were incompatible with the task demands faced by individuals, thereby leading - 218 -to role conflict, dissatisfaction with the task, and reduced liking for group members from whom the conflict stems. These results are consistent with the empirical evidence regarding role conflict; role conflict has been shown to be directly related to psychological withdrawal from the group and job induced tension and inversely related to job satisfaction (e.g.. Brief £ Aldag, 1976; House £ Rizzo, 1972; Kahn et a l . , 1964; Miles, 1976). Individuals in both high and low productivity groups reported neither role ambiguity nor role clarity (M = 24.62; 4.10 on a 7-point scale) (see Table 7). One possible explanation for this result is since the co-workers did not provide task direction, clarify task responsibilities, or obfuscate the leader's description of the task, the participants' task expectations were not influenced. T h i s suggests that in order to increase or to decrease participants' role ambiguity, the group productivity manipulation would need to be expanded to incorporate the provision of task information. In summary, individuals in the high productivity group reported greater task satisfaction, lower role conflict and higher adjustment to the group than individuals in the low productivity group. These findings suggested that group norms for high productivity were congruent with the task demands faced by the individual, thereby leading to positive adjustment to the task and to the group. The lack of significant effects for group productivity norms on individual task performance was su r p r i s i n g . It was suggested that conformity to group productivity norms require more extensive interaction over a long period of time between the individual and the group. Moreover, the salience of the group - 2 1 9 -productivity effect may have been reduced by the presence of a leader who also exerted influence on the participants' behaviour. Therefore, the interpretation of the group productivity main effect needs to be qualified by the leadership style-group productivity interaction. LEADERSHIP S T Y L E x GROUP PRODUCTIVITY Hypothesis 13(a). Individuals exposed to a structuring leader and in a high productivity group will have higher task performance, task adjustment, and adjustment to the leader and to the group than individuals exposed to a structuring leader and in a low productivity group. As predicted, individuals with a structuring leader and in a high productivity group reported higher task adjustment than individuals with a structuring leader and in a low productivity group. Specifically, under structuring leadership, individuals in high productivity groups experienced higher general task satisfaction (M = 10.88; M = 6.33; £ < .01), higher specific task satisfaction (M = 37.25; M = 24.42; £ < .01), and lower role conflict (M = 14.33; M = 24.50; £ < .01) than individuals in low productivity groups (see Table 8). There were no significant interaction effects for role ambiguity (M = 30.92; M = 26.63; £ > .01) (see Table 8). These results suggest for the structuring leader-high group productivity condition, the leader and the group structure the participant's reality in mutually reinforcing ways by advocating task accomplishment, thereby leading to task satisfaction and minimal role conflict. For the structuring leader-low group productivity condition, the contradiction between the - 220 -structuring leader's emphasis on task accomplishment and the group's disinterest in performing the task creates substantial role conflict and dissatisfaction with the exercise. These results are consistent with previous research (e.g., Beehr, Walsh, & Taber, 1976; Brief & Aldag, 1976; House & Rizzo, 1972). The nonsignificant interaction effect for role ambiguity may be attributed to the fact that the structuring leader provided detailed task direction to participants while the group did not offer any information that may have confused or hindered the participants' understanding of these directions; therefore, participants had fairly clear perceptions of the task requirements (M = 28.77; 4.80 on a 7-point scale) (see Table 7). The results further indicated a lack of significant interaction effects for task performance and interpersonal adjustment measures. With regard to task performance, individuals suggested 19 courses of action and had satisfactory quality of performance (M = 2.94 on a 5-point scale) (see Table 7). The interpersonal adjustment data revealed that individuals were moderately adjusted to the leader (M = 16.49; 3.30 on a 7-point scale) and relatively highly adjusted to the group (M = 58.18; 5.81 on a 8-point scale) (see Table 7). These nonsignificant results are not consistent with previous empirical research which suggests that role conflict leads to less effective performance, psychological withdrawal from the group, and reduction in trust, l i k i n g , and respect for role senders from whom the conflict stems (e.g., French & Caplan, 1972; Liddell & Slocum, 1976; Miles, 1976). - 221 -Several possible explanations for the present study's findings may be advanced. For task performance, role clarity may have moderated the deleterious effects • of role conflict, thereby facilitating task performance for