UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Cautiousness in adulthood and old age Holliday, Stephen George 1978

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

Item Metadata

Download

Media
831-UBC_1978_A8 H64.pdf [ 3.07MB ]
Metadata
JSON: 831-1.0094411.json
JSON-LD: 831-1.0094411-ld.json
RDF/XML (Pretty): 831-1.0094411-rdf.xml
RDF/JSON: 831-1.0094411-rdf.json
Turtle: 831-1.0094411-turtle.txt
N-Triples: 831-1.0094411-rdf-ntriples.txt
Original Record: 831-1.0094411-source.json
Full Text
831-1.0094411-fulltext.txt
Citation
831-1.0094411.ris

Full Text

CAUTIOUSNESS IN ADULTHOOD AND OLD AGE by STEPHEN GEORGE HOLLIDAY B.Sc. Northern Michigan Un i v e r s i t y , 1975 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n THE FACULTY OF GRADUATE STUDIES Department of Psychology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 1978 (c) Stephen George Holliday, 1978 In presenting t h i s thesis in p a r t i a l fulfilment of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library shall make it f r e e l y 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 representatives. It is understood that copying or publication of this thesis for f i n a n c i a l gain shall not be allowed without my written permission. Department of Psychology The University of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Orrober 2,7, 1978 i i ABSTRACT The present study examined the question of age r e l a t e d cautiousness i n decision making. T r a d i t i o n a l approaches to the study of age-rel a t e d patterns of choice behaviour and t h e o r e t i c a l explanations of cautiousness i n r i s k y choice s i t u a t i o n s were evaluated. An alternate approach to the study of choice behaviour i n d i f f e r e n t age groups, which emphasizes the r o l e of outcome d i r e c t i o n , p r o b a b i l i t i e s , and payoff values was presented. Peoples' preference for completely predictable and p r o b a b i l i s t i -c a l l y determined response options was assessed. The sample consisted of 12 men and 12 women i n 4 age groups representing young (20-30), middle (35-45) and l a t e (50-60) adulthood, and old age (65-75). The number of completely predictable options chosen by each sub-j e c t across (1) a l l s i t u a t i o n s , (2) gain s i t u a t i o n s only, and (3) loss s i t u a t i o n s only were calculated. An analysis of variance per-formed on the t o t a l number of completely predictable options chosen indicated that people of a l l age groups chose equivalent numbers of c e r t a i n options. There were no s i g n i f i c a n t e f f e c t s for age group, sex, nor the age x sex i n t e r a c t i o n . A repeated measures analysis of variance performed on subjects' scores i n gain and loss s i t u a t i o n s i n -dicated that people chose more completely predictable options i n gain than i n loss s i t u a t i o n s . The e f f e c t s of age group and sex were not s i g n i f i c a n t , nor were the two and three way i n t e r a c t i o n s . A second set of analyses were undertaken to r e l a t e subjects' choices to (1) the outcome d i r e c t i o n of the s i t u a t i o n s , and (2) the i i i expected payoff, or outcome, of the s i t u a t i o n s . Expected payoff was found to be a better predictor of choice behaviour than was outcome d i r e c t i o n . The implications of these findings f o r theories of age-related cautiousness are discussed. I t i s suggested that theories of adult choice behaviour may have to be modified to account for the impact of s i t u a t i o n a l variables on peoples' choice behaviour. i v TABLE OF CONTENTS Page Abstract i i L i s t of Tables v Acknowledgement v i Introduction 1 Method 33 Results 39 Discussion 48 References 55 Appendix A. Selection of the Age Groupings 58 Appendix B. Subject Selection Procedures 61 Appendix C. Questionnaire 64 LIST OF TABLES Socio-Demographic C h a r a c t e r i s t i c s Analysis of Variance Results for the Certainty Index Analysis of Variance Results for the Gain and Loss Scores Analysis of Variance Results for the Predictor Types Percent of Subjects Choosing Certain and Risky Options v i ACKNOWLEDGEMENTS I would l i k e to express my gratitude to Dr. G l o r i a Gutman, Dr. Robert Frender, and Dr. David Lawson for providing me with advice, assistance, and encouragement throughout t h i s project. Thanks are also due to Dr. Ralph Hakstian for advice on s t a t i s t i c a l techniques, to to Dr. Kenneth Prkachin and Timothy McTiernan for t h e i r thoughtful comments on the f i n a l manuscript. I would also l i k e to acknowledge the cooperation of the s t a f f and members of the S i l v e r Harbour A c t i v i t y Centre and Brock House who volunteered t h e i r time to f i l l out my questionnaires, and of Ms. Judy Hawkins who typed t h i s manuscript. F i n a l l y , my s p e c i a l thanks to:.my fiancee, Kathleen Sun, who not only put up with my i d i o s y n c r a t i c behaviour during the past year, but provided assistance at every stage of the project and helped me to achieve t h i s goal. 1 INTRODUCTION During the past two decades, the issue of age-related differences i n patterns of decision making has been the subject of controversy i n the gerontological l i t e r a t u r e (Okum, 1976a). The problems of i n t e r e s t have been whether e l d e r l y people exhibit d i f f e r e n t patterns of choice behaviour than young people i n ris k y s i t u a t i o n s , and whether such d i f -ferences, i f indeed they e x i s t , are i n d i c a t i v e of a general, age re-lat e d , propensity to avoid or minimize the r i s k factor i n choice s i t u a t i o n s . Researchers have been somewhat divided i n opinion on th i s matter. A number of researchers have maintained that differences do ex i s t , and that the e l d e r l y are either generally cautious i n r i s k s i t u a -tions (Wallach and Kogan, 1961; Rabbit, 1977), or averse to taking r i s k s i f other options are a v a i l a b l e to them (Botwinick, 1969, 1973). A l t e r n a t i v e l y , i t has been suggested that the 'cautious 1 behaviour of the e l d e r l y may have been an a r t i f a c t of t h e i r s e n s i t i v i t y to c e r t a i n parameters of r i s k decision s i t u a t i o n s . Although a large body of research dealing with these problems has developed, i t has yielded equivocal r e s u l t s . A number of studies have produced r e s u l t s suggesting that e l d e r l y people w i l l attempt to avoid r i s k y a l t e r n a t i v e s when p a r t i c i p a t i n g i n uncertain or r i s k y tasks, while younger people w i l l seldom exhibit such behaviour (Botwinick, 1966, 1969; Wallach and Kogan, 1961). Other studies have indicated that the performance of the e l d e r l y and the young i s essentially, iden-t i c a l i n c e r t a i n choice s i t u a t i o n s (Sanford, 1968; Griew, 1972; Okum, 1977). . At present the debate i s unsettled, and both sides are able 2 to c i t e reasonable arguments f or t h e i r p o s i t i o n s . For the most part, gerontological research has been conducted independently of studies examining formal models of decision making. At present, there i s a considerable body of l i t e r a t u r e dealing s p e c i f i -c a l l y with decision making under conditions of r i s k , some of which i s d i r e c t l y relevant to the gerontological debate. For example, r i s k seeking and r i s k avoiding are among the taking behaviours i d e n t i f i e d by researchers (Kahneman and Tversky, 1978). In addition, decision th e o r i s t s have offered several explanations of the conditions under which such behaviours can be expected to occur. Unfortunately, such information has not been integrated into the gerontological l i t e r a t u r e . The i s o l a t i o n of these two bodies of research i s apparent when one examines the gerontological l i t e r a t u r e . In t h i s l i t e r a t u r e there are few.specific references to decision making. There are, however, many references to studies examining decision making, some of which deal with decision making. But even i n studies dealing with choice behaviour there i s seldom more than a passing mention of t r a d i t i o n a l d e c i s i o n theory. While i t i s d i f f i c u l t to say why such a s p l i t e x i s t s , the most l i k e l y reason i s that the two areas are concerned with d i f f e r e n t problems. While decision t h e o r i s t s have been interested i n mathematical models of choice behaviour (Slovic and Lichenstein, 1967), the issue which gerontological researchers have focused on i s age re l a t e d changes i n cautiousness (Botwinick, 1973). The dec i s i o n making paradigm has been used more as a convenient technique f o r producing instances of cautious behaviour than as a means of studying the decision process per se. Most of the research on age rela t e d performance has resulted from t h i s 3 s p e c i a l focus of gerontological researchers. The subordination of the study of de c i s i o n making to the study of cautiousness i n the gerontological l i t e r a t u r e i s a d i r e c t r e s u l t of a series of h i s t o r i c a l developments. During the 1950's, several resear-chers (Basowitz and Korchin, 1957) reported that the e l d e r l y , as a group, exhibited a general reluctance to i n i t i a t e a ction when faced with ambiguous conditions. This reluctance to act was interpreted as cautiousness. The presumed cautiousness was then used as an explana-tory v a r i a b l e to account f o r age group performance differences on a var i e t y of perceptual, psychophysical, and i n t e l l i g e n c e t e s t i n g tasks (Okum, 1976a). As other researchers reported instances of cautious behaviour on the part of the e l d e r l y , a t r a d i t i o n of studying age group differences i n cautiousness emerged (Botwinick, 1973). Decision making became important i n the study of cautiousness more through accident that through i n t r i n s i c i n t e r e s t on the part of gerontological researchers. In 1961, Wallach and Kogan, while studying several aspects of decision making, reported age group performance d i f -ferences on a measure of d i s u t i l i t y of f a i l u r e to the i n d i v i d u a l (Wallach and Kogan, 1961). Their r e s u l t s suggested that e l d e r l y people required a higher l e v e l of cert a i n t y before entering a r i s k y s i t u a t i o n than did younger people. The researchers a t t r i b u t e d t h i s to a greater conservatism, or cautiousness, on the part of the e l d e r l y . Subse-quently, other investigators turned to the study of de c i s i o n making (Botwinick, 1966; Okum, 1976b), but unlike Wallach and Kogan, t h e i r primary i n t e r e s t was i n describing the appearance of cautiour behaviour i n e l d e r l y subjects. The decision making paradigm was merely a t o o l 4 to demonstrate age r e l a t e d changed i n cautiousness. As a d i r e c t r e s u l t of t h i s sequence of events, the study of d e c i -sion making i n the gerontological l i t e r a t u r e focused primarily on i d e n t i f y i n g the personal a t t r i b u t e s that could influence choice be-haviour i n r i s k y s i t u a t i o n s . In keeping with the t h e o r e t i c a l tenden-cies of the 1950's and early 1960's, de c i s i o n making was interpreted within a personality framework. Because of t h i s emphasis on persono-l o g i c a l v a r i a b l e s , l i t t l e a t tention was paid to the t r a d i t i o n a l aspects of the decision or choice paradigm. S i t u a t i o n a l parameters such as expected payoffs, outcome values, and outcome d i r e c t i o n s were deemed to be of minimal i n t e r e s t . Researchers also seem to have operated with vague, informal models of the decision making process, and seldom con-sulted the decision making l i t e r a t u r e when designing studies or i n t e r -preting r e s u l t s . The use of such an informal approach seems to have resulted i n a number of inadequacies i n the gerontological research. The l i t e r a t u r e on formal models of choice behaviour (Kahneman and Tversky, 1978; S l o v i c and Lichenstein, 1967) indicates that s i t u a t i o n a l parameters are very important i n determining choice behaviour, and that i t i s neces-sary to p r e c i s e l y c o n t r o l a l l parameters of a d e c i s i o n s i t u a t i o n , including p r o b a b i l i t i e s and payoffs. However, as noted previously, i n t e r e s t i n studying s i t u a t i o n a l parameters has been.visibly lacking i n the gerontological l i t e r a t u r e . Most studies have been very lax with respect to defining the component values of decision s i t u a t i o n s . Of the reported studies on decision making i n r i s k y s i t u a t i o n s , only two (Okum, 1977; B u r k h i l l and Schaie, 1976) analyzed the r e l a t i o n s h i p 5 between parameters of the r i s k taking s i t u a t i o n and people's behaviour. Since peoples' choice behaviour may r e f l e c t the contingencies and information present i n the decision s i t u a t i o n , experimental r e s u l t s may a c t u a l l y have been a function of uncontrolled s i t u a t i o n a l para-meters rather than a function of r i s k taking p r e d i l e c t i o n s on the part of the e l d e r l y subjects. In l i g h t of these considerations, i t seems imperative to begin to integrate the gerontological and formal decision making l i t e r a t u r e . The remainder of th i s s ection w i l l be devoted to that goal. Because the basic differences i n the two areas make i t d i f f i c u l t to simulta-neously comment on research development, representative studies from each area w i l l be presented and reviewed. Following t h i s review, the common features and areas of mutual i n t e r e s t w i l l be i d e n t i f i e d and discussed. The f i r s t study of dec i s i o n making to appear i n the gerontologi-c a l l i t e r a t u r e was a study by Wallach and Kogan (1961). These re-searchers examined the r o l e of four.variables i n dec i s i o n making: ex-tremity of judgement, subjective p r o b a b i l i t y of error i n judgement, confidence i n judgement, and d i s u t i l i t y of f a i l u r e . Subjects were men and women i n two age groups loosely categorized as young and e l d e r l y . The mean age of e l d e r l y subjects was approximately 