UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Measuring completion rate in distance education Wong, Charles Kit Hung 1987

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

Item Metadata

Download

Media
831-UBC_1988_A8 W66.pdf [ 6.69MB ]
Metadata
JSON: 831-1.0055760.json
JSON-LD: 831-1.0055760-ld.json
RDF/XML (Pretty): 831-1.0055760-rdf.xml
RDF/JSON: 831-1.0055760-rdf.json
Turtle: 831-1.0055760-turtle.txt
N-Triples: 831-1.0055760-rdf-ntriples.txt
Original Record: 831-1.0055760-source.json
Full Text
831-1.0055760-fulltext.txt
Citation
831-1.0055760.ris

Full Text

MEASURING COMPLETION RATE IN DISTANCE EDUCATION by CHARLES KIT HUNG WONG B . S c , U n i v e r s i t y of London, 1971 Diploma, London C o l l e g e of P r i n t i n g , 1972 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n THE FACULTY OF GRADUATE STUDIES Department of A d m i n i s t r a t i v e , A d u l t and Hi g h e r E d u c a t i o n We ac c e p t t h i s t h e s i s as c o n f o r m i n g t o the r e q u i r e d s t a n d a r d THE UNIVERSITY OF BRITISH COLUMBIA November 1987 © 1987, C h a r l e s K. H. Wong In p resen t ing this thesis in part ia l f u l f i lmen t o f t h e requ i remen ts fo r an a d v a n c e d deg ree at t h e Univers i ty o f Bri t ish C o l u m b i a , I agree that t h e Library shall m a k e it f ree ly avai lable fo r re ference and s t u d y . I fu r ther agree that permiss ion fo r ex tens ive c o p y i n g o f th is thesis fo r scholar ly p u r p o s e s may b e g r a n t e d by the h e a d o f m y d e p a r t m e n t o r by his o r he r representa t ives . It is u n d e r s t o o d that c o p y i n g o r pub l i ca t i on o f th is thesis f o r f inancia l ga in shall n o t b e a l l o w e d w i t h o u t m y w r i t t e n permiss ion . D e p a r t m e n t o f HbJ*, f^TU* <=h rtitfL&S &*~C*Xf<*~ The Univers i ty o f Brit ish C o l u m b i a 1956 M a i n M a l l Vancouver , Canada V6T 1Y3 DE-6(3/81) ABSTRACT The purpose of t h i s study was to c r e a t e and examine the c o n c e p t u a l and psychometric p r o p e r t i e s of four components that comprise 'programme outcome' sought by d i s t a n c e educators, and to examine the- extent to which student ( s o c i o -demographic) and programme (e.g. d u r a t i o n ) v a r i a b l e s r e l a t e d t o them. T h i s ex post f a c t o study u t i l i s e d the r ecords of 773 correspondence students e n r o l l e d at The Chinese U n i v e r s i t y of Hong Kong for the 1984 Summer s e s s i o n and who submitted one or more assignments. Four v a r i a b l e s were d e r i v e d from the data set - completion r a t e , d e v i a t i o n ( l a t e n e s s i n submitting assignments), turnaround (time taken to r e t u r n marked assignments) and grades. These v a r i a b l e s were more c o n c e p t u a l l y d e f e n s i b l e than the NUEA or other formulae t y p i c a l l y used to measure 'outcomes'. I t was hypothesized t h a t when students had to wait longer f o r the r e t u r n of t h e i r assignments in the f i r s t q u a r t e r of the course, completion r a t e would be lower, but t h i s would not happen a f t e r the course was h a l f over. When students were l a t e s u b m i t t i n g assignments, i t was expected that t h e i r completion rate would be lower than those s u b m i t t i n g on time. I t was found that turnaround had a s i g n i f i c a n t a s s o c i a t i o n with completion rate throughout the course. D e v i a t i o n , that i s , delays i n s u b m i t t i n g assignments, was a l s o r e l a t e d to completion. Each i i of the four variables had s i g n i f i c a n t l y d i f f e r e n t associations with programme outcome. The measures employed here can be used elsewhere as the data that comprise them are found in the records of most distance education programmes. This should f a c i l i t a t e research in distance education and provide practitioners with a way to monitor programmes. i i i TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS i v LIST OF TABLES v i LIST OF FIGURES '. v i i i ACKNOWLEDGMENTS i x CHAPTER ONE: INTRODUCTION 1 Background to the Problem 1 Problems 3 Purpose and Scope 6 S t a t i n g the Hypotheses 7 O r g a n i s a t i o n of Subsequent Chapters 10 CHAPTER TWO: REVIEW OF LITERATURE 11 Def i n i t ion 11 Reviews of Research 12 Chronology of Research on Completion Rate 16 Summary 34 CHAPTER THREE: VARIABLES 39 Independent V a r i a b l e s 39 Dependent V a r i a b l e s ' 4 2 Choice of V a r i a b l e s 48 Summary 50 CHAPTER FOUR: METHOD 52 Data Source • . 52 Data C a p t u r i n g 54 T r a n s p o r t a t i o n of Data 57 Data Transformation 59 V e r i f i c a t i o n 64 Fu r t h e r T r a nsformations 66 CHAPTER FIVE: DATA ANALYSIS 70 S e l e c t i o n of A n a l y t i c a l Procedures 70 Procedures 71 Fu r t h e r Procedures . . . 72 Summary 7 4 iv TABLE OF CONTENTS (Continued) CHAPTER SIX: RESULTS I 77 Characteristics of Participants and Programme . . . 78 Relationships between Student and Programme Variables and 'Outcome' 86 Relationships between Variables Describing 'Outcome' 90 Ana l y t i c a l Tests Across Time and Groups 92 CHAPTER SEVEN: RESULTS II 93 Completion and Turnaround at Various Stages . . . . 94 Completion and Deviation at Various Stages . . . . 97 CHAPTER EIGHT: DISCUSSION 99 Comparison with Previous Research 99 Interpretation of Present Results . . . . 102 Implications for Future Research 104 Implications for Prac t i t i o n e r s 106 REFERENCES 108 APPENDICES 115 v LIST OF TABLES Table 1: Previous findings on relationships between grades, completion, and deviation and student socio-demographic c h a r a c t e r i s t i c s 35 Table 2: Previous findings on relationships between grades, completion, deviation and other outcome variables 36 Table 3: Previous findings on relationships between grades, completion, deviation and programme variables 37 Table 4: Independent variables used in a study of distance education outcome 42 Table 5: Mapping courses of d i f f e r e n t duration (in units) into four parts 73 Table 6: Age groups occupied by 773 students in the programme 78 Table 7: Number and percent of male and female students in the programme 79 Table 8: Educational l e v e l of students in the programme 80 Table 9: Occupational groups of students in the programme 81 Table 10: Courses in the programme grouped by duration . 82 Table 11: Number and percent of students in the programme grouped by language of instruction 82 Table 12: Number and percent of students in the programme grouped by fees paid 83 Table 13: Number and percent of students in the programme grouped by nature of course . . . . 84 Table 14: Number and percent of students in the programme grouped by number of assignments submitted 84 Table 15: Enrolment for each course of the 1984 summer programme at the Chinese University of Hong Kong 86 v i L I S T O F T A B L E S (Continued) T a b l e 16: A n a l y s i s f o r a l l c a s e s , n = 773 88 T a b l e 17: C o r r e l a t i o n - (Pearson product-moment) between d i s t a n c e e d u c a t i o n 'outcome' elements . . . . 91 T a b l e 18: Means s c o r e s of c o m p l e t i o n , d e v i a t i o n , grade and t u r n a r o u n d a f t e r each q u a r t e r of c o u r s e s i n the 1984 summer d i s t a n c e e d u c a t i o n programme a t the C h i n e s e U n i v e r s i t y of Hong Kong ' 94 T a b l e 19: C o r r e l a t i o n s between d e r i v e d v a r i a b l e s a t v a r i o u s s t a g e s and a t end of the c o u r s e . . . 96 T a b l e 20: C o r r e l a t i o n s between c o m p l e t i o n and t u r n a r o u n d 97 T a b l e 21: C o r r e l a t i o n s between c o m p l e t i o n and d e v i a t i o n 98 T a b l e 22: Grades, c o m p l e t i o n & d e v i a t i o n vs dependent v a r i a b l e s , . . . 100 T a b l e 23: Grades, c o m p l e t i o n and d e v i a t i o n v s . s t u d e n t v a r i a b l e s 101 T a b l e 24: Grades, c o m p l e t i o n & d e v i a t i o n vs -programme v a r i a b l e s 102 v i i L I S T OF FIGURES F i g u r e 1 . Enrolment form used by students e n r o l l e d at the Chinese U n i v e r s i t y 53 F i g u r e 2. Computer screen for input of course records . 55 'Figure 3. Computer screen f o r input of enrolment r e c o r d 56 F i g u r e 4. Computer screen f o r input of assignment r e c o r d 57 F i g u r e 5. Sample of enrolment records before and a f t e r t r a n s f o r m a t i o n 61 F i g u r e 6. S e c t i o n of assignment f i l e before t r a n s f o r m a t i o n 62 F i g u r e 7. S e c t i o n of assignment f i l e a f t e r t r a n s f o r m a t i o n . . . . . 63 i v i i i ACKNOWLEDGMENT S I wish to express my s i n c e r e a p p r e c i a t i o n to the two s u p e r v i s o r s of the t h e s i s , Dr Roger Boshier and Dr Thomas Sork, who have provided much needed support and understanding and to The Chinese U n i v e r s i t y of Hong Kong f o r making the data a v a i l a b l e f o r the study. ix CHAPTER ONE INTRODUCTION Background to the Problem • 'Distance education' i s a term commonly used to describe situations where, because of the physical separation of learners and teacher, teaching is mainly done through a variety of print and electronic (audio, v i s u a l , d i g i t a l ) media. The most well known form of distance education involves courses by correspondence, other forms involve radio, t e l e v i s i o n , audio tape, video tape, computer -with or B without t u t o r i a l support - or a variety of combinations of any of the above. In North America, the term 'independent' or 'home' study i s more widely used to describe correspondence courses. Distance education, p a r t i c u l a r l y in the form of correspondence education, has been providing education for large numbers of people for almost a century. The term distance education i s now normally preferred. This was demonstrated by the change of t i t l e of the international professional organisation in t h i s f i e l d when, in 1982, the International Council on Correspondence Education became the International Council for Distance Education. In vast countries l i k e A u s t r a l i a and Russia, whole generations of people in remote areas have received their entire education by correspondence. In 'developing' countries, educators are kept busy planning and implementing 1 new distance education programmes for mi l l i o n s of people in both rural and urban areas. Literacy campaigns in Asia, A f r i c a and Latin America rely on the 'delivery' of programmes by distance education. In China, the need to provide higher education to large number of people quickly fueled the dramatic development of the Television Broadcasting University which currently e n r o l l s 500,000 students, surely the largest single teaching i n s t i t u t i o n in the world. As observed by Kuhanga (1981), "while teaching at a distance is being adopted in many countries in the world, there is cert a i n l y an even greater need for i t in developing countries, i f we are to expand educational opportunities, formal and non-formal, to cover a larger proportion of our populations" (p. 11). In 'developed' countries, the success of the B r i t i s h Open University (UKOU) prompted many countries to examine their own distance education e f f o r t s and to rep l i c a t e the UKOU with adaptation. University extension departments in the United States have been o f f e r i n g independent study programmes to thousands of students since the University of Chicago began correspondence courses in 1891. In Canada, authorities have expanded distance education while education as a whole has been experiencing cut-backs. As noted by Rumble and Harry (1982), "the distance teaching u n i v e r s i t i e s have been seen as the most s t r i k i n g development in higher education structures in recent decades, while the success and 2 acclaim with which some of them have been met suggests that they w i l l constitute a permanent component of the higher education sector of a number cf national educational systems in 1980s and 1990s" (p. 24). Much of the provisions in distance education i s devoted exclusively to adults. The B r i t i s h Open University, the Chinese TV Broadcasting University, the Thai Open University (STOU), the Canadian Athabasca University and B. C. Open Learning In s t i t u t e , the Spanish Open University (UNSTAD) and many others, are serving mostly adults. In Europe and Au s t r a l i a where distance education is more often seen as the channel to improve one's q u a l i f i c a t i o n , there are thousands of distance education courses available to help adults acquire more s k i l l s and technical competence. At the t e r t i a r y l e v e l , there are now more than a dozen open u n i v e r s i t i e s scattered throughout the world and countless other open education systems for lower l e v e l programmes in a wide range of subjects. Problems As distance education plays such a v i t a l role in providing education to so many, i t s outcome and effectiveness become important issues-. From the review of previous research (see Chapter-Two), i t was evident that many of the e f f o r t s were attempts to address these issues. As research topics, these two issues have been translated into student 3 achievement and completion rates. The bulk of the work on achievement was done in the 1950's and 1960's and centred around comparisons between the grades obtained by students studying by correspondence and .those studying by conventional methods. Much of th i s work was flawed because i t involved a comparison of a method (correspondence) with a technique (e.g. lecture) and as Verner (1964) noted, i t is pointless to compare apples and oranges. With regard to completion rates, serious research e f f o r t s were started as long as t h i r t y years ago (see Chapter Two) and continued to the present day. A deluge of findings were published in the 1970's. However, a number of problems persisted. F i r s t , there seemed to be no agreement on the most fundamental d e f i n i t i o n of 'completion' or 'dropout' so i t was d i f f i c u l t to compare findings. A variety of c r i t e r i a were used to determine successful completion: r e g i s t r a t i o n not withdrawn, submission of a s u f f i c i e n t number of assignments and achieving acceptable grades, passing the appropriate tests and examinations and attending s u f f i c i e n t number of t u t o r i a l s . The number and extent of these c r i t e r i a varied from i n s t i t u t i o n to i n s t i t u t i o n . There has always been a percentage of students who had never handed in even one assignment - these were sometimes known as non-starters. In some systems, only those who withdrew their r e g i s t r a t i o n s were c l a s s i f i e d as non-starters. Some argued that completion rate should only take into account those who started the 4 course, i . e . handed in at least one assignment. This eventually came to be known as the National University Extension Association (NUEA) formula. Others argued that completion rate must show a l l those who did not complete, irrespective of whether they were starters or non-starters. This is known as the t o t a l enrolment formula. Calculations based on t h i s p r i n c i p l e would normally show a much lower completion rate than those based on other formulae. Non-completion was another point of contention. One formulation claimed that whoever did not formally withdraw from the course by cancelling r e g i s t r a t i o n was a completer. As a result, higher completion rates would be returned from such .a computation than when inactive students were treated as non-completers. In practice, the most commonly used formulations were the NUEA formula and one derived from div i d i n g 'completers' by ' t o t a l enrolment'. A l l these formulations concerned completion of the whole course. Students were categorised into two groups, those who completed and those who did not. This might su f f i c e for administrative reports, but for the purpose of research, the inadequacy of such categorisation becomes apparent with a simple example. Surely, someone who completes more than 80 percent of the course should f a l l into a di f f e r e n t category from someone who completes ten percent. A more dynamic and precise measure for completion i s needed, one that can show 5 completion rate at each point of the course for each student. One purpose of t h i s study was to create and use such a t o o l . A good measure of completion rate takes on even greater importance in the l i g h t of previous research. Many of the studies reviewed in Chapter Two used completion as a key variable in examining factors that af f e c t student performance. In fact, completion rate and grades were two of the most frequently used indicators of student performance. However, most distance education i n s t i t u t i o n s also regularly keep records which can provide other information on the performance of students and the i n s t i t u t i o n i t s e l f . This concerns how close to the due dates students were submitting their assignments - deviation, and how long the i n s t i t u t i o n s took to mark and return them - turnaround. These variables: completion rate, grades, and deviation describe student performance; and together with the fourth variable, turnaround, as this study w i l l show, provide a comprehensive and precise p r o f i l e of distance education programme outcomes. Purpose and Scope This study was designed to contribute to theory about, and the operation of, distance education. There were two purposes: 1. To create and examine the conceptual and psychometric properties of four components that comprise programme 'outcome' usually sought by distance educators. 6 2. To examine the extent to which four 'student' s o c i o -demographic v a r i a b l e s and four 'programme' v a r i a b l e s were r e l a t e d to programme 'outcome'. These purposes were achieved by reviewing l i t e r a t u r e on 'outcome' and by e m p i r i c a l o p e r a t i o n s performed on a la r g e -data set r e c o r d i n g the 1984 summer s e s s i o n correspondence o p e r a t i o n at the Department of Extramural S t u d i e s , The Chinese U n i v e r s i t y of Hong Kong. The d e f i n i t i o n s of each of these v a r i a b l e s (four components of programme outcome) w i l l be presented and the advantages over previous, measurements d i s c u s s e d . Using the new f o r m u l a t i o n s , the researcher attempted to v e r i f y some of the p r e v i o u s f i n d i n g s concerning programme outcome and, i n p a r t i c u l a r , completion, and f a c t o r s a s s o c i a t e d with i t . Four hypotheses, based on a p r o j e c t i o n of previous f i n d i n g s , were t e s t e d . S t a t i n g the Hypotheses Hypothesis 1 "Students' completion r a t e i s s i g n i f i c a n t l y a s s o c i a t e d with turnaround one quarter way i n t o the course; those e x p e r i e n c i n g s h o r t e r turnaround w i l l submit more assignments than those e x p e r i e n c i n g longer turnaround." I t has been suggested i n p r e v i o u s r e s e a r c h that the time taken by the d i s t a n c e education i n s t i t u t i o n to respond to student assignments - turnaround - has a s i g n i f i c a n t e f f e c t 7 on completion. I t was f u r t h e r suggested that the e f f e c t i s more prominent at the beginning of the course. The hyp o t h e s i s was formulated to determine whether there i s evidence to support such s u g g e s t i o n s . The s u p e r i o r i t y of the new measures d e f i n e d i n t h i s study as a p p l i e d to a r e a l s i t u a t i o n are demonstrated even i n the f o r m u l a t i o n of hypotheses. I t was p o s s i b l e to be more s p e c i f i c about 'at the beginning of the course', i n t h i s case, the f i r s t q u a rter of the course i n terms of time. The f o r m u l a t i o n a l s o r e q u i r e d that there i s a way to measure completion r a t e at the f i r s t q u a r t e r p o i n t . Hypothesis 2 " A f t e r p a s s i n g the halfway p o i n t of the course, turnaround does not c o r r e l a t e with students' completion r a t e s i g n i f i c a n t l y . " I r r e s p e c t i v e of whether Hypothesis 1 i s r e j e c t e d or not, pr e v i o u s r e s e a r c h had not thrown l i g h t on whether turnaround has the same e f f e c t on completion a t the f i r s t and l a s t p art of the course. I f turnaround has an e f f e c t on completion r a t e at the f i r s t but not the l a s t two q u a r t e r s , i t would suggest that an i n s t i t u t i o n can best i n f l u e n c e completion at the beginning of the course. On the other hand, i f turnaround s t i l l has an e f f e c t a f t e r p a s s i n g the halfway mark, the i n s t i t u t i o n should pay a t t e n t i o n to turnaround throughout the programme. 