individuals in low productivity groups. Alternatively, for an individualistic task where the participant is unilaterally responsible for its completion, individuals may have responded to induced role conflict by becoming more involved in the task. Katz (1977) reached the same conclusion in his study of the influence of structuring and considerate leaders on individual performance under conditions of high affective group conflict. Another explanation is that leader structuring behaviour and high group productivity norms may simply not be sufficient to produce outstanding task performance. As discussed earlier, participants anecdotally reported while the structuring leader clearly defined the task, they felt unsure as to how to proceed and were not highly motivated to do so. Therefore, individuals may need to be activated or energized in order to meet the challenge of the task. A final explanation is perhaps that as an individual becomes more established in his/her organizational role and work situation over time, structuring leadership in combination with high group productivity norms would produce higher individual task performance than would structuring leadership in combination with low group productivity norms. - 222 -The lack of significant interaction effects for measures of adjustment to the leader and to the group suggests that the evaluation of inter-personal relations may be separate from the evaluation of the task. That is, individuals' differential adjustment to the task is independent of favourable impressions of the leader and of the group. This contention is supported by the generally weak correlations between adjustment to the leader and task adjustment, and between adjustment to the group and task adjustment (see Table 6). Examination of the operational definitions of the interaction style of the leader and of the co-workers sheds further light on these nonsignificant results. The structuring leader acted in a neutral manner towards participants; the co-workers acted in a friendly and interested manner towards the participants regardless of productivity norms. Therefore, the participants were moderately adjusted to the leader and relatively highly adjusted to the co-workers (see Table 7 ) . Hypothesis 13(b). Individuals exposed to a considerate leader and in a high productivity group will have higher task performance, task adjustment, and adjustment to the leader and the group than individuals exposed to a considerate leader'and in a low productivity group. In accordance with the hypothesis, individuals with a considerate leader and in a high productivity group had significantly higher specific task satisfaction (M = 32 .25 ) than individuals with a considerate leader and in a low productivity group (M = 2 7 . 2 1 ; £ < .01) (see Table 8 ) . The considerate leader, by providing interpersonal support, and the high productivity group, by providing encouragement to do the task, acted in a complementary manner, thereby increasing individual task satisfaction. In - 223 -contrast, in the consideration-low productivity condition, the considerate leader, by providing interpersonal support, and the low productivity group, by providing discouragement to do the task, acted in a discordant manner, thereby decreasing individual task satisfaction. The remaining results were incongruent with this experimental hypothesis. The task performance data indicated that under both group productivity conditions, individuals exposed to a considerate leader recommended 19 courses of action and had fairly satisfactory quality of performance (M = 2.46 on a 5-point scale) (see Table 7). With regard to the task adjustment measures, individuals reported moderate general satisfaction with the task (M = 7.38; 3.69 on a 7-point scale), considerable role ambiguity (M = 16.43; 2.74 on a 7-point scale, reverse scored), and moderate role conflict (M = 22.06; 4.41 on a 7-point scale) (see Table 7). Collectively, these results suggest that individuals felt confused and uncertain about how to approach the task resulting in low task adjustment and marginal task performance. Examination of the operationalizations of considerate leadership and group productivity offers some insights into these results. As discussed in Chapter II, considerate leadership was operationally defined as concern for the personal welfare of the participant, engagement in participative two-way conversations, and emphasis on the comfort, well being, and satisfaction of the" participant. The group productivity manipulation focused on providing task performance cues by displaying either highly productive or minimally productive behaviour on the part of confederate co-workers. Therefore, the participants did not actually receive explicit - 224 -task directions, facilitative of goal accomplishment, from either the leader or the co-workers. Accordingly, the participants' task performance and task adjustment was marginal. The results further indicated that individuals reported positive adjustment to the leader (M = 26.77; 5.35 on a 7-point scale) and fairly high adjustment to the group (M = 53.21; 5.32 on a 8-point scale) (see Table 7). Thi s contradicts previous research which suggests that role conflict and role ambiguity are related to unsatisfactory