70, while.the mean age of younger subjects was approximately 20. Since only the d i s u t i l i t y of f a i l u r e index yielded r e s u l t s r e l e -vant to th i s presentation, i t alone w i l l be discussed. To measure d i s u t i l i t y of f a i l u r e , Wallach and Kogan administered a 12-item choice dilemma instrument to each subject. Each item involved a c e n t r a l 6 character facing a d e c i s i o n of whether or not to take a r i s k y a c t i o n . If the character chose to take the a c t i o n , he/she might either gain or lose money, prestige, or comfort. If the character chose not to take the r i s k y action, the outcome would be completely predictable, i . e . , i t would lead to a continuation of the present s i t u a t i o n . Subjects were asked to i n d i c a t e on a scale ranging from 1/10 to 9/10 the p r o b a b i l i t y of success that they would require before advising the character to take the r i s k y action. If the subject chose to advise the person not to take the action, the response was scored as 10/10. This scoring procedure was based on the assumption that the decision not to take action lay on the same q u a l i t a t i v e continuum as the other choices. Each subject's responses were summed across the 12 items to y i e l d a d i s u t i l i t y of f a i l u r e index. According to Wallach and Kogan, i f the subject's score was high, he required a greater chance of success, and therefore there was a higher d i s u t i l i t y of f a i l u r e . They also suggested that higher scores were associated with conservatism or cau-tiousness . The main fin d i n g of t h i s segment of the study was that, on .the average, older people of both sexes required a higher p r o b a b i l i t y of success before advising someone to enter into a r i s k y s i t u a t i o n than did younger people. This was interpreted as evidence that the e l d e r l y are more cautious or conservative than younger people. Wallach and Kogan also reported that subjects responded d i f f e r e n -t i a l l y to a number of the items. They found that e l d e r l y people, for example, tended to be p a r t i c u l a r l y conservative i n s i t u a t i o n s involving 7 money. The researchers interpreted t h i s as suggesting that the con-f l i c t s being faced by each group influenced t h e i r choice behaviour. They then hypothesized that the d i f f e r e n t i a l response patterns observed i n the study were a function of the relevance of the s i t u a t i o n s to the subjects' present circumstances. Several years elapsed before Wallach and Kogan's work was extended by other gerontological researchers. The next report of cautiousness i n d e cision making was made by Botwinick i n 1966. Unlike Wallach and Kogan, who were interested i n decision making, Botwinick was interested primarily i n the issue of age re l a t e d cautiousness, and h i s work was aimed at c l a r i f y i n g the conditions under which cautious behaviour could be observed. Wallach and Kogan's twin findings of age s p e c i f i c d i f -ferences i n patterns of choice, and d i f f e r e n t i a l responding to item con-tent, led Botwinick to r e p l i c a t e the study using an expanded set of l i f e dilemma s i t u a t i o n s . While Wallach and Kogan had used items approp-r i a t e to younger age groups, such as changing jobs, attending univer-s i t y , and winning a f o o t b a l l game, Botwinick developed items centered on s i t u a t i o n s dealing with aspects of l a t e r adult l i f e , such as remar-rying, r e t i r i n g from work, and v i s i t i n g grandchildren. He then com-bined the new items with Wallach and Kogan's o r i g i n a l items to produce a 24 item l i f e dilemma questionnaire. Using the scoring procedures developed by Wallach and Kogan, Botwinick administered the expanded questionnaire to a group of young (mean age 21) subjects and to a group of e l d e r l y (mean age 75) sub-j e c t s . Both groups contained equal numbers of men and women. The r e s u l t s of t h i s study were s i m i l a r to those obtained by Wallach and 8 Kogan. Regardless of sex or educational l e v e l , older people required higher l e v e l s of assurance of success than younger people before ad-v i s i n g another person to enter into a r i s k y s i t u a t i o n . Botwinick also reported an i n t e r a c t i o n between age and item content, although not of the expected kind. He found that both old and young subjects required a higher l e v e l of assurance of success when the c e n t r a l character was young than when the character was e l d e r l y . The i m p l i c a t i o n of t h i s f i n d i n g was that "both old and young regarded the aged as having less to lose should the r i s k y course of a c t i o n prove i l l advised" (Okum, 1976a). As a secondary a n a l y s i s , Botwinick examined the d i s t r i b u t i o n of responses across choice options. This analysis indicated that the number of e l d e r l y adults choosing the no-action (completely certain) a l t e r n a t i v e was s i g n i f i c a n t l y higher than the number of younger adults doing so. Since the no-action a l t e r n a t i v e received the highest score, th i s phenomenon accounted for the observed differences i n o v e r a l l scores. As Botwinick's procedure was i d e n t i c a l to that of Wallach and Kogan, the i m p l i c a t i o n was that Wallach and Kogan's r e s u l t s were due to a s i m i l a r phenomenon. The presence of t h i s pattern led Botwinick to question the hypothesis that the no action a l t e r n a t i v e lay on the same q u a l i t a t i v e continuum as the other response options. He then went on to suggest that perhaps the e l d e r l y , rather than being cau-tious i n t h e i r general use of decision c r i t e r i o n , were d i s i n c l i n e d to enter into .any r i s k y s i t u a t i o n s regardless of the p r o b a b i l i t y of success, when a safe, non-risky, option was a v a i l a b l e to them. To test the hypothesis that the e l d e r l y ' s behaviour was co n t r o l l e d 9 by such a s p e c i f i c kind of r i s k aversion, Botwinick r e p l i c a t e d h i s study using the 24 item questionnaire, but eliminated the no-action a l t e r n a t i v e . He reasoned that i f there was a continuum of r i s k taking ranging from very cautious to very r i s k oriented, removing the no ac-ti o n choice would r e s u l t i n the el d e r l y people choosing the next safest a l t e r n a t i v e , which would be 9/10. A l t e r n a t i v e l y , i f the older subjects' behaviour was characterized by a very s p e c i f i c form of cautiousness, or r i s k aversion, t h i s pattern would not necessar i l y appear. In t h i s second study, Botwinick assessed the choice behaviour of el d e r l y women (mean age 76), e l d e r l y men (mean age 75), young men (mean age 20), and young women (mean age 21). Using the 24-item question-naire with the no action a l t e r n a t i v e eliminated, Botwinick found no differences between age groups i n choices of preferred l e v e l of pro-b a b i l i t y of success. Regardless of sex, both young and old subjects tended to choose mid-range p r o b a b i l i t y l e v e l s . The age by item i n t e r -a ction reported i n e a r l i e r studies also disappeared. However, Botwinick found that the e l d e r l y were s l i g h t l y less cautious than younger sub-j e c t s when the c e n t r a l character was younger. These findings l e d Botwinick to conclude that old people are i n c l i n e d to avoid r i s k only i s a no r i s k option i s a v a i l a b l e . Following Botwinick's lead, several other researchers used the l i f e s i t u a t i o n dilemma questionnaire to examine age r e l a t e d differences i n d ecision making. As with e a r l i e r studies, the r e s u l t s obtained seem dependent on the experimental procedures used. Tonberg (1970) r e p l i c a t e d Botwinick's second experiment using a 10 30-item l i f e dilemma questionnaire. The 30 items included the 24 used by Botwinick, and 6 new items constructed by Tonberg. This re-vised questionnaire was administered to young, middle-aged, and e l d e r l y adults. The mean age for the young group was 29. For the middle-ages group, the mean age was 41. The mean age for the e l d e r l y sub-j e c t s was 58. As i n the Botwinick study, Tonberg found no evidence of age differences i n l e v e l of assurance required by subjects i n the choice s i t u a t i o n s . Vroon and Pahl (1971), using a 5-item subset of the Wallach and Kogan items, and including the no-action choice, examined the d e c i -sion choices of 1,400 businessmen between the ages of 22 and 58. These researchers reported a small but s t a t i s t i c a l l y s i g n i f i c a n t c o r r e l a t i o n (r_ = .08, p_ = .01) between age and l e v e l of assurance required before advising entry into a r i s k y s i t u a t i o n . Taken i n combination, the data from these f i v e studies seem to provide support for Botwinick's notion of a s p e c i f i c kind of cautious-ness or r i s k aversion i n e l d e r l y people. Only i n studies which i n -cluded a no action option that was scored highest were age differences apparent. In each case, age differences seem to have been due to a disproportionate number of older people choosing the no-action a l t e r -native. On the other hand, i n studies where a r i s k y d e c i s i o n was unavoidable, age differences disappeared altogether. When i n t e r p r e t i n g these r e s u l t s , however, the l i m i t a t i o n s of the choice dilemma instrument must be considered. In a discussion of the flaws inherent i n the choice dilemma questionnaire, Okum (1976) men-tioned the following serious problems. (1) The instrument i s based 11 on hypothetical choices rather than on personal r i s k taking. (2) People's judgements may be made independently of the odds presented i n the s i t u a t i o n . (3) People's perceptions of what constitutes a high r i s k may d i f f e r . (4) The l i f e s i t u a t i o n s contained i r r e l e v a n t or i n s u f f i c i e n t information.. (5) The instrument i n s t r u c t i o n s are d i f f i -c u l t to follow. (6) The instrument i s semi-projective i n nature and the degree of personal involvement by the subjects cannot be c o n t r o l l e d . While such det a i l e d c r i t i c i s m s may seem to be a case of o v e r k i l l , they do serve to underscore a very general problem of the questionnaire. When the choice dilemma instrument was constructed, the s i t u a t i o n a l parameters for each item were not p r e c i s e l y s p e c i f i e d . Points 2, 3, .4, 5, and to some extent, point 6, a l l revolve around the same theme, i . e . , that there i s no way of knowing i f subjects' responses i n d i c a t e an aversion to r i s k , or a r e a c t i o n to some other aspect of the item. A strong argument could be put f o r t h that any one of the factors l i s t e d by Okum, or any combination of these, could have produced the r e s u l t s reported by Botwinick and other researchers. For example, e l d e r l y adults have been reported to be more s e n s i t i v e than younger people to the amount and q u a l i t y of information presented i n experi-mental s i t u a t i o n s (Labouviw-Vief and Chandler, 1978). As Okum (1976) has pointed out, i t i s d i f f i c u l t to d i s t i l l information from the choice dilemma s i t u a t i o n s . If the e l d e r l y were s e n s i t i v e to information con-tent, and reacted negatively to vague or non-specific information, the choice not to take a c t i o n could have r e f l e c t e d a d i s l i k e for s i t u a t i o n s i n which inadequate information was supplied, rather than a t r a i t of cautiousness or r i s k aversion. 12 D i s s a t i s f i e d with the l i m i t a t i o n s of the choice dilemma i n s t r u -ment, Okum attempted to examine choice behaviour within a behavioural framework. In two re l a t e d experiments, Okum and his associates examined r i s k taking i n a personal choice s i t u a t i o n . Using performance on a vocabulary test task as a dependent measure, they systematically mani-pulated several aspects of the r i s k s i t u a t i o n and measured the r e s u l t i n g choice behaviour. In the f i r s t experiment (Okum et a l . , 1976b), the r e l a t i o n between age and condition of i n s t r u c t i o n s i n r i s k taking s i t u -ations was examined. In the second study (Okum et a l . , 1978), the researchers looked at the impact of the payoff matrix associated with choices on people's r i s k taking behaviour. In the f i r s t experiment, the researchers assessed the r i s k taking behaviour of old (mean age 69) and young (mean age 21) males. The subjects were assessed for verbal p r o f i c i e n c y using a standard vocabu-la r y t e s t i n g task i n which they could win points for each correct answer that they gave. In t h i s segment of the experiment, subjects were required to choose the d i f f i c u l t y l e v e l of the items they would work with. They were also informed of the p r o b a b i l i t y of t h e i r an-swering an item of a p a r t i c u l a r d i f f i c u l t y l e v e l c o r r e c t l y . (Proba-b i l i t y estimates were based on each subject's pre-testing performance.) A f t e r each t r i a l , the subjects were given feedback on t h e i r perfor-mance. In addition, the subjects were exposed to supportive, chal-lenging, or neutral i n s t r u c t i o n s . Okum hypothesized that e l d e r l y people would consistently choose low d i f f i c u l t y items and.that they would also exhibit more r i s k taking behaviours under supportive conditions. Consistent with h i s 13 f i r s t hypothesis, he found that e l d e r l y adults tended to choose lower l e v e l s of d i f f i c u l t y than the younger subjects. His second hypothesis, however, was not supported as varying the ins t r u c t i o n s had no impact on the subjects' performances. Okum interpreted these findings as evidence for the notion of age re l a t e d increases i n cautious behaviour. As an explanation of his re-s u l t s , he proposed that the age group differences were a function of d i f f e r e n t l e v e l s of need for achievement i n the two groups. According to Okum, e l d e r l y poeple lowered t h e i r aspirations of success, p a r t l y as a