8 Hypothesis 3 "One quarter way into the course, 'high' deviation (slow in submitting assignments) students have a lower completion rate than 'low' deviation (quick in submitting assignments) students." Students with high deviation rate are those who had been late in the submission of assignments. Previous research suggested that students who have not been able to submit assignments on time are less l i k e l y to complete courses than 'on time' students. This hypothesis was formulated to see i f there is evidence to support the suggestion. Again, i f the hypothesis is not rejected, i n s t i t u t i o n s o f f e r i n g distance education programmes should make every e f f o r t to reduce the extent of deviation at the beginning of the course. Hypothesis 4 "Student variables (sex, age, education and occupation) and programme variables (duration, fees, nature of course and language of instruction) are not s i g n i f i c a n t l y associated with the completion rate, deviation and grades of students." Previous research on relationships between outcome and student variables (sex, age, education and occupation) and programme variables (duration, fees, nature of course and language of instruction) has produced varying r e s u l t s . Usually, only very few variables were selected for 9 examination which made comparisons of results d i f f i c u l t . Moreover, the focus was usually on one 'outcome' variable such as completion or turnaround with some student and programme variables and f a i l e d to show the ove r a l l programme 'outcome' as in this study. Also, some of the inter-relations between the variables were not examined before. Hypothesis 4 has been formulated to v e r i f y previous findings and at the same time to test unexplored relationships. It would only be confirmed when there i s no s i g n i f i c a n t associations in a l l the relationships. O r g a n i s a t i o n of Subsequent Chapters A review of l i t e r a t u r e on research in distance education and, in pa r t i c u l a r , on completion rates i.s presented in Chapter Two. As the four derived variables (completion rate, grades, deviation, and turnaround) are central to this study, Chapter Three i s devoted e n t i r e l y to their d e f i n i t i o n s and advantages they o f f e r . Chapter Four d e t a i l s how data were obtained, v e r i f i e d and prepared for analysis. Chapter Five describes the basis for choosing a n a l y t i c a l procedures to test the hypotheses. Results are presented in Chapter Six and Seven where the hypotheses are confirmed or rejected. A discussion of the implications of the results is presented in Chapter Eight. 10 CHAPTER TWO REVIEW OF LITERATURE The review of l i t e r a t u r e is presented in four parts, sta r t i n g with a d e f i n i t i o n of distance education. Previous reviews of research in distance education are then examined. This i s followed by an annotated chronology of research related to completion rate in distance education. This chapter concludes with a summary highlighting results most relevant to the present study. D e f i n i t i o n The term 'distance education' as used in t h i s study refers to "the various forms of study at a l l levels which are not under the continuous, immediate supervision of the tutors present with their students in the lecture rooms or on the same premises, but which, nevertheless, benefit from the planning, guidance and t u i t i o n of a t u t o r i a l organization" (Holmberg, 1977, p. 9). The notion of distance i s elaborated by Harris and Williams (1977). A working d e f i n i t i o n for 'distance education' must stress that i t would include any planned and regular educational provision where there is distance between teacher (or instructor or educator) on the one hand, and the student (or learner or receptive audience) on the 1 1 other hand. This must mean primarily a distance in place and distance in time, whatever other interpretation, psychological or s o c i a l , might be thought to arise from the f i r s t aspects of distance, (p. 45) 'Distance learning' is often used as a synonym for 'distance'education'. The same confusion also plagues discussion about ' l i f e l o n g learning' and ' l i f e l o n g education'. For present purposes, 'distance learning' i s regarded as a tautology and i t is hard to envisage a case of 'distance learning'. Learning is an intern a l , largely psychological process. 'Education involves the systematic creation of conditions that f a c i l i t a t e learning; i t i s an external process. This i s a thesis about distance education. Reviews of Research There have been a few major reviews of research in distance education. Childs (1971), after preparing a report on recent research developments in correspondence instruction to the Eighth Conference of the International Council on Correspondence Education in 1969, wrote: The volume of writing and reporting on thi s subject i s , I believe, substantially greater than was true only a short time ago.... However, even though the volume of 12 l i t e r a t u r e may be increasing, t h i s does not indicate any great upsurge in research a c t i v i t y . Evidence of c a r e f u l l y done research i s s t i l l hard to find (p. 229). On the topic of completion, Childs reported the work of Sloan (1965), Pulley (1965), Powell (1965), Donehower (1968) and P f e i f f e r (1969) but refrained from making general observations on the state of research on completion as a whole. Most attention in the 1960's was given to comparing student achievements between face-to-face and correspondence instruction and the debate on whether t h i s was a legitimate comparison. Mathieson (1971) agreed with Childs about the lack of substantial research and noted that l i t t l e substantial investigation had been done. Much remained to be done and research had not been one of correspondence study's fortes. He surveyed the work of F a i r i n g and Hughes (1950), Bradt (1954), Hughes (1955), James and Wedemeyer (1959), Spencer (1965), Sloan (1965), Donehower (1968), P f e i f f e r and Sabers (1970) and H a r t s e l l (1971) in the area of completion rates and concluded: The problem can be divided into three related question areas (1) what i s the t y p i c a l completion rate; (2) what factors influence whether or not correspondence students w i l l pursue the course to completion; (3) what can be done to improve the completion rate? (p. 50). 13 Holmberg compiled two bibliographies on distance education in 1974 and 1977 and surveyed recent research (after 1970) for the 1932 Twelfth Conference of the International Council for Distance Education. Like his previous work; the results of the survey (Holmberg, I982a,b) have served as a good bibliographic source. Although Holmberg did not comment on the research i t s e l f , he made known works done in one country to scholars in others while the work of Childs and Mathieson has largely been r e s t r i c t e d to North America. Holmberg also drew attention to work completed by scholars outside the f i e l d . Coldeway (1982) noted that published papers on distance education f e l l into f i v e categories: 1. Position papers that discuss the phenomenon from the authors' perspective with l i t t l e attempt to define terms and variables 2. Descriptions of practice at a p a r t i c u l a r i n s t i t u t i o n 3. Papers reporting general research findings using variables that are so broad and loosely defined (e.g. tutoring versus non-tutoring) that r e p l i c a t i o n would be p r a c t i c a l l y impossible 4. Research studies with precisely defined variables (e.g., a p a r t i c u l a r approach to tutoring) that could be replicated - although they rarely are 5. Research that applies to distance 'learning' (sic) although not conducted with t h i s application in mind. 1 4 The l a s t i s p o t e n t i a l l y the l a r g e s t category of a l l . ' (p. 29) Coldeway a l s o noted some of the f a c t o r s l i m i t i n g the r o l e of r e s e a r c h i n t h i s a r e a : 1. E d u c a t i o n a l r e s e a r c h e r s are r a r e l y present d u r i n g the design of d i s t a n c e l e a r n i n g ( s i c ) systems 2. There i s no c l e a r paradigm for r e s e a r c h i n d i s t a n c e l e a r n i n g ( s i c ) and i t i s d i f f i c u l t to a t t r a c t funds to develop one 3. There have been no consumer groups or p u b l i c a t i o n o u t l e t s f o r such r e s e a r c h (although t h i s i s changing) 4. Some i n s t i t u t i o n s are averse to d e f i n i n g boundaries and v a r i a b l e s c l e a r l y s i n c e p r a c t i t i o n e r s work with macro-level v a r i a b l e s (e.g. t u t o r i n g ) and f e a r t h a t breaking them down i n t o components w i l l c o m p l i c a t e the phenomenon 5. E d u c a t i o n a l r e s e a r c h e r s o f t e n ask q u e s t i o n s of no p r a c t i c a l or even t h e o r e t i c a l r e l e v a n c e . The tendency to ask "What happens when you t r y t h i s ? " d i v e r t s them from the more important i s s u e of "How do you make t h i s happen?" 6. Researchers i n d i s t a n c e l e a r n i n g ( s i c ) t e s t v a r i a b l e s that are r e a l l y c l a s s e s of v a r i a b l e s (e.g., comparisons of d i s t a n c e and classroom l e a r n i n g ) . The r e s u l t s are impossible to r e p l i c a t e and dubious u t i l i t y anyway (p. 30) . 15 It i s evident from these reviews that future research should have well defined variables applicable to a wide range of situations and, at the same time, i n t u i t i v e l y simple so results could benefit other researchers and p r a c t i t i o n e r s . Research should also be r e p l i c a b l e . One should draw widely from the experiences of various countries and not be limited to national or regional p a r t i c u l a r i t i e s . In looking at the results of previous work, one must bear in mind that even when research is i n t e r n a l l y v a l i d , one cannot and should not assume there is external v a l i d i t y . They were c a r r i e d out at di f f e r e n t times, in di f f e r e n t contexts and with very d i f f e r e n t subjects. The extent to which results can be generalised i s usually limited. C h r o n o l o g y of Research on Completion Rate The t r a i l of research on completion rate in distance education i s quite extended and i t is interesting to follow i t s development in a chronological manner. Correspondence courses were very popular among the armed services in the U. S. during and shortly after the Second World War. Because of the large enrolment and low l e v e l of completion achieved in these programmes, an investigation was conducted by Bradt (1954) to determine why service personnel f a i l e d to complete United States Air Forces Institute courses. He blamed ' i n s u f f i c i e n t time' and the 'unsettled conditions' of students in the armed services. 16 One of the f i r s t doctoral dissertations devoted to the study of completion of correspondence study courses was by Hughes (1955) who t r i e d to determine the effects of certain factors on completion. He studied the methods of study, the purpose of enrolment, previous education, the need to f i n i s h a course by a d e f i n i t e date, and the effect of prior experience with correspondence study. The subjects were 441 students enrolled with the General Extension Division of the University of F l o r i d a . Through the use of a questionnaire, usable data were c o l l e c t e d from 249 respondents. The findings showed that completion rate was s i g n i f i c a n t l y related to 'having a regular^ study period', to the completion of work by a d e f i n i t e date, and with previous educational experience. The American National University Extension Association, through a research committee, examined completion rates for various types of students for d i f f e r e n t course areas and u n i v e r s i t i e s (Childs, 1956), and reported an o v e r a l l completion rate of 50 percent. The committee also discussed factors thought to have influenced non-completion - type of course, duration of course, goals of student and frequency of assignment submission. James and Wedemeyer (1959) were interested in the reasons students gave for non-completion. Using students enrolled in high school correspondence courses offered by the University of Wisconsin Extension Divi s i o n , they established 17 by case-interview method that students most frequently gave 'too busy', 'could not find time' and major l i f e events as reasons. They further suggested that the real reason might be 'lack of goal c l a r i t y . ' Childs (1963) at the University of Nebraska Extension D i v i s i o n , was interested in the effects of d i f f e r e n t i a l i n t e l l i g e n c e and aptitude on completion rate. The results f a i l e d to show any s i g n i f i c a n t relationship. However, i t was found that students registered for correspondence courses were above average in a b i l i t y and Childs speculated that the nature of courses might have a more pronounced effect on completion than either i n t e l l i g e n c e or aptitude. Spencer's (1965) study focused on the r e l a t i o n s h i p between motivation and completion. The subjects were students taking credit and non-credit correspondence courses at Pennsylvania State University. The results showed that motivation was important in influencing completion. Highly motivated students completed the courses in shorter periods and earned sati s f a c t o r y grades. It was also noted that men took more time to complete 2-credit and 3-credit courses and earned lower grades than women. Sloan (1965) studied the reasons students gave for non-completion at the University of Kentucky. With working adults in a sample of 135 students, he noted that the most frequently given reasons were (1) the job interfered with study, (2) lack, of time, (3) resident work interfered with 18 study and (4) loss of interest. The causes of d i s i n t e r e s t were given as lack of time, inadequate background in the course material, no instructor contact, too much work for the lessons and the interference of the job. Similar results were obtained by Pulley (1965) at the University of Missouri. He surveyed 243 students who had discontinued their correspondence study enrolment at that i n s t i t u t i o n and the same reasons for discontinuing were found. Donehower (1968) t r i e d to determine i f relationships existed between the degree of success in college l e v e l courses and student variables: age, sex and achievement as represented by grades, completion, withdrawal, previous education and time to submit the f i r s t assignment. The subjects were 905 college l e v e l correspondence students at the University of Nebraska. Twelve hypotheses were tested. r The findings demonstrated s i g n i f i c a n t positive correlations between a number of factors. These included the shorter the time students took to submit the f i r s t assignment and completion; previous education and achievement; previous education and completion; age and completion (older students more l i k e l y to complete), and sex and completion (women more l i k e l y to complete). There were no s i g n i f i c a n t correlations between time to complete and achievement, age and achievement, and sex and achievement. This study became one 19 of the most c i t e d references in a r t i c l e s concerned with drop-outs in correspondence study. P f e i f f e r and Sabers (1970) reported a study c a r r i e d out at the University of Iowa to determine why students enrolled in and why they did not complete correspondence study. The subjects included 6,539 students. There was a s i g n i f i c a n t relationship between completion and submission of the f i r s t assignment, between how early the assignments were submitted and grades. The longer the duration of the course, the lower were the grades. There was no s i g n i f i c a n t r e l a t i o n s h i p between completion and grades. Two reports emerged from a 1971 study conducted at Di v i s i o n of University Extension, University of I l l i n o i s , Urbana. The objective was to identi f y variables a f f e c t i n g the selection, completion of, and achievement in correspondence courses. A questionnaire was sent to 486 students of which 134 were returned. Students were asked about the d i f f i c u l t y of and interest in correspondence material, how they reacted to instructors' comments, whether they would take another course and had they made up a study schedule. The f i r s t report (Anderson & Tippy, 1971) showed that students rated comments.'from instructors as important. Few had a study schedule nor desired the D i v i s i o n to help them establish one. In the second report (Essex & Anderson, 1972), they recommended that personality variables, 20 motivation and perseverance, and turnaround should be included in future studies. Sabers, P f e i f f e r and Ragsdale (1972) used l e t t e r s and postcards of encouragement as e x t r i n s i c motivation and measured the change in the submission of assignments. It was the f i r s t study that looked at completion as 'the number of assignments submitted'. There were three groups of subjects, f i r s t - t i m e enrollees, those who had correspondence study experience with the i n s t i t u t i o n and inactive students already in a course and four weeks behind with their assignments. They found that inactive students increased the submission of assignments after receiving l e t t e r s of encouragement. On the other hand, l e t t e r s had no eff e c t on f i r s t - t i m e enrollees or those who had taken courses before. This was in contrast to the findings of P f e i f f e r and Sabers (1970) who did a similar experiment but found l e t t e r s of encouragement had no eff e c t on any group. Rekkedal (1972) studied the records of 1,417 students registered in correspondence courses at the Norsk Korrespondanseskole School in Norway. The focus was on recruitment, persistence, reasons for discontinuance and how they correlated with completion, discontinuance, active duration, number of lessons submitted, grades, and rates of submission. The concept of completion witnessed a further development here. To categorise various completion groups, Rekkedal used ' f u l l completion', 'active' (submitted at least 21 one lesson), 'stopped submission' and 'cancellation of r e g i s t r a t i o n ' . His population contained more men than women. It was found that the number of active students decreased in time, and the number of students discontinuing decreased with time. There was also a seasonal effect on discontinuance. The reasons for discontinuance given most frequently were 'lack of time' and 'changed to another mode of study'. Sig n i f i c a n t correlations were found between age and discontinuance (younger students were more l i k e l y to discontinue than older ones), more previous education had a positive correlation with achievement, more previous education had a positive affect on persistence, the shorter the time required to submit the f i r s t assignment and higher persistence, and the faster rate of submission with achievement. There were no s i g n i f i c a n t correlations between the rate