work group relationships and inadequate perceived leader behaviour (e.g., French & Caplan, 1972; Rizzo et a l . , 1970; Van Sell et a l . , 1981). However, as discussed earlier, perhaps the adverse effects of role stress do not necessarily generalize to perceptions of interpersonal relations. The considerate leader acted in a reassuring and supportive manner and expressed concern about the individual's personal welfare; the group consistently acted in a friendly manner regardless of productivity norms. Hence the participants had a favourable impression of both the group and the leader. Hypothesis 13(c). Individuals exposed to a charismatic leader and in a high productivity group will have the same level of task performance, task adjustment, and adjustment to the leader and to the group as individuals exposed to a charismatic leader and in a low productivity group. The results of the present study supported this hypothesis. The directionality of group productivity norms appeared to be nullified by the charismatic leader resulting in high task performance, task adjustment, - 225 -and adjustment to the leader and to the group. With regard to task performance, individuals provided 24 courses of action and had satisfactory quality of performance (M = 3.46 on a 5-point scale) (see Table 7). For task adjustment, participants reported high specific task satisfaction (M = 39.44), high general task satisfaction (M = 11.06; 5.53 on a 7-point scale), moderate role clarity (M = 28.65; 4.77 on a 7-point scale), and low role conflict (M = 15.48; 3.09 on a 7-point scale (see Table 7). Finally, individuals reported very positive adjustment to the leader (M = 30.44; 6.01 on a 7-point scale) and moderately high adjustment to the group (M = 54.73; 5.47 on a 8-point scale) (see Table 7). These results lend support to the theoretical literature which suggests that charismatic leaders "are capable of having profound and extraordinary effects on followers" (House, 1977, p.189). As discussed earlier, charismatic leaders, by establishing a strong emotional bond with followers, by providing an ideological goal which heightens followers' work motivation and perceptions of the meaning and significance of their work, and by expressing confidence in followers' ability to meet high performance expectations, facilitate high individual task performance, task adjustment, and interpersonal adjustment. In summary, the foregoing discussion of interaction effects suggests that charismatic leaders, regardless of group productivity norms, foster high individual task' performance, task adjustment, and adjustment to the leader and to the group. Therefore, charismatic leaders overcome group pressures for low task productivity and augment group pressures for high task productivity, thereby facilitating individuals' adjustment and performance in a new work setting. - 226 -In contrast to the charismatic leader, the structuring leader's impact on individual's task adjustment was modified by group productivity norms. Individuals who worked with a structuring leader and in a high produc-tivity group reported higher specific task satisfaction, higher general task satisfaction, and lower role conflict than individuals who worked with a structuring leader and a low productivity group. Therefore, in order to reach the high level of task satisfaction of individuals with charismatic leaders, individuals with structuring leaders may need the support of high group productivity norms. Interestingly, despite the significant interaction effect of structuring leadership and group productivity norms on individual task adjustment, there was an absence of statistically significant effects of structuring leadership and group productivity norms on task performance, role ambiguity, and adjustment to the leader and to the group. Perhaps as individuals become more established in their organizational roles and work situations over time, structuring leadership and high group productivity norms would influence their task performance, role c l a r i t y , and interpersonal adjustment. Finally, individuals with a considerate leader and in a high productivity group had significantly higher specific task satisfaction than those with a considerate leader and in a low productivity group. Thus the considerate leader, by providing interpersonal support, and the high productivity group, by providing encouragement to do the task, acted in a complementary manner, thereby increasing individual task satisfaction. The results further indicated that individuals exposed to considerate leaders, regardless of the directionality of group productivity norms, had marginal task performance and low task adjustment. However, individuals reported positive adjustment to the leader and to the group. - 227 -Overall, individuals with structuring and considerate leaders had highly similar task performance and adjustment to the group. However, individuals in the structuring-high productivity group had more positive task adjustment than those with considerate leaders. On the other hand, individuals with considerate leaders had higher adjustment to the leader than those with structuring leaders. Nonsignificant Results As mentioned earlier, there were no significant effects for three dependent