consequence of f a i l u r e to meet l i f e goals, and consequently were less i n c l i n e d to seek out ris k y tasks. He also suggested that the reported fi n d i n g of r i s k aversion could be seen as attempts by the eld e r l y to avoid negative evaluation. In l a t e r research, Okum (1978) modified t h i s p o s i t i o n . He noted that i n the above study, the expected payoff, or number of points, that a subject could earn on any p a r t i c u l a r t r i a l was unrelated to the d i f f i c u l t y l e v e l of the items the person chose to work with. The ex-pected reward for answering the most d i f f i c u l t item was exactly the same as the expected reward for the easiest one. This suggested that the e l d e r l y people may have chosen the low d i f f i c u l t y items, not be-cause they were cautious, but because there was no reason for them to do otherwise. In other words, t h e i r strategy or pattern of choices may have been no less e f f e c t i v e than that of younger people i n terms of winning the maximum number of points. In f a c t , one can hypothesize that the e l d e r l y subjects may have been more e f f i c i e n t than the younger ones because the r a t i o df energy expended to points won may 14 have been lower. This finding, again raises the p o s s i b i l i t y that the e l d e r l y sub-j e c t s , rather, than being overly cautious or r i s k averse, may instead be very s e n s i t i v e to s i t u a t i o n a l parameters. Their 'reluctance' to take action i n the face of vague information can e a s i l y be interpreted as a sign of mature judgement, while the behaviour of younger subjects can be seen as a f o o l i s h wasteful expenditure of energy. To test the p o s s i b i l i t y that performance on r i s k taking tasks was a function of the payoffs associated with high and low d i f f i c u l t y items, Okum conducted a second experiment. E l d e r l y men (mean age 68) and young men (mean age 21) were assessed using the vocabularly t e s t i n g task described previously. In one experimental condition, the expected reward for a correct item increased with increasing item d i f f i c u l t y so that the most d i f f i c u l t items had the highest payoff. In a second condition, the expected payoff decreased with increasing d i f f i c u l t y l e v e l so-that the easiest items had the highest payoff. A l l subjects were exposed to each condition. When the r e s u l t s were examined, i t was found that a l l subjects per-formed equivalently i n both conditions, and that both groups s h i f t e d t h e i r performance c r i t e r i o n when the payoff conditions were changed. In the face of t h i s data, Okum interpreted the r e s u l t s as suggesting that the e l d e r l y w i l l avoid r i s k , but only under p a r t i c u l a r conditions. He suggested that r i s k avoiding behaviour would be observed i f the payoffs associated with r i s k y choices were inadequate to make the r i s k worthwhile. These conclusions are supported i n part by B u r k h i l l and.Schaie 15 (1975), These authors found that e l d e r l y subjects (mean age 73) took more r i s k s on an i n t e l l i g e n c e test i f there was a monetary incentive for doing so, and i f there was no penalty for f a i l u r e to perform unsuc-c e s s f u l l y . The researchers examined the i n t e l l i g e n c e test performances of eld e r l y males and females (mean age 73) under d i f f e r e n t reinforcement contingencies. The experimental conditions included s i t u a t i o n s i n which high and low r i s k taking was either discouraged or encouraged by making monetary rewards or losses contingent upon c e r t a i n performance patterns. They found that e l d e r l y subjects showed the highest rate of r i s k taking when correct responses were rewarded and f a i l u r e to respond was not penalized. When f a i l u r e was penalized, responding remained low, even when correct performance was rewarded. Although B u r k h i l l and Schaie interpreted these r e s u l t s as suppor-tin g Botwinick's p o s i t i o n that the e l d e r l y h e sitate to become involved i n a r i s k y s i t u a t i o n when the degree of r i s k i s high, they can also be seen as being consistent with Okum's conclusions (1978). The fi n d i n g that e l d e r l y people's performance i s determined by environmental con-tingencies f i t s very w e l l with Okum's report that the e l d e r l y are s e n s i t i v e to s i t u a t i o n a l parameters. Both reports i n d i c a t e that when faced with r i s k y d e c i s i o n s i t u a t i o n s , e l d e r l y adults may be extremely responsive to the p r o b a b i l i t i e s and outcomes associated with the res-ponse options a v a i l a b l e to them. The idea that e l d e r l y people are responsive to the p r o b a b i l i t y and outcome of a choice event has received c o l l a t e r a l support from studies examining the a b i l i t y of people to estimate the p r o b a b i l i t y 16 of events occurring. Sanford, Griew, and O'Donnel (1972) compared the a b i l i t y of old men and women (mean age 68.8) with that of young men and women (mean age 17.8) on two simple p r e d i c t i o n tasks. A l l subjects were asked f i r s t to predict the p r o b a b i l i t y of occurrence of two a l t e r n a t i n g events which occurred with biased p r o b a b i l i t i e s . The s p e c i f i c task was to predict the p r o b a b i l i t y of (a) a head and (b) a t a i l , when a rigged coin was f l i p p e d . They were then asked to per-form a s i m i l a r task i n a s i t u a t i o n with three d i s c r e t e events. The subjects were asked to predict the p r o b a b i l i t i e s of red, white, and blue b a l l s being drawn from an urn. The researchers reported that a l l subjects, regardless of age, quickly i d e n t i f i e d the p r o b a b i l i t y of each class of events occurring. They interpreted t h i s as evidence that the young and the el d e r l y are equivalent i n th e i r s e n s i t i v i t y to and a b i l i t y to detect p r o b a b i l i t y patterns. This f i n d i n g p a r a l l e l s an e a r l i e r report by Griew (1968) that both o l d and young subjects d i s t r i b u t e d viewing time while performing an observation task on the basis of the p r o b a b i l i t y that a s i g n a l would occur. Griew examined the performances of e l d e r l y (age range 62-71) and young (age range 18-23) subjects on a s i g n a l monitoring task. A l l subjects were asked to l i s t e n to a two channel tape recorder with s i g -nals occurring on each channel with d i f f e r e n t p r o b a b i l i t i e s . Although the two channels were presented simultaneously, only one channel could be l i s t e n e d to at a time. Subjects were then asked to detect as many signals on each channel as possible. This forced the subjects to apportion t h e i r l i s t e n i n g time between the two channels. Griew found that channel s e l e c t i o n of the e l d e r l y more c l o s e l y matched the s i g n a l 17 p r o b a b i l i t i e s per channel than did the channel s e l e c t i o n of the younger subjects, although both groups matched the s i g n a l p r o b a b i l i t i e s rea-sonably w e l l . This study i s of p a r t i c u l a r relevance i n that not only did subjects i d e n t i f y the p r o b a b i l i t y patterns, but they did not use the most conservative choice strategy of monitoring the high frequency s i g n a l . Instead they switched channels frequently, i n d i c a t i n g t h e i r willingness to take a r i s k on a lower p r o b a b i l i t y s i g n a l . While neither of these experiments deal s p e c i f i c a l l y with cautious-ness i n decision making, they are of some importance i n demonstrating that the e l d e r l y are not only capable of d i s t i l l i n g p r o b a b i l i s t i c information i n dec i s i o n s i t u a t i o n s , but that they can use that i n f o r -mation when performing choice tasks. The r e s u l t s of the f i v e studies examining r i s k taking i n behavioural s e t t i n g s , Griew's task i n p a r t i c u l a r , are somewhat at odds with the re s u l t s of studies using hypothetical l i f e dilemma s i t u a t i o n s . Both (1978) and B u r k h i l l and Schaie (1976) found that r i s k avoiding behaviour, even when a safe a l t e r n a t i v e was a v a i l a b l e , seemed to be a function of the payoffs and p r o b a b i l i t i e s of success associated with the response a l t e r n a t i v e s , and was not due to the use of a blanket strategy of r i s k avoiding. The Sanford (1972) and Griew (1968) findings that old people respond accurately to p r o b a b i l i s t i c information and can use thi s information i n performing v i g i l a n c e tasks i s also at odds with the notion that people use blanket strategies of avoiding r i s k i n choice s i t u a t i o n s . The el d e r l y subjects could have chosen simply to monitor continuously the high frequency s i g n a l channel i n order to avoid r i s k . In the s i n g l e behavioural r i s k taking study i n which 18 e l d e r l y people appeared to be more cautious than younger people (Okum, 1975), c e r t a i n aspects of the s i t u a t i o n appear to have i n -fluenced the choice behaviours. In summary, the choice dilemma questionnaires, the behavioural r i s k taking s i t u a t i o n s and the p r o b a b i l i t y matching studies have yie l d e d a complex and confusing pattern of r e s u l t s . In experimental tasks where r i s k has been unavoidable, or there has been a high payoff for taking a r i s k , or both, the de c i s i o n making behaviour of the young and the e l d e r l y has been equivalent. In s i t u a t i o n s where r i s k has been avoidable and payoffs for taking a r i s k y a c t i o n have been low or a s u b s t a n t i a l loss might be incurred,.the e l d e r l y have avoided taking r i s k s . This pattern was not evident i n the younger subjects studied. F i n a l l y , the e l d e r l y have been shown to be competent at d i s t i l l i n g p r o b a b i l i s t i c information from experimental s i t u a t i o n s and they are able to integrate p r o b a b i l i t i e s and payoffs into e f f i c i e n t choice strategies (Okum, 1978). The o r e t i c a l explanations of patterns of choice behaviour exhibited by e l d e r l y people have been offered by several researchers. Generally, t h e i r attempts have been less than p e r f e c t l y s a t i s f a c t o r y . Because the study of choice behaviour i n the e l d e r l y developed slowly over a long period of time, and because i t developed independently of the decision making l i t e r a t u r e , these explanations have tended to be ad hoc i n nature. The s h i f t i n g patterns of r e s u l t s i n the studies pre-viously mentioned have been mirrored i n the t h e o r e t i c a l explanations developed over the years. These have, as a r u l e , been l i m i t e d by the tendency to r e f l e c t the s p e c i a l biases of gerontological researchers 19 and have been able to explain only c e r t a i n subsets of the choice behaviour data. Since the f i r s t studies of age r e l a t e d changes i n decision making, gerontological researchers have put f o r t h the claim that old people are more cautious or r i s k averse than younger people, and explanatory attempts have consequently been aimed at e x p l i c a t i n g t h i s development. In addition, researchers have tended to assume a decrement model of human functioning and, i n developing theories, have focused p r i m a r i l y on declines i n f u n c t i o n a l a b i l i t i e s that, i n turn, could produce cau-tious behaviour. The most widely c i t e d formulation of the genesis of cautious or r i s k averse behaviour was developed by Botwinick (1969, 1973). He proposed that e l d e r l y people perceive themselves as being i n e f f i c i e n t or incompetent performers who are l i k e l y to f a i l when attempting to perform complex tasks. He then hypothesized that t h i s perception would cause people to avoid s i t u a t i o n s i n which they could f a i l to perform adequately. In choice s i t u a t i o n s , t h i s would lead them to choose non-r i s k y a l t e r n a t i v e s , as was reported i n choice dilemma studies. In Botwinick's system, performance patterns were seen as evidence of cautiousness or r i s k aversion. This, i n turn, was seen as the mani-f e s t a t i o n of a perception of personal inadequacy which became i n t e r -preted into a blanket strategy of r i s k avoidance. This explanation, which i s based p r i m a r i l y on data obtained from the choice dilemma studies, explicates those r e s u l t s quite w e l l . I t becomes, .less e f f e c t i v e , however, i n describing other f i n d i n g s . For example, the theory cannot account for the pattern of r e s u l t s reported 20 by Okum (1978) i n which the payoff associated with p a r t i c u l a r response options determined people's choice behaviour. Nor can i t e a s i l y ac-count for the findings of B u r k h i l l and Schaie (1976) suggesting that performance i s affected by s i t u a t i o n a l contingencies. Both of these findings i n d i c a t e that c e r t a i n aspects of choice s i t u a t i o n s a f f e c t choice patterns. Botwinick's theory cannot adequately deal with such r e s u l t s as i t does not make allowances for the r o l e of such variables i n choice s i t u a t i o n s . The theory also leaves several unanswered questions. While Botwinick postulated that^the e l d e r l y s u f f e r from f e e l i n g s of personal inadequacy, he does not demonstrate that people a c t u a l l y perceive themselves i n such a manner, or suggest how such a perception could develop. I t i s d i f f i c u l t to imagine why e l d e r l y people would per-ceive themselves as incompetent when, as Botwinick himself demonstrated, t h e i r performance i s equivalent to that of younger people. While i t i s possible to speculate that the e l d e r l y consistently ignore or mis-understand t h e i r performance, or that they reach a state of mind i n which a l l a c t i o n i s seen as p o t e n t i a l l y dangerous, i t seems unneces-sary to do so merely for the sake of t h e o r e t i c a l consistency. Rabbit (1977) has also expressed the opinion that e l d e r l y people