of submission and age, previous education and rate of submission, duration of course and persistence and, most int e r e s t i n g l y , the rate of submission and persistence. Morstain (1974) t r i e d to determine i f there were certain distinguishable c h a r a c t e r i s t i c s of students who preferred an independent study programme when compared to those who preferred a more t r a d i t i o n a l (formal course) approach to education. A questionnaire was administered to 648 undergraduates at University of C a l i f o r n i a , Berkeley. No s i g n i f i c a n t differences in the preference for independent study were found when students were c l a s s i f i e d by sex or 22 entry l e v e l . However, when students were categorised by declared majors, s i g n i f i c a n t differences in the preference for independent study were determined. It was also found that correspondence students are more l i k e l y to support the idea that students rather than teachers should have prime r e s p o n s i b i l i t y in d i r e c t i n g their own studies. H a r t s e l l (1975) found in a simple study with 40 students at the University of Tennessee, Divi s i o n of Continuing Study, that there was no difference in f i n a l grades of students who submitted assignments on-time or l a t e . Baath (1975) carried out an experiment in Sweden to explore the use of submission density, defined as the r a t i o of the students' average number of study hours to the number of submissions in the course. This was another attempt to construct a measure to quantify student performance.. Submission density was found to vary most markedly between di f f e r e n t courses and between students from the same course. Although the idea of submission density has merits, information on student study hours are not easy to obtain and no researcher has applied i t to other studies. Another study concerned with completion and age was conducted by Bradley and Lehman (1975) who found that a higher proportion of younger as compared to older students tended to drop out. Kennedy and Powell (1976) studied reasons for dropping out at the B r i t i s h Open University among 684 students who 23 withdrew, 291 who p a r t i a l l y withdrew and 236 considered 'at r i s k ' . They proposed that student drop-out should be considered in the context of student c h a r a c t e r i s t i c s and circumstances. They found that those with lower previous education did not drop out for academic reasons. Wilson (1976), at the Sydney College of External Studies in A u s t r a l i a , t r i e d to reduce the number of non-starters of correspondence courses. A group of students were given Preliminary Unit Packs (PUP) containing detailed information about the course and a request to do a pre-course assignment. The idea was to screen out non-starters before the course commenced and thus cut down the unproductive work of registering those who would not s t a r t . It was found that non-starters were reduced to 14 percent for the group provided with PUP which contrasted favorably with the normal 37 percent. In comparing the two groups, i t was also noted that those provided with PUP were submitting assignments faster than the control group. Orton (1977) examined the completion and non-start rates for degree-credit correspondence courses offered by the Faculty of Arts and Science at Queen's University in Kingston between 1971 and 1975. Over four years, there were 5,333 regi s t r a t i o n s in 118 courses. The term completion, as used in this study, referred to a student whose r e g i s t r a t i o n was not cancelled before the f i n a l examination. Cancellations meant students who had formally cancelled th e i r r e g i s t r a t i o n . 24 A non-start was a student who cancelled r e g i s t r a t i o n before submitting any assignments, while a drop-out was a student who cancelled r e g i s t r a t i o n after submitting at least one assignment. It was found that completion rate at Queen's compared favorably with other u n i v e r s i t i e s . It was also concluded that completion rates did not change s i g n i f i c a n t l y over time, and that'there were s i g n i f i c a n t differences between four subject areas. Students were more l i k e l y to start submitting work i f registered in a Mathematics or Language course than in a Humanities or Social Science course. However, students were found to be more l i k e l y to complete a Social Science course.' Duby and Giltrow (1978) t r i e d to predict student withdrawal from the Chicago TV College. They defined course completion as 'achieving passing grade' at the end of a 15-week course. F a i l u r e s , incompletes and withdrawals were a l l considered to be withdrawals. Students were required to complete an evaluation questionnaire early in the course as one of their course assignments. From past records, they established baseline information on three main factors. They found that females completed courses more often than males, students from certain campuses consistently completed courses more often than those from others, and students returning completed questionnaires were more l i k e l y to complete courses than non-respondents. Two further secondary relationships were also noted: females who completed the questionnaire were 25 more l i k e l y to complete t h e i r courses than men who completed the q u e s t i o n n a i r e , and females who d i d not complete the q u e s t i o n n a i r e were more l i k e l y to complete t h e i r course than men who d i d not complete the q u e s t i o n n a i r e . With percentage values f o r previous semesters., they were able to develop a formula which s u c c e s s f u l l y p r e d i c t e d withdrawals f o r seven out of e i g h t courses w i t h i n 10 perce n t , and s i x out of e i g h t w i t h i n 6 p e r c e n t . The e i g h t courses ranged widely i n terms of course content and d e l i v e r y mode. The formula c o u l d be f u r t h e r r e f i n e d with the accumulation of more b a s e l i n e i n f o r m a t i o n every year. The r e s e a r c h e r s suggested that the p r e d i c t i o n c o u l d be used to i d e n t i f y , support, and thus reduce the withdrawal of 'at r i s k ' s t u d e n t s . In a r i g o r o u s study on correspondence education combined with systematic telephone t u t o r i n g , F l i n c k (1978) found that students had p o s i t i v e a t t i t u d e s towards the telephone. f Subjects (n=442) were two groups of a d u l t correspondence students i n Sweden and England whose schools were members of the European Home Study C o u n c i l . Two courses with high enrolment and completion r a t e were s e l e c t e d , Economics f o r Swedish students and French f o r E n g l i s h s t u d e n t s . An experimental design was used, and i t was concluded that s i g n i f i c a n t c o r r e l a t i o n s e x i s t e d between telephone t u t o r i n g and achievement with the French, course, completion and courses (higher completion i n language than i n economics) but 26 there was no s i g n i f i c a n t c orrelation between telephone tutoring and studying time with the French course. Hammer and Smith (1979) administered a questionnaire to students at Athabasca University. Students were asked about their reasons for taking a course, the extent to which the r e g i s t r a t i o n services were helpful and their suggestions for course improvements. Their responses were analyses in r e l a t i o n to their demographic c h a r a c t e r i s t i c s . Those who withdrew and those who completed were found to be s i g n i f i c a n t l y d i f f e r e n t only with respect to the r e g i s t r a t i o n service. Yet another study was made by Riches (1979) to determine the reasons for non-completion at the B. C. Institute of Technology. From 446 returned questionnaires, i t was concluded that non-persisters in longer courses dropped out early, and more students on these courses f a i l e d to submit the f i r s t lesson when compared with those on shorter courses. Consistent with previous studies, lack of time and the interference of the job were the most frequently given reasons for discontinuance. Wong and Wong (1979) investigated relationships between the number of assignments submitted by students, completion and f i n a l grades in a study involving subjects enrolled in a fundamental accounting correspondence course at the School of Continuing Education at the University of Toronto. Students who did more than half of the assignments were s i g n i f i c a n t l y 27 more l i k e l y to complete, and the more assignments they submitted, the higher t h e i r f i n a l examination grades tended to be. They suggested that completion of assignments and completion of course were both r e l a t e d to m o t i v a t i o n . There was a f e e l i n g of commitment and involvement i n the course a f t e r completing a few assignments. A well-planned s e r i e s of s t u d i e s was c a r r i e d out at Athabasca U n i v e r s i t y e n t i t l e d P r o j e c t REDEAL (Research and E v a l u a t i o n of D i s t a n c e Education f o r Adult L e a r n e r ) . T h i s began by e s t a b l i s h i n g a t h r e e - p a r t a n a l y t i c a l base that attempted to bridge the gap between the need f o r p r a c t i c a l i n f o r m a t i o n and the d e s i r e to c o n t r i b u t e to a comprehensive t h e o r e t i c a l framework. Coldeway and Spencer (1980a) d i s c u s s e d some of the problems a s s o c i a t e d with the measurement of completion of d i s t a n c e education c o u r s e s . I t was noted that there was g e n e r a l l y no agreement on the f o r m u l a t i o n of completion. T o t a l enrolment formula and the NUEA formula had both been commonly used. There has a l s o been the a d d i t i o n a l problem of q u i t e d i s p a r a t e d e f i n i t i o n s on what c o n s t i t u t e completion, non-start and withdrawal. As completion data were u s u a l l y obtained a f t e r the course had been completed, changes would have taken p l a c e and meaningful comparison would be d i f f i c u l t to make from s e s s i o n to s e s s i o n . For d i s t a n c e education systems which have d i f f e r e n t s t a r t i n g dates f o r each student, and/or no due dates f o r the 28 submission of assignments, measuring completion would be even more d i f f i c u l t . Another report (Spencer, 1980) discussed the use of a computer to generate in d i v i d u a l i s e d study schedules for home study courses. Results showed that while most learners opted to receive such a schedule, they did not l i k e using them. Few followed the schedules and i t appeared to have no effect on completion rates. The more quantitative study in the series (Coldeway & Spencer, 1980b) investigated factors a f f e c t i n g learner motivation, attribute and performance. Four courses offered by Athabasca University with 556 students were selected for an experiment. Twelve hypotheses were tested. S t a t i s t i c a l values are conspicuously absent from the report but i t appeared that learners completing courses had higher 'previous' examination marks than non-completers; women completed more often than men; the age of learner had no bearing on completion (noting that 63 percent of the to t a l sample were under 35 years old); learners with more previous education had s l i g h t l y higher 'rates of completion than those with less; learners who indicated a preference to learn on their own did not have a higher completion rate than those in groups or a combination of other situations-; learners indicated that they anticipated more than.10 hours available each week for studying had a s l i g h t l y higher completion rate than those who anticipated less; learners enrolled for degree 29 programmes did not have a higher completion rate than those not enrolled for degree programmes; learners successfully completing a course had a higher completion rate in subsequent courses and learners enrolled in more than one course did not have a higher completion rate than those in a single course. D i S i l v e s t r o and Markowitz (1982) were interested in the problem of motivating correspondence students, at a distance, to begin promptly and continue to course completion. They set up a study to investigate the use of learning contracts and their effect on lesson submission and completion rates at the Independent Study Division of Indiana University. The population consisted of 430 students in mixed high school and univer s i t y groups. Individuals from each group were randomly assigned to two contract groups and two non-contract groups. The contract stipulated that the student agreed to submit the f i r s t lesson no later than three weeks after receiving the materials, and the second lesson no later than three weeks after completion of the f i r s t lesson. The Independent D i v i s i o n , in return, agreed to return the students' lessons within two and half weeks. There was a s i g n i f i c a n t difference between the contract and non-contract groups, evident in both the university and high school population to begin work on their courses. However, there was no s i g n i f i c a n t difference between the contract and non-contract groups on course completion. It was concluded that contracts 30 had an immediate effect on lesson submission, but not a long-term effect on course completion. Reporting on a study completed in 1973, Rekkedal (1983) examined the effect on the number of assignments submitted and completion by reducing turnaround. Subjects were 127 students at the Norwegian NKI-skolen. Students were randomly divided into two groups, one group experienced 'quick' and the other 'delayed' turnaround. When assignments were received from students on a certa i n day, the tutor corrected a l l assignments. The corrected assignments for the quick group were posted right away and the delayed group had theirs held back for three days before they were posted. The 'quick' turnaround group.had a higher completion rate and more assignments were submitted during the f i r s t three months of enrolment than the 'slow' group. However, for those who completed the work, turnaround had no apparent ef f e c t on the time taken to complete the course and with the f i n a l grade. Smith (1983) conducted a study at Saddleback College to evaluate the retention, achievement, and s a t i s f a c t i o n of students enrolled in the college's telecourse programme. The population included a l l telecourse and on-campus students in four courses offered at the same time on-campus and through telecourses. No differences were found between the personal and academic c h a r a c t e r i s t i c s of telecourse completers and non-completers. But i t found that completers had a preference to study alone, tended to be working towards a 31 degree or for career advancement and had previously taken telecourses. Phythian and Clements (1982) examined the reasons for student dropout from t h i r d l e v e l mathematics courses at the B r i t i s h Open University. Of the 180 withdrawals studied, 50 percent gave personal reasons including "domestic" and "job" factors and 33 percent attributed withdrawal to academic reasons including "courses too hard and too long." Woodley and Parlett (1983) conducted a comprehensive analysis on student records at the B r i t i s h Open University from 1971 to 1982. Non-completion of f i n a l r e g i s t r a t i o n (formally registered as a student) was not featured in the figures. Dropout, or 'overall wastage' as they were c a l l e d , included a l l o f f i c i a l withdrawals and f a i l u r e s . Dropout occurred most frequently at Level 1 and Level 2 (where there are three leve l s for general degree and four le v e l s for the r honours degree) with the highest dropout rate at Level 2. Overall, dropout rate was higher for mathematics and technology courses and r e l a t i v e l y lower for s o c i a l sciences. Men were more l i k e l y to drop out (among new and continuing students) than were women; for new students, the very young and the very old were more l i k e l y to drop out while the least l i k e l y was the 30-39 year old age group. For continuing students there was no difference for the whole age range of 21-59 and the 60-64 age group was least l i k e l y to drop out; low previous education q u a l i f i c a t i o n s was more l i k e l y to drop 32 out and t h i s was especially true for new students; dropout rate was high for students in manual occupation, who were re t i r e d or unemployed and very high for those in i n s t i t u t i o n s ; there was a strong rel a t i o n s h i p between the number of credits held by continuing students at the beginning of the year and their chance' of gaining some cr e d i t during the year; the longer students were with the Open University, the less l i k e l y they were to graduate, and dropout rate increased for those with high workload (credit units) except for those taking h a l f - c r e d i t which had an even higher dropout rate (a normal workload varies between one to two c r e d i t s ) . In a survey of the records of 6,000 students at the Pennsylvania State University Independent Learning service, Chalon ('1985) proposed an evaluation model to improve effectiveness in distance courses. Working with eleven variables: completion rate, passing rate per course, grades, average time to complete course, average time to dropout, previous education, instruction q u a l i t y , guidebook qualit y , workload, months of faculty time working in the programme and quality of communication, and establishing the s i g n i f i c a n t correlations between these variables, a predictor model was constructed. The model was used to predict passing rate per course, completion rate and grades. 