measures: the number of in-basket items completed, self-rated performance, and job related tension. These results indicated that across all conditions the participants attempted on average 15 out of 20 memos, perceived their task performance as adequate (M = 22.31; 3.16 on a 5-point scale) and reported occasional tension associated with the task (M = 36.59; 2.81 on a 5-point scale) (see Table 7). A possible explanation for the lack of significant differences for the number of in-basket items attempted is that individuals' quantitative performance may not have been adequately tested by this measure. That is, due to the short length of the in-basket exercise, the detection of possible group differences may have been attenuated. If, for example, the exercise had consisted of 60 memos, some group differences may have been evident. Another related explanation for the absence of significant differences is that the number of in-basket items attempted was not a sufficiently - 228 -sensitive measure of task performance under ambiguous circumstances. That is , all participants, regardless of treatment conditions, were aware of the number of memos to be completed within the specified time period. Therefore, this aspect of the task was clearly defined and, consequently, the participants handled approximately the same number of memos. However, it is likely that the participants faced ambiguity in deciding on their quality of task performance and on their generation of alternative courses of action to the memos. Therefore, the differences between treatments were reflected in these dimensions of task performance. The absence of significant differences for the self-rated performance measure may be attributed to several factors. For example, there was a lack of task feedback and of criteria against which participants could evaluate their task performance. A related factor is that the ambiguous nature of the task may have made it difficult for participants to judge how well they performed. The nonsignificant differences for the job related tension measure may be due to the fact that participants had difficulty in rating the frequency of work related tension; they experienced tension but were uncertain to what extent. Once again, the participants did not have a set of standards against which to judge their experience of tension. An alternative explanation is that the transitory nature of the task may have attenuated group differences in- work related tension. Over time in a work situation, organizational members could potentially more accurately assess the frequency of job stress. - 229 -Multivariate Analyses To supplement the univariate ANOVAs, a multivariate analysis of variance (MANOVA) was also conducted. Th i s multivariate technique takes into account correlations among dependent variables and compensates for increased T ype 1 errors associated with multiple univariate testing of the same data (e.g.. Green, 1978; Marascuilo & Levin, 1983; Tatsuoka, 1970). Therefore, MANOVA allows simultaneous testing of all the dependent variables and considers the various interrelationships among them. When significant MANOVA results are obtained, discriminant analysis can be subsequently applied in order to construct a linear combination of the set of dependent variables that will maximally differentiate among the groups in question (e.g.. Green, 1978; Marascuilo & Levin, 1983; Tatsuoka, 1970). By examining the relative standardized weights assigned to the different variables in the linear combination, how much (or little) each dependent variable contributed to the differentiation between groups can be determined. According to Tatsuoka (1970, p.4), inspection of the pattern of standardized weights gives a much more accurate account of the nature of group differences in terms of a given set of variables than does looking at each variable separately without regard for their interrelations and partly overlapping information. In the present study, standardized discriminant weights whose absolute values were no less than approximately one-half of the largest weights were interpreted (Tatsuoka, 1970, pp.3-4). There are two approaches to interpreting standardized discriminant weights assigned to each variable in computing the discriminant functions. - 230 -One approach uses the relative magnitude of the standardized discriminant weights as an index of the relative contribution or importance of the dependent variables to the discrimination between the groups, ignoring the positive or negative sign (Hair et a l . , 1979; Huberty, 1975; Klecka, 1980, pp.29-30). Dependent variables with relatively larger weights contribute more to the discriminating power of the function than do variables with smaller weights (e.g., Klecka, 1980; Pedhazur, 1982). Therefore, when the sign is ignored, each weight represents the relative contribution of its associated variable to that function (Hair et a l . , 1979, p.104; Nie, Hull, Jenkins, Steinbrenner, & Bent, 1975, p. 443). After identifying the variables which maximally discriminate between the groups on the basis of the absolute magnitude of the discriminant weights, the group means are used to aid in the interpretation of these variables (e.g., Aaker & Day, 1980; Tabachnick & Fidell, 1983). The