behave cautiously because of performance d e f i c i e n c i e s . Echoing Botwinick's notion that decrements i n performance c a p a b i l i t i e s lead to a reluctance to be involved i n r i s k y s i t u a t i o n s , he noted that, "Per-haps i t (cautiousness) may also be seen as a tendency to choose a com-p l e t e l y predictable outcome because the u t i l i t y maximization strategy fo r the r i s k y a l t e r n a t i v e may be too d i f f i c u l t to work out." He then 21 suggested that cautious behaviour on the part of the e l d e r l y i s a function of information processing d e f i c i e n c i e s a r i s i n g from un-s p e c i f i e d , age rel a t e d , p h y s i o l o g i c a l changes. He also predicted that these d e f i c i e n c i e s would lead to performance differences between the old and young i n si t u a t i o n s where people are required to integrate p r o b a b i l i t y and payoff information into e f f e c t i v e behavioural s t r a -tegies . Again, t h i s theory explains some, but not a l l , of the a v a i l a b l e data. This formulation can adequately account for accurate performance of the el d e r l y on simple p r e d i c t i o n tasks, as these do not require v a high degree of information i n t e g r a t i o n . This would also be true of choice dilemma s i t u a t i o n s where subjects presumably avoid r i s k f or the reasons given previously. But the,theory cannot explain Okum's fi n d i n g that the behaviour of el d e r l y people s h i f t s as a function of payoff and p r o b a b i l i t y information, since such a s h i f t implies that the per-son i s e f f e c t i v e l y i n t e g r a t i n g several sources of information i n order to choose the best option. Rabbit's theory i s also incomplete i n that the s p e c i f i c types of information processing d e f i c i e n c i e s that are the supposed causes of cautious behaviour are not i d e n t i f i e d . Since there i s no i n d i c a t i o n that systematic loss of br a i n function occurs with increasing age, with the possible exception of people s u f f e r i n g from cardio-vascular disease, Rabbit's use of the concept of information processing d e f i -ciencies i s simply too vague and inaccurate to be of use. A less widely known explanation of performance patterns i n d e c i -sion studies was offered by Okum (1976b, 1978). He hypothesized that 22 age differences i n cautiousness were due to d i f f e r e n t i a l need achieve-ment l e v e l s i n the young and the e l d e r l y , with the e l d e r l y behaving l i k e low need achievers. He also suggested that the el d e r l y fear f a i l u r e , ostensibly as a r e s u l t of negative l i f e experiences, and t h i s fear of f a i l u r e leads them to protect themselves from engaging i n p o t e n t i a l l y negative evaluative s i t u a t i o n s . Although Okum (1977) subsequently modified t h i s stance as a r e s u l t of patterns observed i n his own data, he never completely abandoned the need for achievement/ fear of f a i l u r e model of r i s k taking behaviour. The same c r i t i c i s m s applied to Botwinick's and Rabbit's theories also apply to Okum's. The concept of d i f f e r e n t i a l need achievement motivation i s i n t e r e s t i n g but lacks empirical support. Okum's argu-ments would have been much more convincing i f he could have demon-strated that the young and e l d e r l y a c t u a l l y d i f f e r on some measure of need achievement. Okum's hypothesis that the el d e r l y may have ex-perienced more f a i l u r e i n l i f e s i t u a t i o n s than young people i s also i n t e r e s t i n g , but l i k e Botwinick's perceptions of inadequacy and Rabbit's notion of information processing d e f i c i e n c i e s , i t does not have empirical data for support. In general, these explanations of decision making or choice behaviour are inherently un s a t i s f y i n g . Perhaps the most undesirable aspects are that they deal with hypothetical t r a i t s rather than ob-servable events and they necessitate postulating a d d i t i o n a l hypothe-t i c a l states to support t h e i r existence. The p r o c l i v i t y of geronto-l o g i c a l researchers to engage i n such hypothetical speculation has had the unfortunate e f f e c t of d i s t r a c t i n g attention away from the basic 23 process of decision making. Researchers have s h i f t e d attention from the actual process of decision making to an i n v e s t i g a t i o n of person-a l i t y correlates of choice behaviour and the patterns of choices them-selves have been of l i t t l e i n t e r e s t . Another shortcoming of the t h e o r e t i c a l formulations i s that they have been unable to account for c e r t a i n configurations of data. They are generally capable of explaining cases of supposed r i s k avoidance, but are less successful i n dealing with cases of age group equiva-lencies on choice tasks. Thus, the data from the choice dilemma studies supports the model offered by Botwinick (1966) and Okum (1976b), but the data from the p r e d i c t i o n and several of the behavioural r i s k taking studies do not f i t at a l l w e l l . As was noted previously, the theories demand s t a b i l i t y of performance across s i t u a t i o n s , as they are postulating that people use general strategies when facing r i s k y s i t u a t i o n s . Consequently, they cannot cope with the occurrence of systematic change. Since change seems :to occur often i n the geron-t o l o g i c a l studies, the u t i l i t y of speculating that the e l d e r l y consis-te n t l y use any v a r i e t y of r i s k avoiding strategies seems highly ques-tionable. In short, the experiments and theories developed by researchers to explicate age r e l a t e d changes i n choice behaviour have not been successful. Almost 20 years ago, Wallach and Kogan questioned whether the young and e l d e r l y performed equivalently i n r i s k y d e c i s i o n s i t u a -t i o n s . Today, that question has not yet been adequately answered. The v a r i e t y of confusing, often contradictory, experimental data has sparked the controversy, but s t i l l has f a i l e d to c l a r i f y the questions 24 of i n t e r e s t . Theorists are s t i l l unable to agree on whether there i s a generalized or s p e c i f i c tendency on the part of the e l d e r l y to avoid r i s k , or whether people of a l l ages perform equivalently i n decision s i t u a t i o n s . F i n a l l y , the notion that age rel a t e d differences i n choice be-haviour e x i s t i s complicated by the fac t that a l l studies to date have been based on cross s e c t i o n a l designs i n which age and cohort e f f e c t s are hopelessly confounded. Even i f a c t u a l differences were noted, there would be no l o g i c a l necessity that those differences would be a function of age and not some other f a c t o r . In f a c t , the considerable v a r i a t i o n s i n h i s t o r i c a l and s o c i o - c u l t u r a l factors between the age groups suggests that other factors than age could have influenced the r e s u l t s . Given the general confusion surrounding the gerontological l i t e r a -ture on choice behaviour, i t seems imperative to systematically iden-t i f y the conditions that a f f e c t choice behaviour. The t r a d i t i o n a l gerontological paradigms for studying choice behaviour have suffered from several problems that have hindered t h i s study. But. despite these shortcomings, the r e s u l t s of studies examined i n t h i s review i n d i c a t e that one approach to the analysis of choice behaviour i s i n terms of the information contained i n the choice s i t u a t i o n s . Such an approach has been used with some success i n formal studies of decision making. Researchers interested .in formal models of dec i s i o n making have recognized the importance of parametric values i n determining per-formance i n decision s i t u a t i o n s . I t i s commonly accepted by decision t h e o r i s t s that choices i n r i s k y s i t u a t i o n s generally r e f l e c t the 25 subject's evaluation of the subjective worth or value of the a v a i l a b l e a l t e r n a t i v e s (Edwards, 1962). I t has been suggested that these evalu-ations r e f l e c t a procedure i n which the person weights the subjective worth of an outcome with i t s associated p r o b a b i l i t y . This weighting procedure y i e l d s a s i n g l e value describing the worth of the a l t e r n a -t i v e . These subjective value judgements have been demonstrated to c l o s e l y approximate the expected values generated using the a c t u a l outcome and p r o b a b i l i t y values i n a choice s i t u a t i o n (Pitz and Downing, 1967). I t has also been suggested that the person's goal i n d e c i s i o n s i t u a t i o n s i s to maximize gains and minimize losses. Decision making i n t o t a l , then, has been characterized as a process of evaluating a l t e r n a t i v e end states with the goal of f i n d i n g an optimal outcome. Several mathematical models have been offered as characteriza-tions of these evaluative procedures, and have been reported to be quite successful i n describing and p r e d i c t i n g choice behaviour (Edwards, 1962). The primary goal i n developing the models has been to describe the strategies that people use to optimize t h e i r performance i n decision s i t u a t i o n s . The models have tended to assume that people treat pro-b a b i l i s t i c information with a c e r t a i n mathematical exactitude, and that t h e i r choices i n decision s i t u a t i o n s are a d i r e c t function of the outcome values and p r o b a b i l i t i e s i n the choice s i t u a t i o n . Most models have also incorporated the notion that people maximize t h e i r gains and minimize losses using the two step procedure described above. For example, given a choice between a 50% chance of winning $60, and a 60% chance of winning $40, a person i s hypothesized to determine the value or u t i l i t y of each option using a weighting procedure to 26 integrate the p r o b a b i l i t y and outcome information. If the person were to use the information i n a mathematically precise manner, weighting the f i r s t outcome, $60, by. .its associated p r o b a b i l i t y y i e l d s .5 x $60, or $30. Weighting the second outcome, $40, by" i t s associated pro-b a b i l i t y y i e l d s .6 x $40, or $24. The $30 and $24 are mathematical representations of the expected worth of each a l t e r n a t i v e . Next, the person i s assumed to compare the expected outcome or u t i l i t y of the two a l t e r n a t i v e s and choose the one leading to the most desirable end state. The minimization/maximization p r i n c i p l e suggests that people w i l l choose the option with the greatest value, and since the expected outcome of the f i r s t a l t e r n a t i v e s leads to higher expected gains, the person would be expected to choose that a l t e r n a t i v e . S i m i l a r l y , i f a person were faced with a choice dealing with losses, most models of d e c i s i o n making would predict that the most a t t r a c t i v e response option would be the one leading to the smallest expected l o s s . The hypothesized use of a minimization/maximization p r i n c i p l e i n combination with the mathematical weighting procedure implies that a s p e c i a l kind of assymetry exists i n r i s k y choice s i t u a t i o n s , with choice behaviour i n gain and loss s i t u a t i o n s systematically s h i f t i n g as a function of s i t u a t i o n a l values. If people were placed i n a set of gain and loss s i t u a t i o n s i n which the a l t e r n a t i v e s had equal p r o b a b i l i t i e s , equal outcomes and reversed outcome d i r e c t i o n s , the minimization/maximization p r i n c i p l e suggests that peoples' choice behaviour should reverse i t s e l f as the s i t u a t i o n s change from gains to losses or v i c e versa. This should occur because when the outcome 27 of a choice s i t u a t i o n i s reversed from gains to losses, a l t e r n a t i v e s that l ed to maximal gains i n gain s i t u a t i o n s should lead to maximal losses i n loss s i t u a t i o n s . Conversely, a l t e r n a t i v e s that led to maxi-mal losses would lead to maximal gains. For example, i n a choice between a sure gain of $5 and a 40% chance of winning $10, the $5 should be more a t t r a c t i v e as i t leads to the highest expected gain. I f , however, the d i r e c t i o n of the outcome were changed so that the choice was between l o s i n g $5 and a 40% chance of l o s i n g $10, the r i s k y a l t e r n a t i v e should be the most a t t r a c t i v e as i t leads to the lowest expected losses. In general, then, changing the d i r e c t i o n of a choice s i t u a t i o n while leaving the other parameters unaffected should bring about predictable changes i n behaviour, i f people are using minimization/maximization s t r a t e g i e s . Recently, however, researchers have reported that while such s h i f t s do occur, choice patterns deviate from the exact models i n cer-t a i n cases. Kahneman and Tversky (1978) found that people exhibited what they termed r i s k seeking and r i s k avoiding behaviour that was un-r e l a t e d to the expected payoffs calculated i n c e r t a i n choice s i t u a t i o n s . They found that, within c e r t a i n l i m i t s , and regardless of the theore-t i c a l expected payoff or outcome of choice options, people tended to avoid r i s k y options when given a choice between a c e r t a i n and a r i s k y gain, and to seek r i s k options when given a choice between a c e r t a i n and a r i s k y l o s s . According to the researchers, t h i s occurred because people assigned an u n r e a l i s t i c a l l y high value to the c e r t a i n option i n gain s i t u a t i o n s , and an u n r e a l i s t i c a l l y low value to the c e r t a i n option i n the loss s i t u a t i o n s . 