33 Summary While the chronology has shown that completion rate was central to many studies and can be related to a variety of factors and variables, t h i s study was only concerned with the relationships between the four variables: completion, grades, deviation and turnaround (the outcome of programmes) and other variables (age, sex, education, occupation, duration, subject, fee and language of i n s t r u c t i o n ) . The relevant findings are tabulated in Table 1, 2 and 3 to show the state of research up to now. These tables only show the general d i r e c t i o n of relationships. For every study, most variables were defined in a di f f e r e n t manner. The f i r s t impression looking at the tables i s the gaping holes in them. Although completion rate, grades, deviation and turnaround are basic information commonly found in the record of most distance education programmes, i t i s surprising that so l i t t l e investigation has taken place to examine their i n t e r - r e l a t i o n s h i p s to themselves, and to student and programme variables. A pattern which has c l e a r l y emerged from the tables is the s i m i l a r i t y of some of the variab l e s . A variable was termed one thing at one stage of the course and termed another at another stage. If some of these variables which were describing the same phenomenon at d i f f e r e n t stages of the course can be consolidated into one variable, the tables 34 Table 1 Previous findings on relationships between grades,  completion, and deviation and student socio-demoqraphic c h a r a c t e r i s t i c s GRADES COMPLETION DEVIATION AGE Not s i g n i f i c a n t Donehower, -1968 H a r t s e l l , 1974 P o s i t i v e Donehower, 1968 Rekkedal, 1972 Bradley & Lehman, 1 975 Not s i g n i f i c a n t Coldeway et a l , 1 980 Smith, 1983 Not s i g n i f i c a n t Rekkedal, 1972 SEX Female hig h e r Spencer, 1965 Donehower, 1968 H a r t s e l l , 1974 Female hig h e r Donehower, 1968 Duby & Giltrow, 1 978 Coldeway et a l , 1 980 Woodley & Parlett 1 983 Not s i g n i f i c a n t Smith, 1983 EDUCATION LEVEL P o s i t i v e Donehower, 1968 Rekkedal, 1972 P o s i t i v e Donehower, 1968 Rekkedal, 1972 Coldeway et a l , 1 980 Not s i g n i f i c a n t Rekkedal, 1972 OCCUPATIONAL STATUS Not s i g n i f i c a n t Smith, 1983 S i g n i f i c a n t Woodley & Parl e t t 1 983 35 Table 2 Previous findings on relationships between grades,  completion, deviation and other outcome variables GRADES COMPLETION DEVIATION COMPLETION OF PRE-COURSE ASSIGNMENT S i g n i f i c a n t Wilson, 1976 COMPLETION OF FIRST ASSIGNMENT S i g n i f i c a n t P f e i f f e r & Sabers, 1 970 NUMBER OF ASSIGNMENTS SUBMITTED P o s i t i v e S i g n i f i c a n t Wong & Wong, 1979 Wong & Wong, 1979 TURNAROUND Negative Rekkedal, 1983 DEVIATION OF FIRST ASSIGNMENT Not s i g n i f i c a n t Negative Donehower, 1968 Donehower, 1968 Negative P f e i f f e r & Sabers P f e i f f e r & Sabers, 1970 1970 Rekkedal, 1972 DEVIATION P o s i t i v e Not s i g n i f i c a n t Not a p p l i c a b l e Rekkedal, 1972 Rekkedal, 1972 Not s i g n i f i c a n t H a r t s e l l , 1974 GRADES Not a p p l i c a b l e Not s i g n i f i c a n t Not a p p l i c a b l e P f e i f f e r & Sabers, 1 970 36 Table 3 •Previous findings on relationships between grades,  completion, deviation and programme variables GRADES COMPLETION DEVIATION DURATION Negative Negat i v e Not s i g n i f i c a n t P f e i f f e r & Sabers, Riches, 1979 Rekkedal, 1972 1970 SUBJECT(NATURE) S i g n i f i c a n t Childs, 1963 Orton, 1977 FEE LANGUAGE OF INSTRUCTION w i l l be less cluttered and i t i s much easier to compare res u l t s . For example, completion of the f i r s t assignment, number of assignments submitted and course completion are components of completion rate. It can then be further distinguished as completion rate at the beginning of the course, at certain points during the course and at the end of the course. S i m i l a r l y , deviation of the f i r s t assignment, of some of the assignments and of a l l the assignments can be grouped as deviation and further sub-divided as deviation at the beginning, during and at end of the course. Not many studies have been replicated. This i s partly due to the fact the measures used were not compatible or 37 comparable. What was used i n one study was not r e a d i l y a v a i l a b l e i n another s i t u a t i o n . Common v a r i a b l e s which are present i n most d i s t a n c e education programmes should be developed to v e r i f y previous f i n d i n g s and to f a c i l i t a t e f u r t h e r research i n v o l v i n g other f a c t o r s . T h i s review has shown that not much s u b s t a n t i a l research has taken p l a c e i n d i s t a n c e e d u c a t i o n . Of the r e s e a r c h conducted, the r e s u l t s are d i f f i c u l t to compare and r e p l i c a t e . L i m i t a t i o n s a r i s i n g out of a lack of e x t e r n a l v a l i d i t y prevent us from drawing too many c o n c l u s i o n s from them. Desperately l a c k i n g are d e f i n i t i o n s of new v a r i a b l e s which are i n t u i t i v e l y simple and dynamic (ranges throughout the d u r a t i o n of the programme and u n i v e r s a l l y a v a i l a b l e i n most d i s t a n c e programmes) and can be a p p l i e d to most d i s t a n c e e d u c a t i o n o p e r a t i o n s . T h i s would minimize the l i m i t a t i o n s and a l l o w us to make meaningful comparisons. 38 CHAPTER THREE VARIABLES One of the purposes of this study was to create or derive from existing records of a distance education programme four components that comprise programme outcome and to examine their conceptual and psychometric properties. These four components or variables were completion rate, grades, deviation and turnaround. Each of them w i l l be defined in this chapter. The other purpose was to examine how these variables relate to the independent variables (student and programme varia b l e s ) , the l a t t e r w i l l be examined f i r s t . Independent V a r i a b l e s A group of independent variables were associated with the socio-demographic p r o f i l e of students; these were termed 'student variables'. These included age, sex, education and occupation and they were obtained when students f i l l e d out enrolment forms. The second group of independent variables related to the nature of the course selected; these were termed 'programme variables' and included the fee, language of i n s t r u c t i o n , duration and nature of course. Each of these variables are described below and summarised in Table 4 . 39 Age This was a continuous variable. The age range of the student varied from the teens to over 60 years old. On the enrolment form, the students f i l l e d in their actual age in years. The f i n a l coding for age was categorised into ten age groups with a step of five years for each group. Sex This was a nominal variable. In the i n i t i a l coding, F stood for female and M stood for male. Later the coding was changed to 0 for female and 1 for male. Educat ion This was an ordinal variable. Students were asked to indicate their educational l e v e l by choosing from one of the fi v e categories: primary, secondary, post-secondary, university and graduate school. I n i t i a l coding used the characters PR, SE, PS, UN and GS, these were later recoded in numerically 1, 2, 3, 4 and 5. Occupat ion Occupation was an nominal variable. Students had to choose from one of the following eight occupational groups. These were student, c l e r i c a l , technical, s o c i a l work, education, commercial and professional and others. The choice of grouping might not be ideal but that was how the Department categorised students. I n i t i a l coding used the characters OT, ST, TE, SW; ED, CO, PF, these were later recoded in numerals as 1, 2, 3, 4, 5, 6, 7 and 8. 40 Fees The fees varied with the choice of courses. The range for course fees was between 90 and 160 Hong Kong Dollars. It was a continuous variable. Language of instruction The language of instruction varied from course to course but i t was either Chinese or English. If Chinese was the language of instruction, a l l communications, correspondence material, assignments and corrections were conducted in Chinese, likewise the language for communication for courses where the language of instruction was English was, of course, English. As a variable, language of instruction was nominal in nature. The f i n a l coding for the two languages was 1 for Chinese and 2 for English. Duration The duration of the courses varied. The shortest course contained six units and the longest had 21 units. Students were allowed two weeks to complete an assignment for each unit, so a 6-unit course would last for twelve weeks. For the purpose of th i s study, duration was measured in units and not in weeks. Duration was a continuous variable. Nature of course The courses were offered without labels i . e . courses were not presented as academic courses or otherwise. However, for the programme planner, the courses were categorised into language, vocational, enrichment and 41 academic c o u r s e s . The f i n a l c o d i n g f o r t h e s e c a t e g o r i e s was 1, 2, 3, and 4 r e s p e c t i v e l y . Nature of c o u r s e was a nominal var i a b l e . T a b l e 4 Independent v a r i a b l e s used i n a study of d i s t a n c e e d u c a t i o n outcome VARIABLES TYPE RANGE UNITS Age Cont inuous 0 t o 9 year groups Sex Nominal 0, 1 E d u c a t i o n O r d i n a l 1 t o 5 O c c u p a t i o n Nominal 1 t o 8 Fees Cont inuous 90 t o 160 HK$ Language Nominal 1, 2 Durat i o n Cont inuous 6 t o 21 u n i t s N a t ure of c o u r s e Nominal 1 t o 4 Dependent V a r i a b l e s The dependent v a r i a b l e s i n t h i s s tudy were those t h a t d e s c r i b e d the programme outcome. They were c o m p l e t i o n r a t e , s t a n d a r d i s e d grade, d e v i a t i o n , and t u r n a r o u n d . Each i s d e f i n e d below. S t a n d a r d i s e d grade S t a n d a r d i s e d grade (sg) e q u a l l e d the mean of grades over the number of assignments handed i n d i v i d e d by 100. When 4 2 assignments were marked, grades were given i n i n t e g e r s , always a m u l t i p l e of f i v e , and ranged from 50 to 100. • Let g i = grade f o r assignment i s g i = (g1 + g2 + .... + g i ) / a s i * l 0 0 where a s i was the number of assignments handed i n a f t e r i u n i t s of the course. St a n d a r d i s e d grade e q u a l l e d the grade p o i n t average (GPA) d i v i d e d by the maximum p o s s i b l e grade ( f u l l grade, i n t h i s case 100) and was always l e s s than or equal to 1. s g i = GPA(i)/Max(Grade) i f GPA = Max(Grade), s g i =1 i f GPA = 50, sg = 0.5 i f GPA < Max(Grade), s g i < 1 and s g i was the s t a n d a r d i s e d grade a f t e r i u n i t s of the c o u r s e . sg denoted the s t a n d a r d i s e d grade f o r the whole course and sg = (g1 + g2 + ... + g21)/(tassgn * 100) where tassgn was the t o t a l number of assignments submitted and g i was set to 0 i f assignment i was not submitted. For example, i f a student obtained 80 marks in the f i r s t assignment and 60 i n the second, then sg1 = .80 and sg2 = .70. I f the above student d i d not hand i n any more assignments, the s t a n d a r d i s e d grade f o r the whole course would remain at .70 because only assignments handed i n would 43 a f f e c t the s t a n d a r d i s e d grade. If a student's grades improved as the course progressed, the value of s g i would go up, s i m i l a r l y s g i would go down when the grades d e t e r i o r a t e d so that s g i was a true i n d i c a t o r of the grades of the student. Completion r a t e Completion rate (cr) e q u a l l e d the t o t a l number of assignments handed in d i v i d e d by the t o t a l number of assignments r e q u i r e d to be submitted. Let a s i = t o t a l number of assignments handed i n up t i l l u n i t i , and l e t r a s i = t o t a l number of assignments that were r e q u i r e d to be submitted up t i l l u n i t i , normally r a s i = i s i n c e every u n i t of every course r e q u i r e d that one assignment be submitted. then c r i = completion r a t e up t i l l u n i t i c r i = a s i / i and cr = completion r a t e f o r the whole course cr = tassgn / number of u n i t s f o r the course where tassgn was the t o t a l number of assignments submitted at the end of the course. At any p o i n t ( u n i t p o i n t ) of the course, i f a s i = i , c r = 1 , a l l the assignments were handed i n i f a s i < i then c r < 1 , some assignments were handed in but not a l l . 4 4 Since the number of assignments handed i n c o u l d never exceed the number of assignments r e q u i r e d to be handed i n , c r would never be g r e a t e r than 1. If a s i = 0, c r i = 0, no assignments were handed i n . If a student had handed in the f i r s t two assignments and not the r e s t of say twelve assignments, then cr1=1, and cr2=1, cr3=.66, cr4=.50 ... cr12= .16. Again, f o r a course of twelve assignments, i f a student had submitted the f i r s t two assignments and the 5th and the 6th but not the r e s t , then cr1=1, cr2=1, cr3=.66, cr4=.50, cr5=.60, cr6=.67 ... c r l 2 = .33. For the student who handed in four assignments out of the r e q u i r e d twelve, c r = cr12 = .33, the completion r a t e f o r the course. D e v i a t i o n D e v i a t i o n (ad) was d e f i n e d as the number of days that separated the date of handing i n from the due date of the assignment. As a l l courses were s t a r t e d at the same time, and the students were to hand i n one assignment once every two weeks, the due date f o r the handing in of the f i r s t assignment was two weeks or f o u r t e e n days a f t e r the beginning of the course. Subsequent due dates f o r the r e s t of the u n i t s were m u l t i p l e s of f o u r t e e n days from the beginning of the course. T h i s was t r u e f o r a l l students and c o u r s e s . The only d i f f e r e n c e among the courses i n t h i s r e s p e c t was the t o t a l number of u n i t s i n each course. 4 5 Let dvi = 'date in' for assignment i - 'due date' for assignment i , then dvi = deviation from the due date when assignment i was handed in (in days) i f dvi = 0, the assignment was on time i f dvi > 0, the assignment was behind i f dvi < 0, the assignment was ahead. Let adi = the average deviation after assignment i , then adi = (dv1 + dv2 + .... + d v i ) / a s i where asi was the actual number of assignments handed in up t i l l unit i of the course. . If a student had not handed in a p a r t i c u l a r assignment, say assignment i , then dvi was set to 0. ad = average deviation for the whole course and ad = dv1 + dv2 + ... + dvt / tassgn where dvt was the deviation for the l a s t assignment and tassgn was the t o t a l number of assignments submitted. For example, i f a student handed in the f i r s t assignment two days behind schedule,.the second assignment eight days behind, then ad1 = 2/1 = 2, ad2 = 10/2 = 5 i . e . the student was f i v e days behind on average aft e r two units. If the above student did not submit any more assignments after the f i r s t two units, then ad3, ad4, ... ad = 5. This was the case because average deviation only measured the deviation whenever an assignment was handed in, i t did not r e f l e c t the number of assignments handed i n . However, i f a 46 student got progressively behind schedule, the dvi's associated with his or her later assignments would increase and the average deviation would also increase. So the average deviation varied from unit to unit and would give a dynamic measure of the amount of deviation. If a student was ahead a l l the time, the average deviation would be negative. Turnaround Turnaround (at) was the time in days between the date when the student sent in the assignment and the date when the corrected assignment was returned from the Department. The record of assignments used for this study provided the dates showing when the assignments were received from the student and the dates when the corrected assignments were returned to the student. Two mailings were involved, once from the student to the Department and once from the Department to the student. The t o t a l time involved was two and a half days on average. Given that the time required for postal delivery was one day for 90 percent of the mailing and two days for 99 percent of the mailing ( i . e . the mailing time were constant), for the purpose of t h i s study the two and a half days mailing time were ignored because a l l the figures for turnaround were equally o f f . As only correlation analyses would be performed the results would not be affected. Let a t i = (tr1 + tr2 + ... + t r i ) / asi a t i = average turnaround up t i l l unit i 47 where t r i was the turnaround for assignment i and asi was the t o t a l number of assignment submitted up t i l l unit i . and at = (tr1 + tr2 + ... +trt) / tassgn at = average turnaround for the whole course where t r t was the turnaround for the l a s t assignment and tassgn was the t o t a l number of assignments submitted. t r i was always p o s i t i v e and greater than 1 as i t always took longer than one day to correct an assignment. The value of t r i was not affected by an assignment not submitted as t r i was set to 0 when that assignment was not received. For example, i f a student handed in two assignments, the f i r s t assignment took twelve days to correct and the second six days, then at1 = 12/1 = 12 and at2 = (12 + 6)/2 = 9. If the student did not hand in any more assignments then at = 9, the average turnaround for the whole course. Choice of Variables In previous studies of completion rate in distance education programmes, two methods of measuring completion have been used most commonly. They were the Total Enrolment Formula cr = (course completions/total enrolment)/100 where course completions = the number of students who have completed the course and the National University Extension Association (NUEA) Formula 48 cr = (course completions/total enrolment - t o t a l non-starts)/l00 where t o t a l non-starts = number of students who never handed in any assignments. With either formula, students were grouped as completers or non-completers. Those who have completed more than 80 percent of the course were put into the same category as those who have only completed ten percent of the course. This occurred because these formulae were meant for administrative and accounting rather than research or evaluation purposes. The derived variables used in t h i s study had a number of advantages over previous measures. F i r s t , each was an independent measure of a d i f f e r e n t phenomenon and there was no overlap. Second, there was a precise value for each variable at each stage of the course and for every student. Each was a continuous variable which might prove to be advantageous when analysed in r e l a t i o n to other variables and lent i t s e l f to parametric analyses. On the other hand, each variable could also be comprehended in an i n t u i t i v e manner. Completion rate was' simply the percentage of required assignments submitted at any point in the course expressed in decimal form. Standardised grade was the cumulative average grade achieved at any point in the course expressed in decimal form. Deviation was the cumulative average delay of assignment 49 submission at any point in the course expressed in days. Turnaround was the cumulative average time to mark and return assignments by the of f e r i n g i n s t i t u t i o n expressed in days. So the variables defined corresponded to conventional understanding, and the dimensions used had a natural association with them. Completion rate, grade, and deviation, in combination and in the absence of t u t o r i a l support and required examinations and tests as i t was in t h i s case, provided a comprehensive p r o f i l e of student performance at each stage of the course. Turnaround attested to the performance of the offering i n s t i t u t i o n . The four variables, considered together, quantify programme outcome. The raw data that comprised these variables are kept by most distance education programmes so future studies can make use of these variables for comparison among di f f e r e n t programmes and i n s t i t u t i o n s . The variables also r e f l e c t student behaviour so in future, they could be used as tools to study many aspects of distance education. Summary Each of the derived, student and programme variables was presented in this chapter. The advantages of variables defined in this manner were discussed. The next chapter describes how data from an exis t i n g record of a distance education programme were transformed to derive values for 50 these v a r i a b l e s . CHAPTER FOUR METHOD The psychometric part of th i s study involved an ex post facto analysis of extant data of a distance education programme. Unlike studies where 'new' data i s c o l l e c t e d according to established methodologies, the empirical part of this investigation r e l i e d e n t i r e l y on existing data. Consequently, the methods used in t h i s investigation primarily concerned the transformation of raw data. This chapter describes the data source, i t s capturing, transportation, transformation and v e r i f i c a t i o n . Data Source The data for th i s study were the student records of the 52nd term of Correspondence Courses offered by the Department of Extramural Studies, The Chinese University of Hong Kong. Enrolment for the- courses started on July 1, 1984 and the courses commenced on August 1, 1984. When students enrolled, they completed enrolment forms (see Figure 1) which included the usual items l i k e name, address, telephone numbers, age, sex, education background, occupation and the courses selected. Two of these were 52 Name: ENROLMENT FORM Sex: Age: Address: T e l : (Work) EDUCATION Primary Secondary Post-secondary OCCUPATION Professional C l e r i c a l Education Commercial No. (Residence) University Grad. School Technical Social Work Student Others Fee Course T i t l e Total Date: Sign: Figure 1. Enrolment form used by students enrolled at the"Chinese University 53 multiple choice items. With regard to education background, students chose from