other approach to interpreting standardized discriminant weights involves examining both the sign and the magnitude of the weight assigned to each variable in computing the discriminant functions (Tatsuoka, 1970). According to Tatsuoka (1970, p.4), a positive sign indicates the direction which is descriptive of the group having the higher mean standardized score on the linear combination of the set of variables. A negative sign indicates the direction which is descriptive of the group having the lower mean standardized score on the linear combination of the set of variables (Tatsuoka, 1970, p. 1*). With regard to the magnitude of the standardized weights, as discussed above, dependent variables with relatively larger weights contribute more to the discriminating power of the function than do variables with smaller weights (e.g., Kachigan, 1982; Pedhazur, 1982; Thorndike, 1978). - 231 -In summary, one approach to interpreting standardized discriminant weights entails examining the relative magnitude of the weights, irrespective of the sign, to identify the dependent variables which maximally differentiate between the groups and then uses the univariate group means to interpret the differences between the groups. The other approach examines both the relative magnitude of the weights and their direction ( i . e . , the positive or negative sign) to identify the dependent variables which maximally differentiate between the groups and the direction of such differences between the groups. In the present study, with the exception of one finding, the use of these two approaches to interpret the standardized discriminant weights yielded identical results. To give a specific example which will be discussed in the following section, the standardized discriminant function coefficients for the leadership style main effect indicated that adjustment to the leader (-.944) and role ambiguity (.493) maximally separated the considerate and structuring leadership styles on the fi r s t discriminant function (see Table 11). Using the fi r s t approach in which the magnitude of the weights are taken into account but the signs are ignored, the univariate group means are used to identify under which leadership style condition participants had higher adjustment to the leader and higher role ambiguity in comparison to the other leadership style condition. The group means reveal that individuals with considerate leaders have higher When the sign is taken into consideration, the results for the interaction effect indicated that individuals in the structuring-low productivity group recommended more courses of action than individuals in the s t r u c t u r i n g -high productivity group. This finding needs to be subject to further investigation. - 232 -adjustment to the leader (M = 26.77) than individuals with structuring leaders (M = 16.49) (see Table 7). In contrast, individuals with structuring leaders have higher role clarity (M = 28.77) than individuals with considerate leaders (M = 16.43) ( i . e . , the role ambiguity scale is reverse scored). In the second approach to interpreting standardized discriminant weights both the sign and magnitude of the weights are taken into account. The negative sign for the adjustment to the leader variable (-.944) is descriptive of the considerate leadership style which has the lower mean standardized discriminant function score (see Figure 4). The magnitude of the weight indicates that individuals with considerate leaders have higher adjustment to the leader than individuals with structuring leaders. T h i s interpretation using both the sign and magnitude of the weight is consistent with the interpretation using the univariate group means. The positive sign for the role ambiguity variable (.493) is descriptive of the structuring leadership style which has the higher mean standardized discriminant function score (see Figure 4). The magnitude of the weight indicates that individuals with structuring leaders have higher role clarity than individuals with considerate leaders. This interpretation is also supportive of the interpretation using the univariate group means. Therefore, both methods of interpreting the standardized discriminant weights produced the same results. To facilitate the interpretation of the following multivariate results, group means or cell means are cited. It should be underscored that standardized discriminant weights are not unambiguous indices of the relative importance of the variables with - 2 3 3 -w h i c h t h e y a r e a s s o c i a t e d . F o r e x a m p l e , a s m a l l w e i g h t m a y e i t h e r m e a n t h a t i t s c o r r e s p o n d i n g v a r i a b l e i s i r r e l e v a n t i n d e t e r m i n i n g a r e l a t i o n s h i p , o r t h a t i t h a s b e e n p a r t i a l l e d o u t o f t h e r e l a t i o n s h i p b e c a u s e o f a h i g h d e g r e e o f c o r r e l a t i o n w i t h o t h e r d e p e n d e n t v a r i a b l e s . I n a d d i t i o n , s t a n d a r d i z e d d i s c r i m i n a n t w e i g h t s l a c k s t a b i l i t y s i n c e t h e y a r e a f f e c t e d b y t h e v a r i a b i l i t y o f t h e v a r i a b l e s w i t h w h i c h t h e y a r e a s s o c i a t e d a n d b y t h e i n t e r c o r r e l a t i o