28 This i s best seen i n a s p e c i f i c context. Given a choice between a sure gain of $5 and a 60% chance of gaining $10, the expected pay-of f i s $5 for the c e r t a i n option and $6 for the r i s k y choice. On the basis of expected values, then, one might predict that the second a l t e r n a t i v e would be the most valuable as i t leads to the highest ex^ -pected gain. Kahneman and Tversky found that across a v a r i e t y of these s i t u a t i o n s , people consistently chose the safe a l t e r n a t i v e , des-p i t e the f a c t that the r i s k y a l t e r n a t i v e had the highest expected pay^ o f f . Because people seemed to prefer the c e r t a i n to the r i s k y option, they termed t h i s pattern one of r i s k avoiding behaviour. Conversely, they found that when the people were given a choice between a c e r t a i n loss and a r i s k y l o s s , such as between l o s i n g $5 and a 60% chance of los i n g $10, people consistently chose the r i s k y option despite the fact that i t was mathematically expected to y i e l d a greater l o s s . Because people seemed to prefer r i s k y to c e r t a i n losses, they l a b e l l e d t h i s pattern of behaviour r i s k seeking. I t i s beyond the scope of th i s paper to deal with the complexi-t i e s of the formal models of decision making i n an evaluative context. Therefore, the r e l a t i v e merits of the p a r t i c u l a r formulations d e s c r i -bing people's choice behaviour w i l l not be discussed. The l i t e r a t u r e seems to be i n consensus, however, on the point that within c e r t a i n l i m i t a t i o n s , peoples' choice behaviour should exhibit systematic s h i f t s i n gain and loss s i t u a t i o n s . I t i s t h i s hypothesized s h i f t i n g of choice as a function of outcome d i r e c t i o n that should be of p a r t i -cular i n t e r e s t to gerontological researchers. There are two reasons why s h i f t s i n choice behaviour are of 29 p a r t i c u l a r relevance to the gerontological l i t e r a t u r e . F i r s t , the p o s s i b i l i t y that performance changes as a function of outcome d i r e c -t i o n suggests that a thorough analysis of choice behaviour of the el d e r l y under conditions of r i s k must include s i t u a t i o n s i n v o l v i n g gains as well as s i t u a t i o n s involving losses. In the gerontological l i t e r a t u r e , a ttention has been focused p r i m a r i l y on gain s i t u a t i o n s (Okum, 1976b, 1978) or on si t u a t i o n s involving complicated combinations of gains and losses (Wallach and Kogan, 1961). Second, the observation of systematic performance changes on the part of young (univ e r s i t y students) and middle aged (university f a c u l t y members) that were described as r i s k avoiding behaviour, suggests that the use of such strategies i s not simply a consequence of the aging process, but i s probably re l a t e d to d i f f e r e n t aspects of the decision s i t u a t i o n . Studying the behaviour of people i n d i f f e r e n t age groups under conditions that have been reported to stimulate behaviour s h i f t s could enhance our understanding of age group performance patterns. These p o s s i b i l i t i e s have not been se r i o u s l y considered i n the gerontological l i t e r a t u r e , although they are d i r e c t l y r e l a t e d to sug-gestions by both Botwinick and Rabbit that the e l d e r l y prefer completely predictable over r i s k y outcomes.. Although neither of those authors e x p l i c i t l y s tales that s i t u a t i o n a l information and outcome d i r e c t i o n are i r r e l e v a n t factors i n choice behaviour, they neglect the importance of these v a r i a b l e s . Instead, they focus on personality and physio-l o g i c a l variables that are hypothesized to a f f e c t choice behaviour. However, since t h e i r t h e o r e t i c a l formulations have implied behavioural s t a b i l i t y across choice s i t u a t i o n s , the occurrence of behaviour s h i f t s , 30 such as those predicted by decision making theories, would be both unexpected and d i f f i c u l t to handle. Since these two bodies of l i t e r a t u r e suggest that people treat s i m i l a r choice s i t u a t i o n s i n quite d i f f e r e n t ways, several issues of age r e l a t e d decision making might be answered i f the approaches could be intergrated. If i t could be demonstrated that people of d i f f e r e n t ages choose safe or r i s k y a l t e r n a t i v e s because of the outcome d i r e c -t i o n , i t would be d i f f i c u l t to maintain that the e l d e r l y use blanket r i s k avoiding s t r a t e g i e s . On the other hand, i f i t could be shown that e l d e r l y people are not s e n s i t i v e to s h i f t s i n outcome d i r e c t i o n , or other s i t u a t i o n a l values, t h i s could be seen as support for the p o s i t i o n that e l d e r l y peoples' decision strategies are independent of s i t u a t i o n a l f a c t o r s . As has been demonstrated, the gerontological and d e c i s i o n making l i t e r a t u r e s lead to d i f f e r e n t conclusions regarding the nature of per-formance patterns i n r i s k y decision s i t u a t i o n s . D i f f e r e n t i a l predic-tions regarding choice behaviour under conditions of r i s k can be generated from each p o s i t i o n . The overlap of the two f i e l d s suggests the p o s s i b i l i t y of c l a r i f y i n g the questions of age r e l a t e d differences i n decision making through an i n t e g r a t i o n of d e c i s i o n making and gerontological research s t r a t e g i e s . The present study attempted to make that i n t e g r a t i o n i n an exami-nation of the choice behaviour of people i n four age groups (20-30, 35-45, 50-60, 65-75) across gain and loss s i t u a t i o n s involving choices between c e r t a i n and r i s k y options. I t has been suggested that i n these types of s i t u a t i o n s elderly.,, but not younger people, exhibit 31 cautious or r i s k avoiding behaviour (Wallach and Kogan, 1961; Botwinick, 1973). To examine the p o s i t i o n that preference for c e r t a i n options i s a function of age group membership, several aspects of decision making were examined, including preference for completely predictable outcomes and s i t u a t i o n a l s t a b i l i t y of choice behaviour. The i n v e s t i g a t i o n focused on people's preferences i n s i t u a t i o n s involving choices between completely predictable and p r o b a b i l i s t i c a l l y determined gains, and between completely predictable and p r o b a b i l i s -t i c a l l y determined losses. Subjects were presented with a ser i e s of twelve s i t u a t i o n s l o g i c a l l y s i m i l a r to those used by Botwinick (1966, 1969) i n that subjects were presented with a choice between a r i s k y and a c e r t a i n option. However, i n t h i s study, a l l parametric values including p r o b a b i l i t i e s and payoffs and outcome d i r e c t i o n were s p e c i f i e d . In addition, these values were varied across choice situations, to per-mit an analysis of the impact of s i t u a t i o n a l information on choice behaviour. I t was predicted that preference for completely predictable options, as opposed to r i s k y options, would be equivalent for people i n each of the four age groups. I t was further predicted that peoples' preference for c e r t a i n and p r o b a b i l i s t i c a l l y determined options would be r e l a t e d to the outcome d i r e c t i o n of the choice s i t u a t i o n s . The design of the study also permitted an examination of the manner i n which subjects treated information presented i n the choice s i t u a t i o n s . The three pertinent factors i n the s i t u a t i o n s were the p r o b a b i l i t y of occurrence of the event, the payoff associated with the event, and the outcome d i r e c t i o n of the s i t u a t i o n . The analysis 32 focused on whether subjects 1 ..choices i n the s i t u a t i o n s could be seen as a simple function of the outcome d i r e c t i o n , or, a l t e r n a t i v e l y , as a j o i n t function of the p r o b a b i l i t i e s and payoffs associated with the response options. This was done as an exploratory a n a l y s i s , and no hypotheses were offered as to which of the a l t e r n a t i v e s would best describe the subjects' behaviour. In summary, two aspects of r i s k y d ecision making were examined i n a c r o s s - s e c t i o n a l , age group design: preference for completely predictable options and s i t u a t i o n a l s t a b i l i t y of choice behaviour. In addition, subjects' responsiveness to several parameters of choice s i t u a t i o n s , including p r o b a b i l i t i e s , payoffs, and outcome d i r e c t i o n was examined. I t was hypothesized that (a) across a l l age groups, peoples' preferences for completely predictable options would be s i m i l a r , and (b) peoples' preference for r i s k y and completely predictable options would vary as a function of the outcome d i r e c t i o n of the choice s i t u a -t i o n . 33 METHOD Subj ects Subjects were 12 men and 12 women i n each of four age groups re-presenting young adulthood (age range 20-30), middle adulthood (age range 35-45), l a t e adulthood (age range 50-60) and old age (age range 65-75). Subjects i n the young, middle, and l a t e adulthood groups were rec r u i t e d at Langara Community College, the University of B r i t i s h Columbia Summer School, and the Vancouver Public L i b r a r y . Subjects i n the old age groups were r e c r u i t e d at the University of B r i t i s h Columbia Centre f or Continuing Education, Brock House Senior C i t i z e n s Centre, and the S i l v e r Harbour A c t i v i t y Centre. Detailed descriptions regarding the choices of age groupings and the subject sampling procedures are presented i n Appendices A and B. Table 1 shows the socio-demographic c h a r a c t e r i s t i c s of subjects within each age group. Examination of t h i s table indicates that across a l l age groups the average l e v e l of educational achievement was high, with most subjects reporting at least some u n i v e r s i t y education. The educational l e v e l s of the young, middle, and l a t e adults were s i m i l a r , while the e l d e r l y adults received somewhat less education than people i n the other groups. Analysis of variance procedures indicated that there were s i g n i f i c a n t differences between the groups, with the el d e r l y d i f f e r i n g from the other groups, I?(3,88) = 4.61, p_ = .002. Across a l l age groups, the majority of subjects (56%) were married. F i f t e e n percent of the subjects were s i n g l e and 12% separated. Table 1 Socio-Demographic C h a r a c t e r i s t i c s Young Adults Middle Adults Late Adults E l d e r l y Adults Variable M M M M Age ( i n years) Mean S.D. 27.3 2.3 27.7 2.2 38.9 3.4 39.3 3.1 55.5 3.4 55.7 5.1 69.8 3.5 70.4 3.3 Education Mean S.D. 15.7 1.6 15.9 1.7 16.3 2.1 15.4 1.4 15.8 2.5 15.2 1.9 13.3 3.7 13.7 2.2 M a r i t a l Status Married Common Law Divorced/ Separated Widowed Never Married 9 0 1 0 2 4 2 3 0 3 7 1 1 0 3 9 0 1 0 1 9 0 1 0 1 4 0 4 2 0 9 0 2 0 1 3 0 0 6 3 Employ. Status Employed Unemployed In t r a i n i n g Retired Homemaker Unknown 1 3 0 0 0 9 1 2 0 0 0 10 1 1 0 0 0 11 0 0 0 0 0 9 1 0 1 0 1 1 0 0 11 0 0 4 0 0 8 0 0 35 Three percent were divorced and 3% were l i v i n g i n a common-law arrange-ment. As might be expected, a considerable number of e l d e r l y women were widowed (50%). Three percent of the subjects did not report t h e i r m a r i t a l status. Across the three younger age groups, both males and females tended to be employed (56%), while i n the old age group the majority of the people (79%) were r e t i r e d . A small number of people (7%) were i n t r a i n i n g for employment. Five percent of the sample reported being unemployed, 3% l i s t e d themselves as being homemakers, and 2% did not report t h e i r employment status. Apparatus Instrument Development. The r i s k taking instrument was based on a questionnaire developed by the author. In i t s i n i t i a l form, the questionnaire contained 20 two-alternative choice items. The outcome of one a l t e r n a t i v e course of action was completely predictable, while the outcome of the other occurred with some s p e c i f i e d p r o b a b i l i t y . The 20 items were divided into 10 p a i r s , each pair consisting of a s i t u a t i o n i n v o l v i n g a p o t e n t i a l l o s s , and a corresponding s i t u a t i o n involving a p o t e n t i a l gain. For the two members of an item p a i r , the p r o b a b i l i t y and outcome values were equivalent, while the outcome d i r e c t i o n was reversed. I f the p_(winning) i n a gain s i t u a t i o n was .5 and the amount to be won was $50, the p_(losing) i n the corresponding loss s i t u a t i o n was .5 and the amount to be l o s t $50. The parametric values of the item sets were c o n t r o l l e d with pro-b a b i l i t i e s ranging from .5 to 1.0, and amounts ranging from $6 to $10,000. The p r o b a b i l i t i e s and outcome values were balanced across 36 s i t u a t i o n s , so that i n one hal f of the s i t u a t i o n s , regardless of out-come d i r e c t i o n , the expected outcome of the c e r t a i n a l t e r n a t i v e was highest, while i n the remaining h a l f the expected outcome of the r i s k y a l t e r n a t i v e was highest. In scoring the questionnaire, the completely predictable a l t e r -native was assigned the value of 1 and the r i s k y a l t e r n a t i v e the value of 2. Subjects' scores were cumulated for the gain and loss s i t u a t i o n s , y i e l d i n g two scores, one summarizing behaviour i n gain s i t u a t i o n s , the other describing behaviour i n loss s i t u a t i o n s . The 20-item instrument was pretested on a sample of 55 u n i v e r s i t y undergraduate students ranging i n age from 18 to 45. Risk seeking scores were computed for both the gain and loss s i t u a t i o n s . This yielded two subscales focusing on gain and loss s i t u a t i o n s . R e l i a b i l i t y -checks involving d i f f e r e n t combinations of items were performed for each subscale using the Spearman-Brown s p l i t half r e l i a b i l i t y c o e f f i -c i e n t as a target. Items that contributed l i t t l e to, or detracted from the value of the c o e f f i c i e n t were discarded. The 6 item pa i r s y i e l d i n g the best scale c h a r a c t e r i s t i c s were retained and used i n the f i n a l instrument. R e l i a b i l i t y Information. Since the f i n a l questionnaire assessed choice behaviour i n two