primary, secondary, post-secondary, university and graduate l e v e l s . There were eight choices offered for occupation - professional, commercial, educational, social work, technical, c l e r i c a l , student and others. When the forms were received, each student was assigned a unique seven d i g i t student number. If a student enrolled in two courses, he or she would have received two d i f f e r e n t numbers. The f i r s t two d i g i t s of the student number sp e c i f i e d the term and, in t h i s case, a l l the numbers began with 52. The thi r d and fourth d i g i t s related to the course and were the same as the course number advertised in the course brochure (see Appendix 1). The last three d i g i t s of the student number constituted the student number for each course, so the 21th student enrolled in Course No. 7 would have the number 5207021. Data C a p t u r i n g At the Department, a microcomputer and customized student record software were i n s t a l l e d to keep records e l e c t r o n i c a l l y . The software had three input subroutines. The f i r s t f a c i l i t a t e d input of d e t a i l s of courses offered for that session (see Figure 2). The t i t l e , course number, number of assignments, fees, and the commencement date of the courses were entered for each course. The second subroutine 54 CUSR(C52) COURSE TITLE: CHILD DEVELOPMENT & CARE DATE :09/06/84 NUMBER: 5207 NO. OF SESSIONS: 7 FEE: HK$ 90.00 DURATION MONTH: 3 COMMENCEMENT DATE: 01/08 NO.OF STUDENTS: 54 Figure 2. Computer screen for input of course records at the Chinese University of Hong Kong allowed enrolment information to be keyboarded into an enrolment f i l e and the t h i r d f a c i l i t a t e d the input of information related to course assignments. The enrolment subroutine s o l i c i t e d a l l the information from the enrolment form and, in addition, assigned a student number to each enrolment (see Figure 3). The assignment subroutine designated a page of assignment record to each student who handed in one or more assignments. When the operator entered an assignment record (see Figure 4), the software asked for the student number, when that was entered, the page of assignment record associated with the p a r t i c u l a r student was brought onto the screen. On each page, there was a l i n e for each assignment 55 CUSR(C52) ENROLMENT DATE :07/06/84 NAME:CHAN TAI MAN SEX:M AGE:24_ YR. OF BIRTH: 1960 ADDRESS:RM 2316, PEONY HOUSE TEL.:5-242071 WONG TAI SIN EST. TEL. :3~37021 1 KOWLOON  EDUCATION(PR/SE/PS/UN/GS):SE OCCUPATION(PF/CL/ED/CO/TE/SW/ST/OT):CL REMARKS:********** COURSE* :5_207 TITLE: CHI LP DEV CARE STUDENT#: 5207021 # OF ASSIGNMENT:? FEE: HK$ 90.00 DATE: 07/06/84 SUGGESTION: REMARKS: Figure 3. Computer screen for input of enrolment record that the student might hand i n . And on each l i n e , the date when the assignment was received from the student, the grade obtained, and the date when the assignment was returned to the student could be recorded. The assignment records could be updated in piecemeal fashion. The 'rece ived' date could be entered on i t s own as the other information would not be a v a i l a b l e when the assignment was f i r s t received. The 'returned' date and the grade for a p a r t i c u l a r assignment, however, were normally entered together by the operator as assignments were returned once they were recorded. When enrolment forms were received at the beginning of 56 CUSR(C52) ASSIGNMENT DATE :09/06/84 STUDENT#/PG#:5207021/1 COURSE TITLE: CHILD DEV CARE NAME: CHAN TAI MAN AGE:24 EDU.:SE OCC.:CL ASSIGNMENT* DATE RECEIVED GRADE DATE RETURNED £3/09 75_ 13/10 26/09 75_ 06/11 15/10 80 26/11 ~~ / r~ ~ / ~ ~ / ~ ~ / / z / _ _ _ / _ / / #ASS.REC.:03 ~~~/ /~ ~7 / _f~ ~/~ z / _ _ / _ REMARKS: Figure 4. Computer screen for input of assignment record. each session, they were quickly entered into the computer so mail labels could be printed. The input of assignments, however, offered no immediate benefit in terms of saving of work, and i t was easier to record the information on forms by hand. The p a r t i c u l a r s of assignments were then transferred in bulk whenever there was a l u l l in work. In either case, there was an electronic as well as a hardcopy record for v e r i f i c a t i o n . T r a n s p o r t a t i o n of Data A further function of the student record software was the export of data in a form of a f i l e (ASCII or text f i l e ) 57 which could be read on screen and by other common business and s t a t i s t i c a l software. Two types of data were exported in th i s manner. The enrolment data was exported to an enrolment f i l e where each l i n e represented a student enrolment record consisting of eight f i e l d s , the unique student number, name, address, telephones, age, sex, education and occupation. In th i s case the f i l e contained 1731 records. For the purpose of t h i s study, the name, address and telephones were redundant as well as the blank columns that separated the various f i e l d s . These were edited out on a microcomputer using word processing software with a column delete f a c i l i t y . What was'left (see Figure 5) were the seven d i g i t student number from which the session number and the course number could be i d e n t i f i e d , sex (M or F occupying another two columns), the age (positive integers, occupying two columns), education and occupation (in character codes and each taking up two columns). The f i l e containing the assignment records were s i m i l a r l y exported and each l i n e contained a l l the assignment records of one student. For a student who completed a l l assignments, the li n e would start with the student number, the name, the day and month of the receipt of the f i r s t assignment, the grade for that assignment, the day and month of the return of the assignment and s i m i l a r l y for a l l subsequent units (see Figure 6 in thi s chapter). For a student who had not handed in the l a s t three assignments, 5 8 there would be blanks under those columns. If a student missed a few intermediate assignments, blanks would appear in those columns but subsequent assignments would stay in their appropriate positions. In e f f e c t , the assignment f i l e only contained records of students who had handed in one or more assignments. There were 773 such students. For each, the enrolment record was prefixed with the student number and the assignment record would also be s i m i l a r l y prefixed with the same unique number. These two f i l e s were of unequal lengths. The enrolment f i l e contained 1731 records and the assignment f i l e contained 773 records. The enrolment f i l e had a uniform width with missing variables from a few records (see Figure 5). The assignment f i l e had an uneven width, and was dependent on the number of assignments handed in (see Figure 6). There were gaping blanks where a few assignments were not handed in followed by more assignment entries. However, the two f i l e s had a common f i e l d , the student number which would be used to lin k the two parts of each student's record together. Data Transformation The following i s a summary of the raw data obtained from the two f i l e s . Student number 7 d i g i t s , f i r s t 2 d i g i t s denoted the term, the second 2 d i g i t s denoted the 5 9 Sex Age Educat ion Occupat ion Day in Month in Grade Day out Month out course, together the f i r s t 4 d i g i t s denoted a unique course number 1 - character code, (F, M) 2- d i g i t p o s i t i v e integer 2-character code for 8 groups 2-character code for 8 groups 2-digit positive integer (21 units) 2-digit p o s i t i v e integer (21 units) 2-digit p o s i t i v e integer (45-95, step 5, 21 units) 2-digit p o s i t i v e integer (21 units) 2-digit p o s i t i v e integer (21 units) To prepare the data for analysis, a number of transformations were required. F i r s t , the character codes had to be converted into numerals. Second, the f i r s t four d i g i t s of the student number had to be picked up so that the course of each of the students could be i d e n t i f i e d and the duration, language of ins t r u c t i o n , fees and nature of the course could be c o r r e c t l y designated. Third, new variables, 'date in' and 'date out' were reconstructed from the day i n , month in, day out and month out by designating the correct year to make up the complete year, month and day f i e l d for each assignment. As each of the character codes for education and occupation was unique, i t was replaced with pos i t i v e integers using the 'search and replace' function of a word processor. The coding of the fiv e l e v e l s of education became 01 to 05 and the eight 60 d i f f e r e n t choices of occupation became 01 to 08. The coding of sex became 0 for female and 1 for male. Age was recoded into 10 age groups using data base software. The f i n a l version of the enrolment f i l e thus has 14 columns. Columns 1 and 2 represented the session number, columns 3 and 4 the course number, columns 1 to 7 the student number, column 8 the sex, columns 9 and 10 the age, columns 11 and 12 the education level and columns 13 and 14 the occupation. Whenever there were missing variables a blank would occur this would be treated d i f f e r e n t l y from zeros. 12345678901234 12345678901234 5207001F21SEED 52070011020206 5207002F24PSCO 52070021020307 5207003F31SECO 52070031020207 5207004F SEPF 52070041 0208 5207005M22SEOT 52070052020201 5207006F31SEPF 52070061040208 5207007F28SEPF 52070071030208 5207008F29SEOT 52070081030201 5207009F26SECL 52070091030203 Figure 5. Sample of enrolment records before and after transformat ion. Cn the assignment f i l e , the f i r s t four d i g i t s of the student number were read using data base software and the appropriate course d e t a i l s were designated separate f i e l d s for each record. These include duration (ranged from 49 to 147 days), language of instruction (Chinese 1, English 2), 61 fees (ranged from HK$90 to HK$160) and nature of course (Language 1, Vocational 2, Enrichment 3, and Academic 4). The creation of the new variables, date in and date out for each assignment, was also done using data base software. F i r s t , the assignment f i l e (see Figure 6) with over 200 columns was edited with data manipulation software on the UBC mainframe computer. As the structure of th i s f i l e contained many blank columns and redundant characters, e.g. there was a '/' between date and month for each received and returned dates, these were removed and only relevant columns retained. 5207004 NAME 03/09 75 1 3/1 1 26/09 75 06/03 26/09 75 06/03 5207005 NAME 02/10 60 24/1 0 5207006 NAME 26/08 75 24/1 0 5207016 NAME 17/10 75 06/03 5207020 NAME 10/08 75 1 3/1 1 1 0/08 75 1 3/1 1 5207022 NAME 13/09 70 24/1 0 5207023 NAME 13/09 75 1 3/1 1 13/09 75 13/1 1 5207033 NAME 03/09 75 1 3/1 1 5207046 NAME 03/09 75 1 3/1 1 03/09 75 13/1 1 5207054 NAME 03/09 75 1 3/1 1 03/09 75 13/1 1 1 7/09 75 06/03 5207057 NAME 26/09 75 06/03 Figure 6. Section of assignment f i l e before transformation (long lines truncated to 70 columns) If there were missing values, these remained as blanks throughout the whole manipulation process. Student names were also removed as they served no useful purpose in subsequent analysis. What remained (see Figure 7) on each record (line) was column 1 to 7, the student number, columns 8 to 11 the received date for the f i r s t assignment, columns 62 12 to 13 the grade for that assignment, columns 14 to 17 the returned date for the same assignment and the same ten column cycles for subsequent assignments. Whenever there were missing assignments, blanks would appear. The assignment f i l e d i f f e r e d from the enrolment f i l e in that for each enrolment there was always a record of the student, but i f he or she had not handed in any assignment, there would be no record of this student in the assignment f i l e . 12345678901234567890123 456789012345678901234 567890123456789 520700403097513112 60975060326097 506031010752805 52070050210602410 52070062608752410 52070161710750603 52070201008751311100875131 1 52070221309702410 520702313097513111309751311 5207033030975131 1 520704603097513110309751311 5207054030975131103097513111709750603 52070572609750603 Figure 7. Section of assignment f i l e a f t e r transformation Using appropriate software, the student number, and the day i n , month i n , grade, day out, month out of each assignment were read into a data base f i l e . The f i e l d s defined in t h i s data base f i l e were numeric for the student number and the grade for each assignment and date f i e l d s for the date in and date out (each requiring day, month and year). The year attached to each date f i e l d was 1984, the sta r t i n g year, whenever the month was more than 8, the 63 s t a r t i n g month. The date f i e l d was assigned 1985 whenever the month was less than 8 which would correspond to an assignment handed in more than four months into the course. As due dates for each of the assignments of a l l the courses were the same, due dates for each assignment were designated, After the preceding transformations the second generation data consisted of the following. Student number 7 d i g i t s , remained unchanged Sex Age Education Occupat ion Date in Date out Due dates Durat ion Language Fees Nature Grades 1 d i g i t , (0, 1) 1 d i g i t , 10 age groups (0-9) 1'digit, 5 education lev e l s (1-5) 1 d i g i t , 8 occupation groups (1-8) dd/mm/yy (21 units) dd/mm/yy (21 units) dd/mm/yy (21 units) 2 d i g i t s , (in units) 1 d i g i t , (Chinese 1, English 2) 3 d i g i t s , (90 to 200 Hong Kong dollars) 1 digit,(Language 1, Vocational 2, Enrichment 3, Academic 4) 2 d i g i t s , remained unchanged, (21 units) Verif ication The two data f i l e s (enrolment and assignment) were thus ready for checking. Frequencies for each variable were 64 obtained with the SPSS ( S t a t i s t i c a l Package for Social Scientists) software.- It revealed the t o t a l number of missing values in the variables. Since there were less than five missing values in the enrolment f i l e , each missing variable was checked against the o r i g i n a l enrolment form. No errors were detected. The student record software also had the capacity to do frequencies analysis on enrolment for each course showing the number of males and females, their age, education and occupation groups. Two errors were discovered and corrected. The operator of the student record system had entered two persons twice. With the unique student number, the two extra entries could also be i d e n t i f i e d in the assignment f i l e . In fact, the two persons did not hand in any assignment, and they had no record in the assignment f i l e and no corresponding change needed to be made. Other figures were r checked against the frequency tests from SPSS and they were a l l perfect matches. The checking of the assignment f i l e required a more elaborate procedure. The received dates were subtracted from returned dates. Negative or zero values would indicate errors in data entry. Several such errors were discovered, the mistakes were corrected against the assignment forms. A search was also made to seek out records on the assignment f i l e which had either one or two of the three f i e l d s missing. These might be the ' i n ' dates, the grades, or the 'out' 65 dates. When such errors were revealed, i t seemed that the operator had missed entering some data. They were again checked against the assignment forms and corrections were subsequently made to the assignment f i l e . After a l l the changes were made, the same tests were repeated to make sure that the corrections were correct and no additional errors had been introduced. F u r t h e r Transformations The data contained in the assignment f i l e were kept in a data base f i l e for easy manipulation. It was also possible to export an ASCII f i l e for checking and analysis by SPSS at any time. From t h i s f i l e , deviations from due dates and turnarounds (time taken for the assignments to be corrected) for each unit and the t o t a l number of assignments handed in for each student were computed. The following were the new variables derived. dvi = deviation from due date for unit i = date in for unit i - due date for unit i t r i = turnaround (correction time) for unit i . = date out for unit i - date in for unit i tassgn = t o t a l number of assignment handed in = count for grades not equal 0 As the tutors doing the correction never assigned 0 to an assignment, any assignment handed in would attract a grade 6 6 greater than zero so that the t o t a l assignment count would y i e l d the correct number of assignments handed i n . A l l other variables'remained unchanged in thi s transformation. A further check for unreasonable or i l l o g i c a l values of dv and t r were made and any unusual occurrences were compared with the assignment forms and corrected i f the i n i t i a l keyboarding was found to be wrong. As the dates i n , dates out and due dates would not feature in further computation, they were dropped from the latest version of the f i l e . To prepare the f i l e for s t a t i s t i c a l analysis, new variables were again created and unnecessary ones dropped. The following variables were created in the l a s t transformation. asi = number of assignments handed in up to unit i = assignment count up to unit i for grade not equal to zero ad = average deviation = sum(dv1 to dv21)/tassgn dvi= 0 for assignment not handed in and for i greater than the number of assignments required for the course adi = average deviation at unit i = sum(dv1 to dv i ) / a s i cr = average completion rate = tassgn/total number of assignments required for the course c r i = completion rate at unit i 67 = asi/number of assignments due up to unit i at = average turnaround = sum(tr1 to tr2l)/tassgn tri=0 for assignment not handed in and for i greater than the number of assignments required for the course a t i = sum(tr1 to t r i ) / a s i sg = standardised grade = sum(g1 to g21)/(tassgn * 100) sgi = 0 for assignment not handed in and i greater than the number of assignments required for the course sgi = sum(g1 to g i ) / a s i * 100) After the preceding procedure, dv1 to dv2l, tr1 to t r 2 l and the grades g1 to g21 were dropped and the remaining variables saved. These variables remaining were 1. Given variables (some were recoded) Student number 7 d i g i t s Sex 1 d i g i t , (0, 1) Age 1 d i g i t , 10 age groups (0-9) Education 1 d i g i t , 5 education levels (1-5) Occupation 1 d i g i t , 8 occupation groups (1-8) Duration 2 d i g i t s , (in units) Language 1 d i g i t , (Chinese 1, English 2) Fees 3 d i g i t s , (90 to 200 Hong Kong dollars) Nature 1 d i g i t , (Language 1, Vocational 2, Enrichment 3, Academic 4) 2. Derived variables 68 tassgn Total assignments handed i n , (1 to 21) as1 to as21 number of assignments handed in up to that unit (positive integer, 1 to 21) cr average completion rate for whole course, (positive r e a l number between 0 and 1) cr1 to cr21 average completion rate up to that unit (positive real number between 0 and 1) ad average deviation for whole course (real numbers, days) ad1 to ad21 average deviation up to that unit (positive real numbers, days) at average turnaround for whole course (real number, days) at1 to at21 average turnaround up to that unit (positive real numbers, days) sg average standardised grade for whole course (positive real number between 0 and 1) sg1 to sg21 average standardised grade up to that unit (positive real number between 0 and 1) These were the given and derived variables submitted for s t a t i s t i c a l analysis. 