d i s t i n c t types of s i t u a t i o n s , Spearman-Brown s p l i t h a l f r e l i a b i l i t y c o e f f i c i e n t s were computed separately f o r each subscale. A l l c o e f f i c i e n t s were based on pre-test data. The r e l i -a b i l i t y c o e f f i c i e n t for the gain subscale was .64 and the c o e f f i c i e n t for the loss subscale was .58. Because the length of a test strongly influences the value of the r e l i a b i l i t y c o e f f i c i e n t , and because the 37 scales were of limited, length, a v a r i a t i o n of the Spearman-Brown c o e f f i c i e n t was used to estimate scale r e l i a b i l i t y independently of scale length. As recommended by Ferguson (1971), r e l i a b i l i t y co-e f f i c i e n t s were calculated for tests with 100 items. The r e l i a b i l i t y estimates obtained with this procedure were .97 for the gain items, and .94 for the loss items. The estimates obtained with t h i s pro-cedure indicated that the i n i t i a l moderate estimated were due to te s t length, rather than to inconsistencies i n the questions. The F i n a l Instrument. The 6 items pairs contained i n the f i n a l instrument are l i s t e d i n Appendix C. The 12 s i t u a t i o n s were designed to be analogues of r e a l l i f e economic dilemmas. E f f o r t s were made to make the items as r e a l i s t i c as possible, with item topics such as moving to a new neighbourhood, buying a car, and investing money. Each s i t u a t i o n has two p o t e n t i a l outcomes, one completely pre d i c t a b l e , the other associated with some s p e c i f i e d degree of r i s k . For each s i t u a -t i o n , subjects were asked to decide which a l t e r n a t i v e was most approp-r i a t e to the s i t u a t i o n . Separate scores were calculated f o r the gain and loss s i t u a t i o n s . A t h i r d score based on the number of completely predictable options chosen was also computed. Procedure A l l subjects were approached and asked to p a r t i c i p a t e i n a study of d ecision making. If they expressed willingness to p a r t i c i p a t e , they were given a 12 item forced choice questionnaire containing 6 gain and 6 loss s i t u a t i o n s . The ..experimenter b r i e f l y explained the instrument and the i n s t r u c t i o n s f o r recording choices. I t was empha-sized that the questionnaire was not a te s t , that there were neither 38 ri g h t nor wrong answers, and that the experimenter was only interested i n the choice patterns of the people being tested. A f t e r completing the t e s t , the subjects were informed i n more d e t a i l of the hypotheses being tested. A l l interested subjects were sent a b r i e f d e s c r i p t i o n of the findings and the interp r e t a t i o n s of the r e s u l t s . 39 RESULTS In the f i r s t stage of the ana l y s i s , three indices of r i s k taking were computed. The f i r s t , preference for c e r t a i n t y , was based on the t o t a l number of completely predictable options chosen by each subject. Completely predictable options were scored as 1 and summed across s i t u a t i o n s y i e l d i n g an index with a value range of 0-12, with higher scores associated with a stronger preference for c e r t a i n t y . The second and t h i r d indices measured peoples' choice preferences i n gain and loss s i t u a t i o n s . Choices of the completely predictable option were scored as 1 and of the r i s k y option as 2. Scores were summed over each type of outcome y i e l d i n g two indices with value ranges of 6-12. Thus low scores were associated with preference for c e r t a i n options and high scores with preference for r i s k y options. The measures were analyzed using analysis of variance procedures as described i n the S t a t i s t i c a l Package for the So c i a l Sciences. A l l measures were analyzed using a group by sex design. T h e o r e t i c a l l y , there was l i t t l e reason to suspect that sex differences existed. Nevertheless, common stereotypes often depict women as i n f e r i o r i n deci s i o n s i t u a t i o n s . To determine i f performance differences existed between men and women, sex was included as a design f a c t o r . To test the hypothesis that people i n older and younger age groups have s i m i l a r preferences for completely predictable options, a 4x2 (age x sex) analysis of variance was performed, using the cer-t a i n t y index as the dependent measure. The r e s u l t s of t h i s analysis are shown i n Table 2. 40 Table 2 Analysis of Variance Results for the Certainty Index Source df SS MS F Age 3 22..59. 7.53 1.82 .15 Sex 1 0.95 0.85 0.23 .63 Age x Sex 3 3.09 1.03 0.25 .86 Error 88 363.71 4.13 T o t a l 95 390.34 41 This analysis revealed no s t a t i s t i c a l l y s i g n i f i c a n t e f f e c t s f o r eith e r age, sex, or the age by sex i n t e r a c t i o n . Thus, the hypothesis that there are differences i n members of d i f f e r e n t age groups' pre-ferences for c e r t a i n options was not supported. Examination of the group means indicated that the e l d e r l y group scored s l i g h t l y higher than the other groups. In addition, the mean values tended to increase with increasing age. The middle adults chose the fewest completely predictable options (6.04), followed by the young adults (6.54), the l a t e adults (6.84), and the e l d e r l y adults (7.39). As noted e a r l i e r , however, these differences are not s t a t i s -t i c a l l y s i g n i f i c a n t . The indices of r i s k taking i n gain and loss s i t u a t i o n s were analyzed using a 4x2x2 (age group by sex by outcome type) analysis of variance, with repeated measures on the f i n a l f actor of outcome type. This was done to test the hypothesis that people's choices are affected by outcome type. The r e s u l t s of t h i s analysis are shown i n Table 3. The analysis revealed a s t a t i s t i c a l l y s i g n i f i c a n t effect, for outcome type, supporting the hypothesis that outcome type influences choice behaviour. There were no s t a t i s t i c a l l y s i g n i f i c a n t e f f e c t s f o r age groups or sex. Neither the two nor the three way in t e r a c t i o n s were s t a t i s t i c a l l y s i g n i f i c a n t . Examination of the mean values of the indices indicated that the s i g n i f i c a n t e f f e c t of outcome type was due to subjects choosing more ri s k y options i n loss s i t u a t i o n s than i n gain s i t u a t i o n s . In loss s i t u a t i o n s , people chose on the average 3 r i s k y and 3 c e r t a i n a l t e r n a -t i v e s . In gain s i t u a t i o n s , subjects chose, on the average 4 c e r t a i n 42 Table 3 Analysis of Variance Results for the Gain and Loss Scores Source df SS MS Group 3 11.60 3.87 1.87 Sex 1 0.48 0.48 0.23 Group x Sex 3 1.55 0.52 0.25 Subjects (Groups) 88 181.86 2.07 Outcome 1 24.08 24.08 14.78 Group x Outcome 3 4.45 1.48 0.91 Sex x Outcome 1 0.52 0.52 0.32 Sex x Group x Outcome 3 3.18 1.06 0.65 Outcome x Subjects (Groups) 88 143.41 1.63 ,14 .63 .86 ,001 .44 .57 .58 43 and 2 r i s k y options. This pattern was very consistent across the age groups. A second set of analyses were conducted to examine whether people's choices were a simple function of outcome d i r e c t i o n or a j o i n t function of p r o b a b i l i t y and outcome values. To achieve t h i s end, two indices were constructed. The f i r s t index represented agreement between subjects' choices and predictions generated based on the outcome d i r e c -t i o n of the choice problems. From th i s index, subjects' responses for each item were assigned the value 1 i f they chose either a completely predictable option i n a gain s i t u a t i o n or a r i s k y option i n a loss s i t u a t i o n . The second index represented agreement between subjects' choices and predictions generated based on the p r o b a b i l i t y and payoff values of each s i t u a t i o n . Subjects' scores were assigned a value of 1 i f they chose the option leading to either the highest expected gain or the lowest expected l o s s . For each index, scores were summed across the 12 si t u a t i o n s y i e l d i n g two scores with value ranges of 1-12. To determine i f the number of correct predictions generated by the two sets of c r i t e r i o n d i f f e r e d , a 4x2 (age group by predictor type) repeated measures analysis of variance was conducted on the two above mentioned indices. The r e s u l t s of t h i s analysis are presented i n Table 4. The analysis revealed a s i g n i f i c a n t e f f e c t for predictor type. Neither the group e f f e c t nor the group by predictor type i n t e r a c t i o n was s i g n i f i c a n t . Examination of the means indicated that more cor-rect predictions were generated using both p r o b a b i l i t y and payoff 44 Table 4 Analysis of Variance Results for Predictor Types Source df MS F Age 3 4.03. 1.34 0.67 .57 Subjects (Age) 92 184.80 2.01 Type 1 8.08 8.08 7.97 .006 Age x Type 3 5.34 1.74 1.72 .17 Type x Subj. (Age) 92 93.95 1.01 45 information than were generated using outcome d i r e c t i o n alone. Although there was a s t a t i s t i c a l l y s i g n i f i c a n t e f f e c t f o r pre-d i c t o r type, the absolute number of matching predictions was not par-t i c u l a r l y high f o r either model friean values of 7.5 and 6.3 for the probability/payoff and outcome predictions r e s p e c t i v e l y ) . To determine i f the mean number of correct predictions d i f f e r e d from values expected due to random f l u c t u a t i o n , the mean values were converted to propor-tions and compared to a random expectation of .50 correct p r e d i c t i o n s . Before conducting t h i s a n a l y s i s , however, Chi Square analyses were conducted to determine i f the age groups treated the i n d i v i d u a l items d i f f e r e n t l y . . It.was found that, with one exception, the age groups responded i n l i k e manner to each s i t u a t i o n . Because of the s i m i l a r i t y across groups, i t was decided to collapse the four groups into one. The following analyses are based on the response patterns of the en t i r e sample. J:-tests for the differe n c e between two proportions were then con-ducted for the two indices of matching predictions, using procedures described i n Glass and Stanley (1971). The proportion of matches generated using p r o b a b i l i t y and outcome values (.63) d i f f e r e d s i g n i -f i c a n t l y from a chance expectation of .50 (Z = .26, _p_ = .05). The proportion of matches generated using.outcome d i r e c t i o n (,56) did not d i f f e r s i g n i f i c a n t l y from chance expectations (Z - 1,2, _p_ = .05). While the analysis of proportions indicated a s i g n i f i c a n t e f f e c t for predictions based on p r o b a b i l i t i e s / p a y o f f s , the absolute number of correct matches generated was only 7.5 of a possible 12 pre d i c t i o n s . This less than perfect success of the p r e d i c t i o n techniques led to a 46 decision to further examine the i n d i v i d u a l items. As i n the above analyses, the following discussion i s based on the choice patterns of the e n t i r e sample. Across the majority of p o s i t i v e items, there seems to have been a tendency for subjects to choose c e r t a i n rather than r i s k y options (Table 5). This was p a r t i c u l a r l y true when the c e r t a i n option led to the highest expected gain. When the r i s k y option was associated with the highest expected outcome, the c e r t a i n option was s t i l l chosen close to 50% of the time. Acorss the negative items, there was a tendency for subjects to choose r i s k y options when they led to the lowest expected l o s s . How-ever, t h i s did not appear to be as strong as the tendency to choose c e r t a i n options i n gain s i t u a t i o n s when they led to the greatest gains. When the ri s k y options were associated with the highest expected l o s s , people tended to prefer the c e r t a i n a l t e r n a t i v e . However, a substan-t i a l number chose the ri s k y option i n these s i t u a t i o n s . 47 Table 5 Percent of Subj ects Choosing Certain and Risky Opt ions Gain Situations Loss Situations Item Certain Risky Item Certain Risky 1 56% * 44% 7 16% A A 84% 2 45% A 55% 8 52% A A 48% 3 * 84% 16% 9 A A 73% 27% 4 A 55% 45% 10 44% A A 56% 5 46% * 54% 11 A A 55% 45% 6 * 84% 16% 12 60% 40% greatest expected gain lowest-expected loss 48 DISCUSSION The data supported the p o s i t i o n that adults of a l l ages tre a t decision s i t u a t i o n s s i m i l a r l y . The r e s u l t s of the two primary analyses indicated no s t a t i s t i c a l l y s i g n i f i c a n t differences between age groups on any of the indices of r i s k taking. The e l d e r l y people chose cer-t a i n options no more often than the younger people and, l i k e the younger people, tended to choose more c e r t a i n options i n gain than i n loss s i t u a t i o n s . The complete absence of between group differences suggests that formulations of age r e l a t e d r i s k aversion are incomplete. The analysis of the certainty index revealed that there was no s p e c i a l tendency f o r the e l d e r l y to choose completely predictable a l t e r n a t i v e s . Although there were some differences between the groups, with the e l d e r l y receiving s l i g h t l y higher scores, the differences were neither s t a t i s t i c a l l y s i g n i f i c a n t nor of the magnitude reported i n previous research (Botwinick, 1969). The f i n d i n g of s i m i l a r choice preferences under such conditions for a l l age groups i s d i r e c t l y contrary to Botwinick's report that the e l d e r l y prefer completely predictable over r i s k y options and are also inconsistent with his p o s i t i o n that the e l d e r l y employ r i s k avoiding s t r a t e g i e s . In Botwinick's work, i t was reported that a sub-s t a n t i a l number (Botwinick does not give an exact figure) of the e l d e r l y chose a completely predictable option when one was a v a i l a b l e . In the present study, subjects i n each of four age groups tended to divide t h e i r choices between r i s k y and c e r t a i n options. The f a c t that the e l d e r l y did not choose a majority of completely predictable options 49 i s d i f f i c u l t , i f not impossible, to r e c o