69 CHAPTER FIVE DATA ANALYSIS S e l e c t i o n of A n a l y t i c a l Procedures The focus of th i s study was to examine relationships between independent and dependent variables and between various components which comprised the dependent variables. The independent variables included ( 1 ) student variables: age, sex, education l e v e l , and occupation; and ( 2 ) programme variables: fees, language of instruction, duration and nature of course. The independent variables were divided into three groups. F i r s t , those variables which were continuous or o r d i n a l . These included age, education l e v e l of students, duration and fees. The second group consisted of nominal variabl e s : occupation and nature of course. The t h i r d group of variables were dichotomous - sex and language of in s t r u c t i o n of the course. . The dependent variables were a l l derived and continuous variables. These were completion rate, deviation, turnaround and standardised grades. To examine the relationship between the f i r s t group of independent variables and the four dependent variables, Pearson product-moment correlations were computed. Between the nominal variables and continuous dependent variables, one-way analysis of variance was used. Between the 70 dichotornous variables and dependent variables, the T-test was applied. The relations between the dependent variables themselves were also of interest and as a l l of them were continuous, Pearson's correlations were computed for the six relations between the four variables. Procedures 1. Frequencies tests were applied to the age, sex, education, occupation of the 773 students, and to duration, language of instruction, fee, and nature of courses, and the number of assignments submitted and student enrolment in each course. The result showed the p r o f i l e s of the students and courses. 2. Pearson's correlation analyses were performed on age, education, duration and fee with completion rate, deviation, turnaround and standardised grade. 3. One-way analyses of variance were applied on occupation, nature of course with completion rate, deviation, turnaround and standardised grade. 4. T-test analyses were applied on language of in s t r u c t i o n , sex with completion rate, deviation, turnaround and standardised grade. 5. Pearson's correlation analyses were performance on completion rate, deviation, turnaround and standardised grade. 71 F u r t h e r Procedures Analyses over time Procedures 2 to 4 were conducted to examine relationships between the independent and dependent variables and between the dependent variables themselves for a l l 773 cases, hereafter to be known as analyses on a l l cases. However, the dependent variables were formulated so they ref l e c t e d the progress of students at various stages of the course. To take advantage of t h i s , a similar set of procedures were performed at quarter-points into the courses and together with the f i r s t set of procedures, t h i s provided analyses at 25, 50 and 75 percent into and at the end of the course. Performing analyses at quarter points into the courses offered a cross-sectional view of the progress of students during the course and showed changing relationships between a l l the variables. However, as courses were of d i f f e r e n t durations and varied between six and 21 units, ( i . e . 12 to 42 weeks), percentages into the courses were only approximations as shown in Table 5. Analyses across d i f f e r e n t groups By selecting groups of students who completed d i f f e r e n t numbers of assignments and applying the same set of a n a l y t i c a l procedures, another view of how d i f f e r e n t groups of students performed was available. For example, students 72 Table 5 Mapping courses of d i f f e r e n t duration (in units) into four parts Quarter points into course Number of courses Durat ion 1/4 2/4 3/4 4/4 3 6 2 3 4 6 2 7 2 4 5 7 5 10 3 5 8 1 0 7 1 2 3 6 9 1 2 1 1 3 3 6 9 1 3 1 1 6 4 8 1 2 1 6 1 20 5 10 1 5 20 1 21 5 10 5 21 who submitted very few assignments might have had very di f f e r e n t deviation values when compared with those who had submitted many assignments or when compared with the student population as a whole. Beginning with students who had handed in two assignments or less, followed by four assignments or less, and six assignments or less, the same set of a n a l y t i c a l procedures was applied. These groups were considered to have submitted very few assignments. By selecting students who had submitted six or more, eight or more, ten or more, and twelve or more assignments, the same set of a n a l y t i c a l procedures were again applied. 73 These groups were considered to have submitted many assignments. These two groupings (students who had submitted very few or many assignments) were made without regard to the duration of the course. A student who had completed six assignments for a course with only six units had a completion rate of 1, but a student who had completed six assignments for a 12-unit course only had a completion rate of .50. Results from analyses across these groupings would be d i f f e r e n t . For example, a l l those who had completed six assignments might have had very similar deviation values when compared to those who completed only two assignments. Inter-correlations between derived variables One further set of analyses was also performed. This concerned the relationships between completion rate, deviation, turnaround and standardised grade at 25, 50, and 75 percent into the courses and those same variables at 100 percent into the courses. As a l l these' variables were continuous, Pearson's c o r r e l a t i o n tests were applied and thi s would show us how soon and how close the values of the variables would approximate their terminal values. Summary To summarise, analyses were performed as follows: 1. Frequencies tests on age, sex, education, occupation, 74 duration, language, fee, nature of course, number of assignments submitted and student enrolment for each course. 2. Pearson's cor r e l a t i o n analyses on age, education, duration and fee with completion rate, deviation, turnaround and standardised grade at 25 percent, 50 percent, 75 percent, 100 percent into the courses, and for the groups of students who had submitted two assignments or less, four assignments or less, six assignments or less, and for the groups of students who had submitted six assignments or more, eight assignments or more, ten assignments or more and twelve assignments or more. 3. One-way analyses of variance on occupation and nature of course with completion rate, deviation, turnaround and standardised grade at 25, 50, 75 and 100 percent into the courses, and for the groups of students who had submitted two assignments or les s , four assignments or les s , six assignments or les s , and for the groups of students who had submitted six assignments or more, eight assignments or more, ten assignments or more and twelve assignments or more. 4. T-test'analyses on language and sex with completion rate, deviation, turnaround and standardised grade at 25, 50, 75 and 100 percent into the courses, and for the groups of students who had submitted two assignments or less, four assignments or less, six assignments or les s , and for the groups of students who had submitted six assignments or more, 75 eight assignments or more, ten assignments or more and twelve assignments or more. 5. Pearson's cor r e l a t i o n analyses on completion rate, deviation, turnaround and standardised grade at 25, 50, 75 and 100 percent into the courses, for the groups of students who had submitted two assignments or less, four assignments or l e s s , six assignments or less, and for the groups of students who had submitted six assignments or more, eight assignments or more, ten assignments or more and twelve assignments or more. 6. Pearson's cor r e l a t i o n analyses on completion rate, deviation, turnaround and standardised grade at 25, 50 and 75 percent into the courses with the same variables at the end of the course. However, in terms of off e r i n g d i f f e r e n t views on the variables, there were five groups of procedures: (1) frequencies tests for student and programme variables, (2) a n a l y t i c a l tests for a l l cases ( i . e . 100 percent into the courses and a l l values of the variables were at their terminal values), (3) a n a l y t i c a l tests across time, (4) a n a l y t i c a l tests across d i f f e r e n t groups determined by the number of assignments submitted, (5) cor r e l a t i o n tests for the values of the four derived variables at d i f f e r e n t stages of the course with their own terminal values. 76 CHAPTER SIX RESULTS I The q u e s t i o n s posed i n Chapter One were answered by performing a s e r i e s of s t a t i s t i c a l procedures on the data. There were f i v e groups of procedures as f o l l o w s , (1) The c h a r a c t e r i s t i c s of p a r t i c i p a n t s and the programme were a s c e r t a i n e d by running SPSS f r e q u e n c i e s t e s t s . (2) The r e l a t i o n s h i p s between the dependent (derived) v a r i a b l e s ( i . e . completion r a t e , d e v i a t i o n , grades and turnaround) and the independent' v a r i a b l e s ( i . e . age, educatio n * d u r a t i o n , fee, o c c u p a t i o n , nature of course, language of i n s t r u c t i o n and sex) at the end of the course when a l l v a l u e s of the v a r i a b l e s were at t h e i r t e r m i n a l v a l u e s were e s t a b l i s h e d by running ANOVA, Pearson's c o r r e l a t i o n and the g e n e r a t i o n o f . t - v a l u e s . (3) The same a n a l y t i c a l t e s t s were a p p l i e d at q u a r t e r p o i n t s i n t o the co u r s e s . (4) The same a n a l y t i c a l t e s t s a c r o s s d i f f e r e n t groups determined by the number of assignments submitted were a p p l i e d . (5) C o r r e l a t i o n t e s t s f o r the values of the four d e r i v e d v a r i a b l e s at d i f f e r e n t stages of the course with t h e i r own t e r m i n a l v a l u e s were performed. 77 C h a r a c t e r i s t i c s of P a r t i c i p a n t s and Programme It can be seen from Table 6 that of the nine age groups, the largest involved participants between the age of 20 to 24 years. In fact, 85 percent of the participants in the programme were under 30 years old. This was consistent with Table 6 Age groups occupied by 773 students in the programme Cum Value Label Value n Percent Percent <=19 1 210 27.2 27.2 20-24 2 301 38.9 66. 1 25-29 •3 1 46 18.9 85.0 30-34 4 62 8.0 93.0 35-39 5 29 3.8 96.8 40-44 6 1 1 1 .4 98.2 45-49 7 6 .8 99.0 = >55 9 8 1 .0 100.0 Total 773 1 00.0 1 00.0 V a l i d Cases 773 Mi ssing Cases 0 the most active age group who p a r t i c i p a t e in adult education in Hong Kong. They have started work for a few years, are aware of their need for more education and most would not have started their own families yet and would have time for education. Another factor which could account for the high percentage of young participants (27.2 percent were under 19 years old) was that the programme was offered in the summer and this seasonal high number of secondary students occurred 78 every year. Only 7 percent of the participants were over 30 years old. Table 7 shows the number of female and male par t i c i p a n t s . There were more female (55.1 percent) than male (44.9 percent) p a r t i c i p a n t s . This was consistent with Table 7 Number and percent of male and female students in the programme Value Label Value n Percent Cum Percent Female Male 0 1 426 347 55. 1 44.9 55.1 100.0 Total 773 100.0 Va l i d Cases 773 Missing Cases 0 the o v e r a l l enrolment in the Department (over 60 percent of the 40,000 enrolment were taking face-to-face courses). The courses offered were promoted as being for those who had completed secondary education so i t was not surprising that the largest group were those who had a secondary school le v e l of education. Table 8 shows that t h i s group accounted for 76.8 percent of the pa r t i c i p a n t s . On the other hand, although there was no admission requirement for any course, there were only four participants who had no more than a primary school education. Post-secondary education in 79 Table 8 Educational l e v e l of students in the programme Cum Value Label Value n Percent Percent Primary 1 4 .5 .5 Secondary 2 594 76.8 77.4 Post-secondary 3 1 27 16.4 93.8 University 4 46 6.0 99.7 Graduate Study 5 2 .3 100.0 Total 773 1 00.0 Va l i d Cases 773 Missing Cases 0 Hong Kong i s normally interpreted to mean various forms of technical, teacher and nurse t r a i n i n g . Those with post-secondary education background accounted for 16.4 percent of the population. Together with those with secondary education, they comprised over 92 percent of the population. In the enrolment form, the categories of education l e v e l were not s t r i c t l y delineated, for example, for secondary l e v e l , someone who had attended secondary school, irrespective of the exit point (anything between Form 1 and Form 5) would belong to thi s category, other categories were also defect ive. Table 9 shows participants in various occupation groups. Usually, the largest occupation group had been the c l e r i c a l workers. However, the programme was offered in the summer which had a history of attracting a large number of senior secondary school students who believed that taking 80 Table 9 Occupational groups of students in the programme Cum Value Label Value n Percent Percent Others 1 1 34 17.3 17.3 Student 2 214 27.7 45.0 C l e r i c a l 3 180 23.3 68.3 Technical 4 59 7.6 75.9 Social Work 5 1 1 1 .4 77.4 Education 6 49 6.3 83.7 Commerc i a l 7 62 8.0 91 .7 Professional 8 64 8.3 1 00.0 Total 773 100.0 V a l i d Cases 773 Missing Cases 0 courses in the summer with a university extension department would improve their academic work. On the other hand, many c l e r i c a l workers would be taking th e i r holidays in the summer and they were outnumbered by students in t h i s session. Although the duration of the courses varied between six and 21 units, i t can be seen from Table 10 that the popular courses were mostly of ten and twelve units duration. Participants taking these courses accounted for 65.5 percent of the population examined. There were only 14.1 percent taking short courses (six and seven u n i t s ) , and 5.8 percent taking long courses (20 and 21 u n i t s ) . The great majority (80 percent) were taking courses between ten and sixteen units which lasted between twenty and 32 weeks. 81 Table 10 Courses in the programme grouped by duration Cum Value Label n Percent Percent 6-unit course 74 9.6 9.6 7-unit course 35 4.5 14.1 10-unit course 247 32.0 46. 1 12-unit course 259 33.5 79.6 13-unit course 27 3.5 83. 1 16-unit course 86 11.1 94.2 20-unit course 1 0 1 .3 95.5 21-unit course 35 4.5 100.0 Total 773 100.0 Va l i d Cases 773 Mi ssing Cases 0 The most popular were language courses as can be seen from Table 13. Most language courses were about and taught in English. This accounted for the more than half the courses conducted in English (see Table 11). .The courses with Chinese as the language of instruction consisted of Chinese Language and Lit e r a t u r e , enrichment and two vocational courses. Table 11 Number and percent of students in the programme grouped by  language of instruction Cum Value Label Value n Percent Percent Chinese 1 322 41 .7 41 .7 English 2 451 58.3 100.0 Total 773 100.0 V a l i d Cases 773 Missing Cases 0 82 The course fees varied between HK$90 to HK$160. There were four which charged $140, a l l of them popular English Language courses, which would account for the high Table 12 Number and percent of students in the programme grouped by  fees paid Cum Value Label n Percent Percent Course fee $90 1 56 20.2 20.2 Course fee $95 39 5.0 25.2 Course fee $100 1 4 1 .8 27.0 Course fee $ 110 73 9.4 36.5 Course fee $120 32 4. 1 40.6 Course fee $135 61 7.9 48.5 Course fee $140 306 39.6 88. 1 Course fee $150 20 2.6 90.7 Course fee $160 72 9.3 100.0 Total 773 1 00.0 V a l i d Cases 773 Mi ssing Cases 0 39.6 percent for participants paying $140 (see Table 12). Again, there were two popular courses at $90 and would account for the high 20.2 percent' of t o t a l . As can be seen from Table 13, the most popular courses were in languages, both the English and Chinese ones which comprised 78.8 percent of the student population examined. How the courses were categorised can be seen in Appendix 1. Table 14 grouped the students by the number of assignments submitted irrespective of the duration of course selected. The shortest course required the submission of six 83 Table 13 Number and percent of students in the programme grouped by nature of course Cum Value Label Value n Percent Percent Language 1 609 78.8 78.8 Vocational 2 62 8.0 86.8 Enr ichment 3 72 9.3 96. 1 Academic 4 30 3.9 1 00.0 Total 773 1 00.0 Va l i d Cases 773 Missing Cases 0 Table 14 Number and percent of students in the programme grouped by  number of assignments submitted Value Label n Percent Cum Percent 1 assignment 133 17. .2 17. 2 2 assignments 247 32, .0 49. 2 3 assignments 64 8, .3 57. 4 4 assignments 95 12, .3 69. 7 5 assignments 33 4 , .3 74. 0 6 assignments 68 8, .8 82. 8 7 assignments 22 2, .8 85. 6 8 assignments 1 9 2, .5 88. 1 9 assignments 1 1 1 , .4 89'. 5 10 assignments 23 3, .0 92. 5 11 assignments 18 2, .3 94. 8 12 assignments 25 3, .2 98. 1 13 assignments 1 . 1 98. 2 14 assignments 1 . 1 98. 3 15 assignments 3 .4 98. 7 17 assignments 3 .4 99. 1 18 assignments 2 .3 99. 4 19 assignments 1 . 1 99. 5 20 assignments 4 .5 100. 0 Total 773 100, .0 V a l i d Cases 773 Missing Cases 0 84 assignments while the longest required 2 1 . Nearly half the students submitted two assignments or less and 6 9 . 7 percent submitted four or I G S S (see Table 1 4 ) . It i s evident from these figures that most students f a i l e d to complete their courses and most of them gave up at a f a i r l y early stage. Table 14 also shows there were few who submitted more than thirteen assignments. That i s to be expected since, as shown in Table 10 , 83 .1 percent of the students were taking courses with thirteen units or le s s . Table 15 shows the number of students (who handed in one assignment or more) enrolled for each course offered in the 1984 summer distance education programme at the Chinese University of Hong Kong. There were four popular courses, each accounted for more than ten percent of the population examined while none of the other courses exceeded 5 . 2 percent. Information concerning the fees, duration and language of instruction for each of these courses can be found in Appendix 1. 85 Table 15 Enrolment f o r each course of the 1984 summer programme at the  Chinese U n i v e r s i t y of Hong Kong Value Label n Percent Cum Percent Chinese Language 94 12. ,2 12. ,2 P r a c t i c a l Chinese W r i t i n g 20 2. ,6 14. ,7 Approach to Ch. Poetry 19 2. ,5 17. ,2 T r a n s l a t i o n from E to C 34 4. ,4 21 . ,6 Business A d m i n i s t r a t i o n 27 3. ,5 25. , 1 P r i n c i p l e s of Economics 20 2. ,6 27. ,7 C h i l d Dev and Care 9 1 . ,2 28. ,8 P r i n . of 2-D Design 17 2. ,2 31 . ,0 Chinese P a i n t i n g 14 1 . 8 32. ,9 Prob. i n E n g l i s h for Asians 1 03 13. .3 46. ,2 E f f . E n g l i s h W r i t i n g 97 12. .5 58. ,7 E n g l i s h Comprehension 16 2. . 1 60. ,8 Advance E n g l i s h W r i t i n g 27 3, .5 64, .3 E n g l i s h Vocabulary What's wrong with your Eng. 40 5, .2 69, .5 86 1 1 , . 1 80, .6 I n t r o . Bus. Correspondence 27 3, .5 84, . 1 I n t r o to Mod. Maths 10 1 . .3 85, .4 Business Correspondence 26 3, .4 88, .7 Idiom-Magic 20 2, .6 91 , .3 Monochrome Drawing 32 4, . 1 95, .5 Gen. P r i n . of Eng. Law 35 4, .5 100, .0 T o t a l 773 100, .0 V a l i d Cases 773 M i s s i n g Cases 0 R e l a t i o n s h i p s b e t w e e n s t u d e n t and programme v a r i a b l e s and 'outcome' Table 16 shows the r e s u l t s of a p p l y i n g s t a t i s t i c a l procedures to determine the r e l a t i o n s h i p s between the d e r i v e d v a r i a b l e s : completion r a t e , d e v i a t i o n , grades and turnaround and the independent v a r i a b l e s : age, e d u c a t i o n , d u r a t i o n , fee, occupation, nature of course, language of i n s t r u c t i o n and sex 86 at the end of the course ( i . e . a l l values of the variables were at their terminal values). As we are examining the relationships between the derived (dependent) and the independent variables, the independent variables are l i s t e d down and the derived variables across the page in Table 16. The descriptive s t a t i s t i c s were l i s t e d at the bottom of the table. Students were on average submitting 36 percent of the required assignments (X=.36, S.D.=.28). The average age of students f e l l within the 20-24 and 25-29 age groups (X=2.36, S.D.=1.39). Their average education l e v e l , was between the secondary and post-secondary education groups (X=2.29, S.D.=.59). The average duration of course taken was just over eleven units or very near the very popular 10 to 12-unit courses (X=11.56, S.D.=3.46). The average fees paid by students were about HK$125 (X=124.98, S.D. 23.54). On average, students were almost three weeks behind in submitting assignments for every unit (X=19.95, S.D.=29.33). The average time taken by the Department to mark and return assignments was almost fourteen weeks (X=97.06, S.D.=86.34). Students achieved an average grade of 73 percent of the f u l l mark (X=.73, S.D.=.09). We now examine the results of correlation analyses between age, education, duration and fee and the derived variables. In Table 16, we can see that age correlated s i g n i f i c a n t l y with completion rate (r=.08, p <.01); older 87 Table 16 A n a l y s i s f o r a l l cases, n=773 Pearson's Complet ion C o r r e l a t i o n D e v i a t i o n Turnaround Grades Age .08* .05* -.07 .10* Education .04 -.08 .10 . 1 5** Durat ion -.11** -.12** -.15** .28** Fee -.12** -.01 . 28** .01 One-way Occupation N.S. N.S. F=2.57 p <.05 F=6.26 p <. 01 Nature F=3.83 p <.01 F=2.74 p <.05 F=37.27 p <.01 F=22.38 p <.01 T-Test Language N.S. N.S. t=-25.05 p < .01 N.S. Sex N.S. N.S. N.S. N.S. * p < . 0 1 , * * 2 < -001 c r age educ dura fee ad at sg Mean 0.36 2.36 2.29 11.55 124.98 19.95 97.06 0.73 S.D. 0.28 1.39 0.59 • 3.46 23.54 29.33 86.34 0.09 students were sub m i t t i n g more assignments than younger ones. D u r a t i o n c o r r e l a t e d s i g n i f i c a n t l y with completion r a t e (r=-.11, p <.001 ) ; s h o r t e r courses had higher completion r a t e s . Completion r a t e c o r r e l a t e d s i g n i f i c a n t l y with fee (r=-.12, p <.001 ); l e s s expensive courses had a higher completion r a t e than more expensive ones. D e v i a t i o n c o r r e l a t e d s i g n i f i c a n t l y 88 w i t h age (r=.05, p <.01); o l d e r s t u d e n t s took s l i g h t l y l o n g e r t o submit assignments than younger ones. D e v i a t i o n c o r r e l a t e d s i g n i f i c a n t l y w i t h d u r a t i o n (r=-.12, p <.001); s t u d e n t s on s h o r t e r c o u r s e s took l o n g e r t o submit assignments than s t u d e n t s on l o n g e r c o u r s e s . Age c o r r e l a t e d s i g n i f i c a n t l y w i t h grades ( r = . l 0 , p <.01); o l d e r s t u d e n t s o b t a i n e d h i g h e r grades than younger ones. Grades c o r r e l a t e d s i g n i f i c a n t l y w i t h e d u c a t i o n ( r = . l 5 , p <.001); s t u d e n t s w i t h a h i g h e r e d u c a t i o n a l background o b t a i n e d h i g h e r grades than those w i t h l o w e r . Grades c o r r e l a t e d s i g n i f i c a n t l y w i t h d u r a t i o n (r=.28, p <.001); s t u d e n t s on l o n g e r c o u r s e s s c o r e d h i g h e r than those on s h o r t e r ones,. I t i s i n t e r e s t i n g t o note t h a t e d u c a t i o n a l background d i d not c o r r e l a t e s i g n i f i c a n t l y w i t h c o m p l e t i o n r a t e or d e v i a t i o n . D i f f e r e n t o c c u p a t i o n groups d i d not have s i g n i f i c a n t c o m p l e t i o n r a t e s nor d e v i a t i o n and none of the groups completed s i g n i f i c a n t l y more ass i g n m e n t s nor took s i g n i f i c a n t l y l o n g e r t o submit them than any o t h e r group. The groups, however, s c o r e d s i g n i f i c a n t l y d i f f e r e n t g r ades, (F=6.26, p <.01). When grouped by n a t u r e of c o u r s e , p a r t i c i p a n t s had s i g n i f i c a n t l y d i f f e r e n t c o m p l e t i o n r a t e s (F=3.83, p <.01), d e v i a t i o n s (F=2.74, p <.05), and grades (F=22.38, p <.01). I t took the Department s i g n i f i c a n t l y l o n g e r t o r e t u r n a s s i g n m e n t s t o s t u d e n t s f o r the c o u r s e s conducted i n E n g l i s h (t=-25.05, p <.01) than those i n C h i n e s e . Men and women d i d 8 9 not achieve s i g n i f i c a n t l y d i f f e r e n t completion rate, deviation or grades. The results obtained has shown that sex had no s i g n i f i c a n t c o r r e l a t i o n with student performance. Age was associated with completion rate, deviation and grades. Education had no s i g n i f i c a n t c o r r e l a t i o n with completion rate and deviation but correlated s i g n i f i c a n t l y with grades. Duration had s i g n i f i c a n t association with completion rate, deviation and grades. Fee has s i g n i f i c a n t association with completion rate but not with deviation and grades. Occupation was s i g n i f i c a n t l y associated with grades but not completion rate or deviation. Nature of course had s i g n i f i c a n t influence on a l l three. Language of instruction had no s i g n i f i c a n t association with any of the four derived var iables. Although a number of conditions of the hypothesis have been s a t i s f i e d , the hypothesis could only be accepted i f a l l the conditions are s a t i s f i e d , which i s not the case here. Hypothesis 4 was rejected. R e l a t i o n s h i p s between V a r i a b l e s D e s c r i b i n g 'Outcome' Recall that one purpose of t h i s study was to examine the relationship between d i f f e r e n t elements (completion rate, deviation, grades and turnaround) that usually comprise the 'outcome' sought in distance education. Table 17 shows the inter-relationships between the elements in the form of a 9 0 Table 17 Correlation (Pearson product--moment) between distance education ' outcome' elements Completion Deviation Turnaround Grades Complet ion 1 .00 Deviation -.09* 1 .00 Turnaround -.14** .03 1.00 Grades -.04 -.02 -.16** 1.00 * 2 <.01, ** £ <.001 corr e l a t i o n matrix. Completion rate correlated s i g n i f i c a n t l y with deviation (r=-.09, p <.01); those completing more assignments took shorter time to submit them than those completing fewer assignments. Completion rate correlated s i g n i f i c a n t l y with turnaround (r = -.14, p <.001); those who experienced shorter turnaround completed more assignments than those who experienced longer turnaround. There was no s i g n i f i c a n t correlation between grades and completion, between deviation and turnaround, and between deviation and grades. There was s i g n i f i c a n t c o r r e l a t i o n between turnaround and grades (r=-.16, p <.001); those who experienced shorter turnaround had better grades than those experienced long turnaround. In a follow-up to this study, i t w i l l be necessary to factor analyse this correlation matrix to ascertain the extent to which common variance underlies these relationships. If a general factor pervades the matrix, i t 91 might be possible to drop one or more of the outcome elements from future analyses. However, note the lack of i n t e r -c o r r e l a t i o n between three pairs of derived variables. In other words, deviation i s not a redundant exemplar of turnaround or grades. A n a l y t i c a l T e s t s Across Time and Groups Appendices 2 to 4 tabulates the results of performing the same a n a l y t i c a l tests at quarter points into the courses. Appendices 5 to 11 tabulates the results of administering the same set of a n a l y t i c a l tests for d i f f e r e n t groups of students determined by the number of assignments submitted. These results might be relevant to other researchers and are included in the appendices. 92 CHAPTER SEVEN RESULTS I I Of the fiv e sets of results obtained from the a n a l y t i c a l tests, four were presented in Chapter Six. However, the res u l t s of the tests at the various stages of the course were d i f f i c u l t to compare as tabulated. Bearing in mind that Hypotheses 1 to 3 a l l involved comparisons over time, they w i l l be re-tabulated here across time and in pairs of variables pertinent to what was stated in the hypotheses i . e . between completion and turnaround, and between completion and deviation. This w i l l help us to determine whether Hypotheses 1 to 3 are to be rejected. Hypothesis 4 had already been rejected. The next set of results concerned correlations between the four derived variables at d i f f e r e n t stages of the course with their own terminal values. The l a s t set of res u l t s related to the time and 'variable' e f f e c t on programme outcome in repeated measures. Hypotheses 1 to 3 are l i s t e d here for reference. 1. Students' completion rate is s i g n i f i c a n t l y associated with turnaround a quarter way into the course; those experiencing shorter turnaround submit more assignments than those experiencing longer turnaround. 2. After passing the halfway point of the course, 93 turnaround does not correlate with students' completion rate s i g n i f i c a n t l y . 3. One quarter way into the course, 'high' deviation (slow in submitting assignments) students have a lower completion rate than 'low' deviation (quick in submitting assignments) students. Completion and Turnaround at Vari o u s Stages Recall that courses were divided into four quarters in terms of time. Table 18 shows the mean completion, deviation, grades and turnaround scores for each quarter. Perusal of this table shows that only completion varied s i g n i f i c a n t l y from quarter to quarter. The mean deviation Table 18 Means scores of completion, deviation, grade and turnaround  after each quarter of courses in the 1984 summer distance  education programme at the Chinese University of Hong Kong Quarters 1 St 2nd 3rd 4th completion Mean .71 .52 .42 .36 S.D. .26 . 30 .30 .28 deviat ion Mean 20.64 20.40 20. 17 20. 1 1 S.D. 27.77 28.34 28.47 29.22 grades Mean .73 .73 .73 .73 S.D. . 1 0 .09 .09 .09 turnaround Mean 97.61 98.40 97.84 97. 14 S.D. 94.03 90.70 88.51 86.61 94 score was always approximately 20 days, i r r e s p e c t i v e of which q u a r t e r . Grades were always very c l o s e to .73 and turnaround only v a r i e d between 97 to 100 days. Thus, with the exception of completion, students behaved much the same with respect to d e v i a t i o n and grades through the courses and so d i d the i n s t i t u t i o n (with respect to turnaround). However, what t h i s t a b l e f a i l s to show are the r e l a t i o n s h i p s between student and i n s t i t u t i o n behaviour. The matter w i l l be d i s c u s s e d i n some d e t a i l i n Chapter E i g h t . Now i t i s necessary to examine the c o r r e l a t i o n s between the d e r i v e d v a r i a b l e s with themselves at v a r i o u s stages of the course. Table 19 c l e a r l y shows the four v a r i a b l e s q u i c k l y approximated t h e i r t e r m i n a l v a l u e s very e a r l y on i n the course. T h i s was e s p e c i a l l y t r u e f o r d e v i a t i o n , turnaround and grades. A f t e r the f i r s t q u a r t e r , t h e i r values a l r e a d y showed q u i t e s u b s t a n t i a l c o r r e l a t i o n w i t h the t e r m i n a l values and i n c r e a s e d f u r t h e r a f t e r the 2nd and 3rd quarter to almost the p e r f e c t c o r r e l a t i o n of value 1. In other words, the p a t t e r n of d e v i a t i o n , turnaround and grades d i d not vary a great d e a l throughout the d u r a t i o n of the course. For completion r a t e , the approach' to the t e r m i n a l value was not as dramatic but s t i l l s u b s t a n t i a l . A f t e r the f i r s t q u a r t e r , the c o r r e l a t i o n was a l r e a d y r=.67, p <.001, q u i c k l y turned r=.90, p <.001, a f t e r the second q u a r t e r . The p a t t e r n f o r these four v a r i a b l e s were f i r m l y set a f t e r the 95 second quarter when each correlation attained r=.90, p <.001 or better. Table 19 Correlations between derived variables at various staqes and at end of the course Overall completion Completion after 1st quarter 0.67** Completion after 2nd quarter 0.90** Completion after 3rd quarter 0.97** Overall deviation Deviation after 1 St quarter 0.88** Deviation after 2nd quarter 0.95** Deviation after 3rd quarter 0.99** Overall turnaround Turnaround after 1 St quarter 0.96** Turnaround after 2nd quarter 0.98** Turnaround after 3rd quarter 0.99** Overall grades Grades af t e r 1 St quarter 0.93** Grades after 2nd quarter 0.97** Grades after 3rd quarter 0.98** * 2 < . 0 1, ** p_ <-001 96 As can be seen from Table 20, the correlation between completion and turnaround at the end of the course was negative and s i g n i f i c a n t (r=-.14, p <.001). After the f i r s t quarter of the course, the cor r e l a t i o n increased but i t was s t i l l negative and s i g n i f i c a n t (r=-.21, p <.001). Students who had experienced shorter turnaround submitted more assignments. Hypothesis' 1 was confirmed. Table 20 Correlations between completion and turnaround By time into the course 1st quarter ,-.21** 2nd quarter -.20** 3rd quarter -.16** 4th quarter -.14** * 2 <•0 1r ** p <.001 After the second quarter, correlation between completion and turnaround decreased s l i g h t l y , but s t i l l negative and s i g n i f i c a n t (r=-.20, p <.001). After the t h i r d quarter, the cor r e l a t i o n decreased further, but s t i l l negative and s i g n i f i c a n t (r = - . l 6 , p <.001). After passing the halfway point, changes in turnaround s t i l l had a s i g n i f i c a n t influence on completion. Hypothesis 2 was rejected. Completion and D e v i a t i o n at V a r i o u s Stages The co r r e l a t i o n between completion and deviation at the end of the course, as shown in Table 21, was negative and s i g n i f i c a n t (r=-.09, p <.01). 97 Table 21 C o r r e l a t i o n s between completion and d e v i a t i o n By time i n t o the course 1st q u a r t e r 2nd q u a r t e r 3rd q u a r t e r 4th q u a r t e r -. 1 2** -.11** -.12** -.09* * 2 < « 0 1 f ** 2 <- 001 A f t e r the f i r s t q u a r t e r , c o r r e l a t i o n was higher (r=-.12, p <.001), students who were d e v i a t i n g more from due dates i n the submission of assignments handed in fewer assignments than those who were d e v i a t i n g l e s s . Hypothesis 3 was con f i rmed. Hypotheses 1 to 3 were time dependent, i . e . they s t a t e d a r e l a t i o n s h i p at some p o i n t i n the course. However, even i f the time element was taken out, Hypotheses 1 and 3 were s t i l l not to be r e j e c t e d and Hypothesis 2 were s t i l l to be r e j e c t e d . 98 CHAPTER EIGHT DISCUSSION What this study has shown, f i r s t of a l l , i s that with each element of 'outcome' defined (completion, deviation, grades and turnaround), i t was possible to test hypotheses which were time based. This provided us with an instrument to look at what was happening at various stages of the distance education course. The amount of results from various analyses were substantial. Although only a small proportion of this information was necessary to examine the four hypotheses, a treasure chest of usable data was uncovered. Those which were not d i r e c t l y relevant to the hypotheses were included in Appendices. Comparison with Previous Research Table 2 to 4 in Chapter Two summarised some of the results from previous research. We can now f i l l in most of the gaps s t i l l outstanding. Some variables w i l l be consolidated. Completion of f i r s t assignment becomes i n i t i a l completion rate and completion of assignments becomes completion rate while the course i s in progress. Deviation of f i r s t assignment becomes i n i t i a l deviation. 99 Table 22 Grades, completion & deviation vs dependent variables GRADES COMPLETION DEVIATION COMPLETION OF PRE-COURSE ASSLGNMENT S i g n i f i c a n t Wilson, 1976 INITIAL COMPLETION RATE I n s i g n i f i c a n t S i g n i f i c a n t Wong, 1987 P f e i f f e r & Sabers, 1970 Wong, 1987 Negat ive Wong, 1987 COMPLETION WHILE COURSE IN PROGRESS P o s i t i v e S i g n i f i c a n t Negative Wong & Wong, 1979 Wong & Wong, 1979 Wong, 1987 Not s i g n i f i c a n t Wong, 1987 Wong, 1987 TURNAROUND Not s i g n i f i c a n t Wong, 1987 Negative Rekkedal, 1983 Wong, 1987 Not s i g n i f i c a n t Wong, 1987 INITIAL DEVIATION Not s i g n i f i c a n t Donehower, 1968 Wong, 1987 Negative P f e i f f e r & Sabers, 1 970 Negative Donehower, 1968 P f e i f f e r & Sabers 1 970 Rekkedal, 1972 Wong, 1987 S i g n i f i c a n t Wong, 1987 DEVIATION P o s i t i v e Rekkedal, 1972 Not s i g n i f i c a n t H a r t s e l l , 1974 Wong, 1987 Not s i g n i f i c a n t Rekkedal, 1972 Negative Wong, 1987 Not a p p l i c a b l e GRADES Not a p p l i c a b l e Not s i g n i f i c a n t P f e i f f e r & Sabers, 1970 Wong, 1987 Not a p p l i c a b l e 100 Table 23 Grades, completion and deviation vs. student variables GRADES COMPLETION DEVIATION AGE Not s i g n i f i c a n t Donehower, 1968 Ha r t s e l l , 1974 P o s i t i v e Wong, 1987 P o s i t i v e Donehower, 1968 Rekkedal, 1972 Bradley & Lehman, 1 975 Wong, 1987 Not s i g n i f i c a n t Coldeway et a l , 1 980 Smith, 1983 Not s i g n i f i c a n t Rekkedal, 1972 P o s i t i v e Wong, 1987 SEX Female higher Spencer, 1965 Donehower, 1968 H a r t s e l l , 1974 Not s i g n i f i c a n t Wong, 1987 Female higher Donehower, 1968 Duby & Giltrow, 1 978 Coldeway et a l , 1 980 Woodley & Parlett, 1 983 Not s i g n i f i c a n t Smith, 1983 Wong, 1987 Not s i g n i f i c a n t Wong, 1987 EDUCATION P o s i t i v e Donehower, 1968 Rekkedal, 1972 Wong, 1987 P o s i t i v e Donehower, 1968 Rekkedal, 1972 Coldeway et a l , 1980 Not s i g n i f i c a n t Wong, 1987 Not s i g n i f i c a n t Rekkedal, 1972 Wong, 1987 OCCUPATION S i g n i f i c a n t Wong, 1987 Not s i g n i f i c a n t Smith, 1983 Wong, 1987 S i g n i f i c a n t Woodley & Parlett, 1 983 Not s i g n i f i c a n t Wong, 1987 101 Table 24 Grades, completion & deviation vs programme variables GRADES COMPLETION DEVIATION • DURATION Negative P f e i f f e r & Sabers, 1 970 P o s i t i v e Wong, 1987 Negative Riches, 1979 Wong, 1987 Not s i g n i f i c a n t Rekkedal, 1972 Negative Wong, 1987 SUBJECT(NATURE) S i g n i f i c a n t Wong, 1987 S i g n i f i c a n t Orton, 1977 Wong, 1987 S i g n i f i c a n t Wong, 1987 FEE Not s i g n i f i c a n t Wong, 1987 Negative Wong, 1987 Not s i g n i f i c a n t Wong, 1987 LANGUAGE OF INSTRUCTION Not s i g n i f i c a n t Not s i g n i f i c a n t Wong, 1987 Wong, 1987 Not s i g n i f i c a n t Wong, 1987 In this study, we explored an array of in t e r -relationships and f i l l e d in some gaps l e f t by previous research. From Table 22 to 24, i t can be seen that on the whole, many findings from previous research have been confirmed. This is surprising considering the vast differences in contexts and operation. There were also no diametrically opposite re s u l t s . I n t e r p r e t a t i o n of Present R e s u l t s The hypotheses were in fact drawn up to investigate the hunch that there might be variations in the derived variables 1 02 across time. Previous research strongly suggested that student performance at the i n i t i a l stage of the course was a very t e l l i n g indicator of things to come. This was explained by Wong and Wong (1979) as student commitment to a programme of study, students made their - commitment to complete the course after the programme started and were aware of the amount of work and the d i f f i c u l t i e s involved. It was assumed that after the f i r s t quarter of the course, commitments would have been made. Once students have made the commitment, variations in turnaround could be seen as posi t i v e and negative reinforcement to the commitment. One is tempted to speculate that shorter turnarounds would give feedback to the previous.assignment and signalled to students that the o f f e r i n