n c i l e with the notion that the el d e r l y are averse to r i s k . Since both r i s k y and safe options were equally a v a i l a b l e , the el d e r l y should have chosen the safe options i f they were t r u l y r i s k averse. That they did not exhibit a strong pre-ference f o r the c e r t a i n options can only i n d i c a t e that they were not using s p e c i a l r i s k avoiding s t r a t e g i e s . One possible reason for the discrepancies between Botwinick's r e s u l t s and those reported by the author i s that the qu a l i t y and amount of information presented i n the studies d i f f e r e d . One of the c r i t i c i s m s of the choice dilemma instrument mentioned e a r l i e r i s that the s i t u a -tions were vague. Subjects may, consequently, have been put i n the p o s i t i o n of having to make a judgement based on vague and inadequate information. The behaviour patterns observed by Botwinick may have been caused by the manner i n which each age group reacted to the struc-ture of the choice item. As Basowitz and Korchin (1957) noted, e l d e r l y people may react negatively to ambiguous stimulus configura-tio n s . In the choice dilemma studies, i f the e l d e r l y perceived the items as ambiguous, they could have chosen non-risky options not be-cause they were cautious, but because they perceived that there was not enough information for them to make an adequate judgement. I t i s e n t i r e l y possible that under such circumstances they may have adopted a strategy i n which ambiguous but c e r t a i n a l t e r n a t i v e s were given a higher value than a l t e r n a t i v e s that were both uncertain and r i s k y . This suggests that differences i n responsiveness to the qu a l i t y of stimulus materials, rather than r i s k aversion could have produced Botwinick's r e s u l t s . In the present study, i n which a l l values of the 50 response options were e x p l i c i t l y stated, r i s k avoiding was not ob-served. The s i m i l a r i t i e s i n choice behaviour observed under these conditions may have occurred because the information was s u f f i c i e n t l y clear to prevent subjects' responses to the decision s i t u a t i o n s from being confounded with t h e i r r eaction to the qua l i t y of information i n the s i t u a t i o n s . My contention that e l d e r l y people respond d i f f e r e n t i a l l y to s i t u a t i o n a l information also receives some support from the studies of vocabulary test taking behaviour reported by Okum (1975, 1978), who studied choice behaviour when the expected outcome values of high and low r i s k options assumed several values. He found that behaviour which i n i t i a l l y suggested that e l d e r l y people were more cautious than younger subjects was ac t u a l l y a function of the way that the two group responded to the expected outcomes of the items. When the payoffs associated with low and high r i s k options were equal, the e l d e r l y chos more low r i s k options than the younger people, making them appear more cautious. But when the expected outcome of the high r i s k option was made greater than the outcome for the low r i s k option, the e l d e r l y people chose the high r i s k option as frequently as the younger people. The supposedly cautious behaviour of the el d e r l y was a c t u a l l y due to the age groups' d i f f e r e n t i a l responses to the expected outcome of the options. The analysis of subjects' response patterns across outcome types supported the notions that (a) people i n a l l age cohorts tend to use si m i l a r s trategies i n evaluating r i s k y s i t u a t i o n s , and (b) people's choices can be viewed to some extent as a function of the ,outcome 51 d i r e c t i o n of the choice s i t u a t i o n . In addition, the lack of s i g n i f i c a n t between group differences on either measure again points out that age group membership i s an i r r e l e v a n t v a r i a b l e i n simple decision tasks. Of most i n t e r e s t was the f i n d i n g that for a l l age groups, choice patterns s h i f t e d as a function of outcome d i r e c t i o n , with the response d i s t r i b u t i o n being comparable for each age group. Since the groups d i f f e r e d i n age and other v a r i a b l e s , i t seems l i k e l y that these si m i -l a r i t i e s i n performance were due to the f a c t that people used s i m i l a r strategies i n evaluating the options. The f i n d i n g that people's preferences s h i f t e d as a function of outcome type i s consistent with Okum's report (Okum, 1978) that choice behaviour s h i f t s as a function of expected payoffs of the a v a i l a b l e choices. I t i s also consistent with B u r k h i l l and Schaie's f i n d i n g that s i t u a t i o n a l contingencies influence the performance of the e l d e r l y . These studies provide evidence that the e l d e r l y are s e n s i t i v e to s i t u -a t i o n a l information and seem to formulate decision or choice strategies on the basis of that information. The findings of the two main analyses indicated that there was no evidence of cautious or r i s k avoiding behaviour on the part of the e l d e r l y . In addition, the e l d e r l y were shown to behave s i m i l a r l y to young, middle, and l a t e adults. These r e s u l t s support the p o s i t i o n that the e l d e r l y are not necessarily cautious, r i s k averse, and/or i n e f f i c i e n t i n r i s k y choice s i t u a t i o n s . Attempts to l i n k people's choice patterns to a p a r t i c u l a r com-bination of s i t u a t i o n a l parameters were less than t o t a l l y successful. Predictions generated using the expected values of the items both 52 matched subjects' choices more c l o s e l y than did predictions based on outcome d i r e c t i o n and d i f f e r e d s i g n i f i c a n t l y from chance expectations. However, neither set of predictions matched the data very w e l l . In retrospect, t h i s i s not completely s u r p r i s i n g . I t seems l i k e l y that people, even i n s i m p l i f i e d s i t u a t i o n s such as the analogues of l i f e events used i n the present study, use more complex de c i s i o n strategies than those used to generate predictions. In f a c t , the .analyses suggested that people's responses may have been influenced by aspects of the s i t u a t i o n s incorporated into each of the p r e d i c t i v e methods. I t was found that when the expected payoff for a c e r t a i n a l -ternative i n a gain s i t u a t i o n was higher than the expected payoff for a ri s k y option, people tended to choose the c e r t a i n option. In cases where the r i s k y option had the highest payoff, however, people's choices tended to be d i s t r i b u t e d between the r i s k y and c e r t a i n a l t e r n a t i v e s . A s l i g h t l y weaker version of t h i s phenomenon was observed i n loss s i t u -ations. People tended to choose the r i s k y option i f i t led to the smallest expected l o s s . When the r i s k y option led to the greatest ex-pected l o s s , people preferred the c e r t a i n option, although a substan-t i a l number s t i l l chose the r i s k y option. This f i n d i n g suggests that people may simultaneously prefer par-t i c u l a r types of response options and p a r t i c u l a r response values. People appear to prefer completely predictable options i n gain s i t u a -t i o n s , p a r t i c u l a r l y when the expected outcomes of the options agree with t h e i r choices. The converse may hold true f or loss s i t u a t i o n s , although the case i s not nearly as c l e a r , with people p r e f e r r i n g r i s k y options, p a r t i c u l a r l y i f they have the lowest expected values. The 53 s p l i t s between r i s k y and c e r t a i n options may have occurred because on some of the items, people's preference for a s p e c i f i c type of a l t e r n a -t i v e may have c o n f l i c t e d with t h e i r preference for a s p e c i f i c outcome value. In b r i e f , the r e s u l t s of the present study provide a d d i t i o n a l e v i -dence that the e l d e r l y are neither more cautious nor r i s k averse than younger adults. Thus the r e s u l t s stand beside an increasing array of evidence that people of a l l ages perform s i m i l a r l y i n many choice s i t u -ations. The present r e s u l t s are also of p a r t i c u l a r value i n that they provide the f i r s t i n d i c a t i o n that the e l d e r l y do not necessarily choose completely predictable options x^hen such options are a v a i l a b l e to them. This f i n d i n g counters the claim of gerontological researchers that the e l d e r l y use s p e c i a l r i s k avoiding s t r a t e g i e s . The r e s u l t s of t h i s and other studies have indicated that older adults are neither cautious i n decision s i t u a t i o n s , nor p a r t i c u l a r l y prone to use r i s k avoiding strategies to e x t r i c a t e themselves from r i s k y s i t u a t i o n s . Much, of ...the a v a i l a b l e evidence suggests that differences i n choice behaviour, when they are observed, are more l i k e l y to be a function of s i t u a t i o n a l variables than personality t r a i t s such as cautiousness. These findings do not lead to the conclusion that differences i n decision making strategies or choice behaviour across age groups are non-existent. In f a c t , as w i l l presently be pointed out, such differences may e x i s t . The r e s u l t s do suggest that i t i s a mistake to attempt to measure i n -stances of cautiousness using d e c i s i o n making tasks. The present research suggests that d e c i s i o n making i s a multi-dimensional pheno-menon, and that a l l dimensions must be considered i f research i s to 54 c l a r i f y the conditions under which performance differences may occur. The r e s u l t s of th i s and other studies suggest that there may be some i n t e r e s t i n g differences i n choice behaviour across age groups. For example, the findings of Okum (1976b, 1977) indic a t e that people of d i f f e r e n t ages may weight information differently, when choosing between r i s k y options. Since i n h i s study, expected payoff was asso-ciated with d i f f e r e n t i a l patterns of performance across ages, i t seems reasonable to suggest that one l i n e of research could begin there. Systematically manipulating the expected payoffs of d i f f e r e n t types of choice options could y i e l d information that could help paint a more accurate p i c t u r e of-the meaning of t h i s v a r i a b l e to d i f f e r e n t age groups. Such a strategy has been su c c e s s f u l l y employed i n a study of decision making, although not i n a l i f e span context. S l o v i c and Lichenstein (1967) devised a set of gambles that enabled them to assess people's preferences for p r o b a b i l i t i e s and payoffs independently, of each other. Their r e s u l t s helped to c l a r i f y the meaning of those values to people. There i s a clear need f or the a p p l i c a t i o n of s i m i l a r strategies to the issue of choice behaviour across the l i f e span. Only by applying such systematic strategies w i l l we be able to understand the ways i n which people treat d e c i s i o n s i t u a t i o n s at d i f f e r e n t points i n t h e i r • l i v e s . 55 REFERENCES Baltes, P.B. Longitudinal and cross s e c t i o n a l sequences i n the study of age and generation e f f e c t s . Human Development, 1968, 11, 145-171. Basowitz,. H. and Korchin, S.J. Age differences i n the perception of closure. Journal of Abnormal and S o c i a l Psychology, 1957, 54, 93-97. B u r k h i l l , W.R. and Schaie, K.W. The ef f e c t s of d i f f e r e n t i a l r e i n f o r c e -ment of cautiousness i n i n t e l l e c t u a l performance among the e l d e r l y . Journal of Gerontology, 1975, _5, 578-583. Botwinick, J. Cautiousness i n old age. Journal of Gerontology, 1966, 21, 347, 353. Botwinick, J. D i s i n c l i n a t i o n to venture versus cautiousness i n res-ponding: Age diff e r e n c e s . Journal of Genetic Psychology, 1969, 115, 347-353. Botwinick, J. Aging and behavior. New York: Springer Publishing Co., 1973. Buhler, C. The course of human l i f e as a psychological problem. Human Development, 1968, VI, 184-200. Edwards, W. Subjective p r o b a b i l i t i e s i n f e r r e d from decisions. Psychological Review, 1962, 69, 109-135. Erikson, E. Childhood and society. Toronto: W.W. Norton & Co., Inc., 1963. Griew, S. Age and the matching of s i g n a l frequency i n a two channel detection task. Journal of Gerontology , 1968, 23, 93-96. 56 Kahneman, D. and Tversky, A. Prospect theory: An analysis of d e c i -sion making under r i s k . Unpublished manuscript. Kimmel, D.C. Adult and aging. New York: John Wiley & Sons, Inc., 1974. Nunnally, J.C. Research strategies and measurement methods for i n -v e s t i g a t i n g human development. In J.R. Nesselroade and H.W. Reese (eds.) L i f e span developmental psychology: Methodological  issues . New York: Academic.Press, 1973. Okum, M.A. Adult age and cautiousness i n decision: A review of the l i t e r a t u r e . Human Development, 1976, 19_, 220-233. Okum, M.A. and DiVesta, F.J. Cautiousness i n adulthood as a function of age and i n s t r u c t i o n s . Journal of Gerontology, 1976, 31, 571-576. P i t z , G.F. and Downing, L. Optimal behavior i n a d e c i s i o n task as a function of i n s t r u c t i o n s and payoffs. Journal of Experimental Psychology, 1967, 73, 549-555. Rabbit, P. Changes i n problem solving a b i l i t y i n old age. In J. Birr e n and K.W. Schaie (eds.) Handbook of the psychology of aging . New York: Van Nostrand Rheinhold Company, 1977. Sanford, A.J., Griew, S., and O'Donnell, L. Age e f f e c t s i n simple p r e d i c t i o n behaviors. Journal of Gerontology, 1972, 27_, 259-264. Schaie, K.W. A general model for the study of developmental problems. Psychological B u l l e t i n , 1965, J54, 92-107. Sl o v i c , P. and Lichenstein, S. Relative importance of p r o b a b i l i t i e s and payoffs i n r i s k taking. Journal of Experimental Psychology Monograph, 1968, 78, 3, part 2. 57 Wallach, M.A. and Kogan, N. Aspects of judgement and decision making: Int e r r e l a t i o n s h i p s and changes with age. Behavior Science, 1961, _6, 23-36. Vroon, V.H. and Pahl, B. Relationships between age and r i s k taking among managers. Journal of Applied Psychology, 1971, J5_5, 399-405. 58 APPENDIX A Selection of the Age Groupings The age ranges used i n t h i s study are based on t h e o r e t i c a l con-ceptions of the l i f e span. Af t e r reviewing the d i f f e r e n t conceptua-l i z a t i o n s of adult development (Buhler, 1968; Erikson, 1963; Kimmel, 1974; Peck, 1969), four developmental groups were defined. The groupin while approximate, represent an i n t e g r a t i o n of the adult l i f e stages and associated time periods i d e n t i f i e d by these t h e o r i s t s . In keeping with the primary emphases of the theories, the groupings are based pr i m a r i l y on s o c i a l markers i d e n t i f i e d with c e r t a i n periods of adult l i f e . Each of the periods i d e n t i f i e d has i t s own p a r t i c u l a r context. Young adulthood i s often a time for s t a r t i n g a career, marrying or e s t a b l i s h i n g a permanent r e l a t i o n s h i p , and s e t t l i n g into an adult r o l e . This i s the time that Buhler associates with s e l f determination of goals and which Erikson r e l a t e s to the need to e s t a b l i s h a t r u l y intimate r e l a t i o n s h i p . Estimates of the age range of t h i s category from age 15 to an end point somewhere between 20 and 30. Middle adulthood has been a period l a r g e l y ignored by l i f e span researchers, but perhaps the most c e n t r a l c h a r a c t e r i s t i c of t h i s period was captured by Shakespear who i d e n t i f i e d the age of the s o l -d i e r . There seems to be a c e r t a i n v a l i d i t y i n conceptualization that the c e n t r a l c h a r a c t e r i s t i c of t h i s period i s a concern with adventure or power. In modern l i f e , middle adulthood i s often a time when one's attention i s turned to furthering ambitions through career and family a c t i v i t i e s . 59 Late adulthood i s often a time for assuming positions of career and l i f e importance. During t h i s period the person's power and influence are often at t h e i r zenith. Along with middle adulthood, t h i s time i s often associated with Erikson's notion of ge n e r a t i v i t y , and with Buhler's idea of s e l f assessment. Buhler suggested that the time boundaries for t h i s period were approximately between ages 45 and 65. Old age often means r e t i r i n g from career and family a c t i v i t i e s and developing new a c t i v i t i e s or i n t e r e s t s . Theorists t y p i c a l l y view . old age as a time for summing up and pronouncing judgement on one's l i f e . Buhler has envisioned old age as a time to experience the feel i n g s of either success or f a i l u r e regarding l i f e . Erikson has stated that the c e n t r a l problem of the e l d e r l y person i s to make sense of his l i f e and to accept i t as a meaningful experience. Almost with-out exception, t h e o r i s t s have set the boundary for entry into old age at or near age 65. It i s recognized that these groupings and age ranges are some-what a r b i t r a r y . This a r b i t r a r i n e s s i s i n e v i t a b l e , as there i s no agreement on the most appropriate way to d i v i d e up the l i f e span. The groupings have a degree of content v a l i d i t y , however, i n that they r e f l e c t the speculations of leading l i f e span t h e o r i s t s . In addition, the age ranges possess a c e r t a i n face v a l i d i t y . People f a l l i n g into each group can often be d i f f e r e n t i a t e d from members of other groups. Thus while one might confuse a 35 and a 40 year old, there i s l i t t l e l i k e l i h o o d that one would confuse a 35 year.old with either a 25 year old or a 50 year o l d . In short, while these groupings are not 60 offered as d e f i n i t i v e l i f e stages, they are seen as a reasonable way to divide the l i f e span into meaningful groups. Defining the Age Ranges To keep the periods as d i s t i n c t as possible, i t was decided to place 5 year buffer zones between the age groupings. This was done pr i m a r i l y for s t a t i s t i c a l reasons. In cases where the path of development i s unknown and developmental periods cannot be c l e a r l y s p e c i f i e d , allowing age groupings to border too c l o s e l y upon each other may lead to d i s t o r t i o n s i n the data analyses. If performance chnages occured near a boundary, the changes would be r e f l e c t e d i n the performance of a l l people l y i n g near the boundaries regardless of group membership. This would r e s u l t i n increased v a r i a b i l i t y within groups and decreased v a r i a b i l i t y between groups. P h y s i c a l l y separating the groups helped to ensure that performance patterns could be more c l e a r l y seen as a func-t i o n of group membership. 61 APPENDIX B Subject S e l e c t i o n Procedures Because of the problems involved i n r e c r u i t i n g subjects from a v a r i e t y of age groups, a s p e c i f i c r a t i o n a l e for s e l e c t i n g subjects was used i n t h i s study. I t has long been recognized that cohort s p e c i f i c e f f e c t s may a f f e c t subjects' performance on dependent measures i n un-predictable ways and a considerable body of l i t e r a t u r e has accumulated analyzing experimental designs that separate age and cohort e f f e c t s . There i s a smaller amount of l i t e r a t u r e dealing with the problems of cohort differences i n sampling procedures. Nunally (1976) has sug-gested that i t i s quite easy to obtain comparable samples for cross s e c t i o n a l studies as long as the researcher follows randomization or quasi-randomization procedures. While i t may be true that randomiza-t i o n i s the best procedure, at a t h e o r e t i c a l l e v e l , i t i s nearly im-possible to put into p r a c t i c e . The l o g i s t i c a l d i f f i c u l t i e s involved i n implementing randomization procedures i n l i f e span research, and p a r t i c u l a r l y i n cross s e c t i o n a l studies, put i t beyond the..reach of most researchers. Stated most simply, adults seem to r e s i s t attempts to 'randomly sample' them. They are usually d i f f i c u l t to f i n d and often unwilling to p a r t i c i p a t e a f t e r contact has been made with them. Because recruitment then becomes li m i t e d to that very s e l e c t subgroup of adults that i s w i l l i n g to put up with the psychological researcher, the basic tenets of randomiza-t i o n procedures are v i o l a t e d . Gerontological researchers have often resorted to an a l t e r n a t i v e 62 strategy involving matching groups of adults of d i f f e r e n t ages on some v a r i a b l e of global importance, such as education or IQ. While t h i s strategy may s u p e r f i c i a l l y appear to produce comparable groups, cohort e f f e c t s operate d i r e c t l y against t h i s . Mismatching a c t u a l l y occurs because the d i s t r i b u t i o n of values for the matching v a r i a b l e often d i f f e r s between cohort groups. For example, 12 years of educa-t i o n i s the modal amount of education i n the 20-30 year old cohort group, while 12 years of education i n the 65 year old cohort i s i n the top of the d i s t r i b u t i o n of educational achievement. If between group comparisons were done a f t e r matching on the basis of education, per-formance comparisons would a c t u a l l y be between the top performers i n the e l d e r l y cohort and the modal performers i n the younger cohort. A s i m i l a r s i t u a t i o n would occur i f IQ, SES, or any other v a r i a b l e suc-c e p t i b l e to h i s t o r i c a l forces were used as a^matching v a r i a b l e . These problems r u l e out matching as a s a t i s f a c t o r y way of achieving compar-able groups i n l i f e span research. For the reasons mentioned above, both randomization and matching were deemed inappropriate strategies f o r t h i s study. Instead, another method of subject s e l e c t i o n involving a quasi-matching procedure was used. Only subjects who had f i r s t s e l f selected to p a r t i c i p a t e i n s p e c i a l educational or enrichment a c t i v i t i e s were asked to p a r t i c i p a t e i n the study. The assumption underlying t h i s approach i s that subjects seeking s i m i l a r a c t i v i t i e s were more l i k e l y to occupy s i m i l a r positions within t h e i r age cohorts. For example, people from each cohort who are interested i n i n t e l l e c t u a l a c t i v i t i e s are more l i k e l y to f a l l into s i m i l a r a b i l i t y l e v e l s within t h e i r cohorts than people chosen by other 6 3 means. Recruiting subjects who might be expected to be similar with respect to their cohort distributions helped to ensure that between group comparisons would not be confounded by systematically mismatched variables. 64 APPENDIX C INSTRUCTIONS On the following pages are a number of stories about situations that might occur in l i f e . In each situation a person is faced with two courses of action. The outcome of one course of action is completely predictable. The outcome of the other involves a certain amount of risk. For each situation I would like you to read over the story and decide which course of action seems most appropriate for the situation. When you have made a choice, please make a check mark on the line in front of your choice. There are no right or wrong answers for these situations. We only wish to see what types of alternatives people choose. Thank you very much for helping me with this study. 65 DEMOGRAPHIC INFORMATION PLEASE ANSWER THE FOLLOWING QUESTIONS AGE IN YEARS; SEX; MALE FEMALE MARITAL STATUS; MARRIED SINGLE SEPARATED WIDOWED DIVORCED COMMON LAW EDUCATION; PLEASE CHECK THE HIGHEST LEVEL OF EDUCATION YOU HAVE REACHED. UNIVERSITY-UNDERGRADUATE UNIVERSITY-POST GRADUATE 1 YEAR 1 YEAR 2 YEARS 2 YEARS 3 " 3 " 4 " 4 " 5 " 5 " 6 or more - Specify 6 or more - Specify EMPLOYMENT; WHAT IS YOUR PRESENT EMPLOYMENT STATUS? IN TRAINING FOR EMPLOYMENT RETIRED HOMEMAKER EMPLOYED UNEMPLOYED IF NOT RETIRED. WHAT IS YOUR PROFESSION? IF YOU ARE, OR HAVE BEEN, MARRIED, WHAT IS (WAS) YOUR SPOUSES PROFESSION? GRADE 8 OR LESS (SPECIFY) GRADE 9 GRADE 10 GRADE 11 GRADE 12 GRADE 13 IF YOU ARE RETIRED. WHAT WAS YOUR PROFESSION DURING YOUR WORKING CAREER? 66 1. A man is considering investing in the grocery business, and is trying to decide which of two small stores to buy. Store 1: There is a guaranteed annual return of $5000. Store 2: There is a 60% chance of making $10,000. annually, and a 40% chance of making no profits. 2. A person has inherited a sum of money and wishes to invest i t for a term of one year. ______ Plan 1: Purchase a blue chip stock with a guaranteed annual return of $1000. _______ Plan 2: Purchase stock in a company with 70% chance of making a profit of $2000. and a 30% chance of making no profit . 3. A person works in a travel agency and can vacation for free in one of two company owned places, provided that paying customers are not occupying the space. Place 1: It is absolutely certain that he w i l l be able to stay there for 8 days. ______ Place 2: There is a 60% chance of staying for 10 days, and a 40% chance the reservations w i l l be cancelled. 4. A person is trying to decide which of two neighbourhoods to move.to. • Place 1: It is absolutely certain that living expenses w i l l be $700. less per year. Place 2: There is a 70% chance that living expenses w i l l be $900i less per year, and a 30% chance that they w i l l remain the same. 67 5. A person has decided to attend university and must choose which of two schools to attend. Both schools are excellent and w i l l lead to good careers. School 1: The school w i l l provide a summer laboratory assistantship paying $1500. ________ School 2: The school reports that there is an 80% chance of getting a summer assistantship worth $2500., and a 20% chance that there w i l l be no assistantship available. 6. Aperson is playing a card game and is in danger of losing a certain amount of money. He is trying to decide which strategy to follow. Strategy 1: It is absolutely certain that this w i l l results in losing $80. Strategy 2: With this strategy there is a 60% chance of losing $100. and a 40% chance of breaking even. 7. A person is being forced to relocate to a new place, and is trying to decide which of two places to move to. ________ Place 1: It is absolutely certain that the cost of home maintenance w i l l increase by $700. _______ Place 2: There is a 70% chance that the cost of home maintenance w i l l increase by $900., and a 30% chance that the costs wi l l be the same as they are now. 8. A person has been offered a choice between a special lottery ticket and a sum of money. ______ Choice 1: $900. Choice 2: A ticket with a 50% chance of winning $1500,, and a 50% chance of being worthless. 6 8 9. A person has decided to attend a private school, and must decide which of two schools to attend. _____ School 1: It is certain that he/she w i l l have to pay $1500. in tuition and fees. School 2: There is an 80% chance of having to pay $2500. in tuition and a 20% chance of receiving a scholarship and not having to pay anything. 10. A person is trying to decide how to declare his income on a business form. There are two legal options open to him. _____ Option 1: It is absolutely certain that he wi l l have to pay $900. in corporate fees. Option 2: There is a 50% chance of having to pay $1500., and a 50% chance of not having to pay anything. 11. A man is trying to decide which of two equally priced care to buy, and would like to make the best purchase. One is a standard model, and the other may become a collectors item. _____ Car 1: It is absolutely certain that the value of the car wi l l decrease by $1000. a year. Car 2: There is a 70% chance that the value of the car w i l l decrease by $2000. a year, and a 30% chance its value w i l l not decrease at a l l . 12. A person wishes to dissolve a business and is trying to decide when is the best time to do so. Time 1: Dissolving the business now means a certain loss of $5000. _____ Time 2: Waiting for 6 months means there is a 60% chance of losing $10,000., and a 40% chance of breaking even. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0094411/manifest

Comment

Related Items