g i n s t i t u t i o n was ready to respond and expected the next assignment to be submitted, that the student was not l e f t to fe e l e n t i r e l y without attention. On the other hand, deviation or how quickly the assignments were submitted could be seen as conducive to the forming of behaviour patterns. Once the pattern has been established, more of the same would happen. This led the researcher to speculate that completion rate would not be affected by turnaround more than halfway into the course. This was not supported by the r e s u l t s . However, as turnaround also did not varied s i g n i f i c a n t l y , one can say that the established pattern persisted. The s e l f -correlations for the four variables at d i f f e r e n t stages of 103 the course convincingly demonstrated that once a pattern was set, in t h i s case, the response to turnaround, more of the same would happen l a t e r . Those who had experienced extended turnaround at the beginning of the course took longer to submit their assignments. In the second half of the course, turnaround remained extended and students' response remained the same, they took 'just as long to submit th e i r assignments. The results reinforced the contention that students performance at the i n i t i a l stage of the course was a good predictor of subsequent performance. The problem then i s how to get students to quickly e s t a b l i s h a desirable pattern of study at the beginning of the programme. I m p l i c a t i o n s f o r Future Research Looking again at the mean scores for each of the derived variables (in each quarter) were displayed in Table 18 (Chapter Seven), i t i s important to note that each of these variables was, to some extent, correlated with the other. For example, as shown in Table 17 (Chapter S i x ) , completion was correlated with turnaround (r =-.14, p <.001 ) . If the four time periods are regarded as categories within the variable 'time', and 'outcome' i s construed to have the four components (e.g. completion rate, deviation, grades and turnaround) in the v e r t i c a l axis of the table, we have a 4 x 4 f a c t o r i a l design. E a r l i e r , the descending nature of the 104 completion scores was noted, as was the absence of any d i s c e r n i b l e increase or decrease i n the remaining 'components' over time. I t was beyond the scope of the present study to i n v e s t i g a t e the 'main' or ' p a r t i a l ' e f f e c t s of each of the components, or of time. In a subsequent i n v e s t i g a t i o n i t w i l l be necessary to employ a MANOVA or s i m i l a r procedures to i n v e s t i g a t e the extent to which, fo r example, the descending nature of the 'completion' scores was due j u s t to 'completion' or to i t s i n t e r a c t i o n with d e v i a t i o n , turnaround and grades. The other i m p l i c a t i o n f o r r e s e a r c h i n d i s t a n c e edu c a t i o n , p a r t i c u l a r l y in the area of completion r a t e , i s that there i s a l o t of e x i s t i n g data f o r r e t r o s p e c t i v e study. The source of data f o r t h i s study i s commonly found i n most d i s t a n c e education programmes. I t can a l s o be e a s i l y r e p l i c a t e d . Through the use of the new measures, a host of analyses c o u l d be performed and new i n s i g h t s gained. Meaningful comparative s t u d i e s a c r o s s programmes, i n s t i t u t i o n s and c u l t u r e s c o u l d be and should be made. The psychometric p r o p e r t i e s of the d e r i v e d v a r i a b l e s i n i n v e s t i g a t i o n had been amply demonstrated and should be employed by other r e s e a r c h e r s . In t h i s study, q u e s t i o n s were posted across time and i t was p o s s i b l e to post q u e s t i o n s a c r o s s groups. T h i s c o u l d be groups of students determined by the number of assignments submitted or i t c o u l d have been grouping by student or 105 programme variables. Information in the appendices showed that the influence of some of the factors varied with time and with grouping. T r a d i t i o n a l studies freezed a frame of r e a l i t y for study. An appropriate analogy would be the taking of an X-ray picture of the human body. With the new measures, i t is possible to take clearer and more revealing pictures, l i k e CAT scans, and also from many angles. Another possible avenue of research into completion i s prediction, the precision of the measures makes precision of prediction possible. The d i f f i c u l t task i s in the further interpretation of these r e s u l t s . There i s no "off the shelf" theory that one can turn to. However, the investigation has provided a foundation from which further i n q u i r i e s can be made. I m p l i c a t i o n s f o r P r a c t i t i o n e r s The measures used in t h i s study could be everyday measures of a distance education programmes. They provide for close monitoring of student performance and programme outcome at any stage of the programme. If changes were introduced - whether changes in course design, course material, course arrangement, administrative policy or other factors - the measures could track what changes in outcome would occur. They would help to operate better programmes. One should also make attempts to encourage students to e s t a b l i s h a desirable study pattern at the beginning of the 106 programme and see i f that would improve completion rate. In the control of programme outcome, the findings in th i s study could form the basis of developing a strategy for intervention to improve completion. In short, p r a c t i t i o n e r s do have the psychometric tools to monitor, investigate and the information to develop appropriate e f f e c t i v e strategies to improve programme outcome. 107 REFERENCES Anderson, T. H., & Tippy, P. H., (1971). An exploratory study of correspondence students. Unpublished manuscript, University of I l l i n o i s , Urbana. (ERIC Document Reproduction Service No. ED 070 960). Baath, J . A., (1975). Submission density in nine correspondence courses. Pedagogical Reports, 2. Lund: University of Lund. Bartels, J., (1982). Drop-out at the distance university in  the Federal Republic of Germany. Paper presented at the Annual Forum of the Association for I n s t i t u t i o n Research, Denver, Colorado. (ERIC Document Reproduction Service No. 220 037). Bittner, W. S., & Mallory H. F., (1933). University  teaching by mail. New York: Macmillian. Bradley, A. P. & Lehman, T., (1975). A t t r i t i o n at a  non-traditional I n s t i t u t e . Saratoga Springs: State University of New York. Bradt, K. H., (1954). Why service personnel f a i l to  complete USAFI courses. Washington, D.C.:Government Printing O f f i c e . Chalon, F., (1985). An evaluation model to improve effectiveness in distance courses. Paper presented at the Thirteenth World Conference of the International Council for Distance Education, Melbourne. Childs, G. B., (1956). Preliminary report of the Committee  on Research of the Correspondence D i v i s i o n of the National  University Extension Association. Paper presented at the National University Extension Association Convention. Childs, G. B., (1963). An analysis of c e r t a i n factors which may aff e c t completion in supervised correspondence study. Journal of Experimental Education, 3_2(1), 101-105. Childs, G. B., (1971). Recent research developments in . correspondence inst r u c t i o n . In Mackenzie and Christensen (Eds.), The changing world of correspondence study: • international readings. University Park: Pennsylvania State University Press. 108 Coldeway D. 0., (1982). Recent research in distance learning. In J. S. Daniel & M. A. Stroud & J. S. Thompson (Eds.) Learning at a distance. Edmonton: Athabasca University. Coldeway, D. 0., & Spencer, R., (1980a). The measurement of  a t t r i t i o n and completion in distance learning courses. (REDEAL Technical Report No. 8). Edmonton: Project REDEAL, Athabasca University. Coldeway, D. 0., & Spencer, R., (1980b). Distance education: Interaction between learner attributes and  learner course performance"! (REDEAL Project No. 9) . Edmonton: Project REDEAL, Athabasca University. Daniel, J. S., & Stroud, M. A., & Thompson, J. S., (1982). Learning at a distance: a world perspective. Edmonton: Athabasca University. Davies, T. C , (1977). Open learning system for mature  students: the report on an investigation undertaken for  the Council for Educational Technology. London: Council for Educational Technology. De Goede, M. P. M., & Hoksbergen, R. A. C , (1978). Part-time education at t e r t i a r y l e v e l in the Netherlands. Higher Education, 7, 443-455. Di S i l v e s t r o , F. R., & Markowitz, H., J r . , (1982). Contracts and completion rates in correspondence study. Journal of  Educational Research, 75(4), 218-221. Donehower, G. M., (1968). Variables associated with  correspondence study enrollment at the University of  Nevada, 1963-1965. Unpublished master's thesis, University of Nevada. (ERIC Document Reproduction" Service No. ED 016 925). Duby, P. B., & Giltrow, D. R., (1978). Predicting student withdrawals in open learning courses. Educational  Technology, February, 43-47. Essex, D. L., & Anderson, T. H., (1972). Some correlates of  success in correspondence study. (ERIC Document Reproduction Service No. 070 952). Fa i r i n g , R. L., & Hughes, C. R., (1950). An analysis of  students' reasons for f a i l i n g to complete correspondence  study. G a i n s v i l l e : Correspondence Study Department, General Extension D i v i s i o n , University of F l o r i d a . 109 Flinck, R. , (1978). Correspondence education combined with  systematic telephone tutoring. (ERIC Document Reproduction Service No. ED 168 535). Glatter, R. , & Wedell, E. G., (1971). Study by  correspondence. London: Longman. Graholm, G. W., & Ljosa, E., (1977). ICCE Research Survey 1976. ICCE Newsletter, 7(4), 3-30. Hammer, P., & Smith, W. A. S.,' (1979). A t t r i t i o n -completion study in an Open University. Paper presented at the conference of The Association for I n s t i t u t i o n a l Research, San Diego. Harris, W. J. A., & Williams, J . D. S., (1977). A handbook  on distance education. Manchester: University of Manchester, Department of Adult and Higher Education. H a r t s e l l , C , (1971). Correspondence dropouts. Why? In D. E. Mathieson (Ed.), Correspondence study: A review of  research and development l i t e r a t u r e . New Yorl<:: Syracuse University. H a r t s e l l , C. W., (1975). Knowledge - Instant or delayed? N.U.E.A. Spectator, 38(19), 26-30. Holmberg, B., (1974). Distance education. Malmo: Hermods. Holmberg, B., (1977). Distance education: Survey and  bibliography. London: Kogan Page. Holmberg, B., (1982a). Recent research into distance education. Hagen: FernUniversitat, Zentrales I n s t i t u t fur Fernstudienforschung. Holmberg, B., (1982b). Recent research into distance educat ion 11. Hagen: FernUniversitat, Zentrales I n s t i t u t fur Fernstudienforschung. Hornik, R. C , (1976). Useful evaluation designs for evaluation the impact of distance learning systems: methodology. Educational Broadcasting International, 9, 6-10. Hughes, C. R., (1955). The influence of some selected factors upon the completion of correspondence study courses. Dissertation Abstracts International, 15, 2459-2460. James, B. J., & Wedemeyer, C. A., (1959). Completion of university correspondence courses by adults: the eff e c t s 110 of g o a l - c l a r i t y and other factors. Journal of Higher  Education, 30, 87-93. Kaye, A., & Rumble, G., (1981). Distance teaching for  higher and adult education. London: Croom Helm. Keegan, D. J., (1980). Drop-outs at the Open University. Australian Journal of Education, 24(1), 44-55. Keegan, D. J., (1983). Six distance education theorists. Hagen: FernUniversitat, Zentrales I n s t i t u t fur Fernstudienforschung. Kennedy, D., & Powell, R., (1976). Student progress and withdrawal in the Open University. Teaching at a  Di stance, 1_, 61-75. Kuhanga, N. A., (1981). The concept of adult education at a distance and i t s application in developing countries. In M. W. Neil (Ed.), Education of adults at a distance. Milton Keynes: Kogan Page. Ljosa, E., (1978). Trends and p r i o r i t i e s in distance education research. ICCE Newsletter, 8(4), 4-8. MacKenzie, 0., & Christensen, E. L., (1971). The changing  world of correspondence study: international readings. University Park, Pennsylvania: Pennsylvania State University. MacKenzie, 0., Christensen, E. L., & Rigby, P. H., (1968). Correspondence instruction in the United States. New York: McGraw-Hill. Mathieson, D. E., (1971). Correspondence study: A summary  review of the research and development l i t e r a t u r e . New York: Syracruse University. (ERIC Document Reproduction Service No. ED 047 163). Monstain, B. R., (1974). Students who desire independent study: some distinguishing c h a r a c t e r i s t i c s . College  Student Journal, 8(4), 85-92. Moore, M. G., (1976). Attitudes to independent study and  cognitive style of adult learners who use correspondence  independent study and self directed independent study. Unpublished manuscript. N e i l , M. W. (Ed.), (1981). Education of adults at a  distance. Milton Keynes: Kogan Page. 11 1 Orton, L. J., (1977). Completion and non-start rates in correspondence courses. Journal of University Continuing  Education, 4, 21-26. Pe r c i v a l , F. , & Ell i n g t o n , H., (1981). Distance learning  and evaluation. London: Kogan Page. P f e i f f e r , J. W. , (1969). The effect of l e t t e r s and postcards of encouragement on the submission of lessons in correspondence study courses. Doctoral d i s s e r t a t i o n , University of Iowa. P f e i f f e r , J. W., & Sabers, D., (1970). A t t r i t i o n and achievements in correspondence study. Washington, D.C: National Home Study Council. (ERIC Document Reproduction Service No. ED 036 707). Phyhtian, T. , & Clements, M., (1982). Drop-out from t h i r d l e v e l Maths courses. Teaching at a Distance, 21, 35-45. Powell, E., (1965). Survey of students enrolled for correspondence in s t r u c t i o n . Georgia Center for Continuing Education, University of Georgia, Athens. Pulley, J . , (1965). A study of correspondence student  dropouts. University Extension D i v i s i o n , Columbia, Missouri. Rekkedal, T. , (1972). Correspondence studies, recruitment,  achievement and discontinuation. Oslo: NKI - Skalen Undervisningssentrum. (ERIC Document Reproduction Service No. ED 154 115). Rekkedal, T., (1981). Introducing the personal tutor/counsellor in the system of distance education.  Project Report 1: Experiment description. Paper presented at Annual Conference of the European Home Study Council, Madrid. (ERIC Document Reproduction Service No. ED 235774) . Rekkedal, T. , (1983). The written assignments in correspondence education. Effects of reducing turn-around time. An experimental study. Distance Education, 4(2), 231-251 . Riches, E., (1979). A t t r i t i o n from correspondence courses. B. C. Institute of Technology, Div i s i o n of Continuing Education and Industrial Services, Vancouver, B. C. Riley, J . , (1984). The problems of draf t i n g distance education materials: the problems of revising drafts of distance education materials; and draf t i n g behaviours in 1 1 2 the production of distance education m a t e r i a l . B r i t i s h  Journal of Educational Technology, J_5(3), 1 9-2-238. Ruggles, R. H. et a l , (1982). Learning at a distance and  the new technology. Vancouver: Educational Research Ins t i tute of B . C . Rumble G . , & Harry, K. (Eds . ) , (1982).. The distance  teaching u n i v e r s i t i e s . London: Croom Helm. Sabers, D. L . , P f e i f f e r , J . W. , & Ragsdale, J . P . , (1972). The effects of l e t t e r s and postcards of encouragement on the submission of lessons in correspondence study. Journal of Experimental Education, 4J_( 1 ) , 87-90. Sewart, D . , Keegan, D . , & Holmberg, B. (Eds . ) , (1983). Distance education: in t erna t iona l perspect ives . Croom Helm: London. Shale, D . , (1980). Course completion rates . Edmonton: Athbasca Un ivers i ty , Off ice of I n s t i t u t i o n a l Studies . Sloan, D . , (1965). Survey study of un ivers i ty dropouts and  cance l l a t i ons . Lexington: Univers i ty Extension, Univers i ty of Kentucky. Smith, J . , (1983). Evaluation of the Telecourse program at  Saddleback Col lege: Student retent ion and academic  achievement. Doctoral d i s s e r t a t i o n , Nova U n i v e r s i t y . (ERIC Document Reproduction Service No. ED 239 684). Spencer, 0 . , (1965). Factors associated with persons who completed correspondence courses . Home Study Review, 5, 10-24. Spencer, R. E . , (1980). The e f fec ts of computer generated  schedules on the performance of Athabasca U n i v e r s i t y  learners"! (REDEAL Research Report No. 5~H Edmonton: Project REDEAL, Athabasca U n i v e r s i t y . Sweet, R . , (1982). Distance education: The personal response. Paper presented at the Annual Meeting of the  American Educational Research Assoc ia t ion , New York. (ERIC Document Reproduction Service No. ED 216 658). Sweet, R . , (1984). Applying T i n t o ' s model of student dropout to distance education. Unpublished manuscript. Tate, P. J . , & Kresse l , M. ( E d s . ) , (1983). The expanding  role of telecommunications in higher education. San Francisco: Jossey Bass. 1 1 3 Thompson, G., (1984). The c o g n i t i v e s t y l e of field-dependence as an explanatory c o n s t r u c t i n d i s t a n c e education drop-out. D i s t a n c e Education, 5(2), 286-293. Verner, C , & Booth, A., (1964). Adult Education. Washington, D.C: Centre f o r A p p l i e d Research i n Educat i o n . Wedell, E. G., " (1970). The p l a c e of education by correspondence i n Permanent E d u c a t i o n . Strasbourg: C o u n c i l f o r C u l t u r a l Co-operation of the C o u n c i l of Europe. W i l l e n , B., (1981). D i s t a n c e education at Swedish u n i v e r s i t i e s . An e v a l u a t i o n of the experimental programme  and a follow-up study. Uppsala: Uppsala U n i v e r s i t y . (ERIC Document Reproduction S e r v i c e No. ED 243 458). Wilson, H. G., (1976). A quest f o r c o n t r o l of a t t r i t i o n  from correspondence study. Sydney: C o l l e g e of E x t e r n a l S t u d i e s . (ERIC Document Reproduction S e r v i c e No. ED 144 546 & ED 144 547). Wilson, R. R., (1968). The e f f e c t s of s e l e c t e d programming - analog techniques and v o i c e - c o n t a c t on completion behavior in correspondence e d u c a t i o n . D i s s e r t a t i o n  A b s t r a c t s I n t e r n a t i o n a l , 29, 2579A. Wong, A. T., & Wong, S. C. 0., (1979). The r e l a t i o n s h i p between assignment completion and the a t t r i t i o n and achievement in correspondence c o u r s e s . J o u r n a l of  E d u c a t i o n a l Research, 72(3), 165-168. Woodley, A., & P a r l e t t , M., (1983). Student drop-out. Teaching at a D i s t a n c e , 24, 2-23. 1 1 4 APPENDICES i 1 15 Appendix 1 Courses offered in the 1984 summer session distance education programme of the Chinese University of Hong Kong No. T i t l e Duration Fee Nature 5201 Chinese Language (C) 10 $90 L 5202 P r a c t i c a l Chinese Writing (C) 10 $95 L 5203 Approach to Chinese Poetry (C) 10 $95 L 5204 Translation-'from English to Chinese (C) 12 $135 L 5205 Business Administration (C) 13 $160 V 5206 Pr i n c i p l e s of Economics (C) 12 $150 A 5207 Child Development & Care (C) 7 $90 E 5208 Pr i n c i p l e s of Two Dimensional Design (C) 10 $110 E 5209 Chinese Painting (C) 12 $100 E 5220 Introduction to Modern Mathematics (C) 20 $160 A 5221 Business Correspondence (C) 7 $90 V 5223 Monochrome Drawing (C) 12 $120 E 5210 Problems in English for Asians (E) 12 $140 L 5211 E f f e c t i v e English Writing (E) 10 $140 L 5213 English Comprehension (E) 12 $110 L' 5214 Advanced English Writing (E) 6 $135 L 5215 English Vocabulary (E) 12 $110 L 5216 What's Wrong with Your English (E) 16 $140 L 5218 Introductory Business Correspondence (E) 6 $90 V 5222 Idiom-Magic (E) 6 $140 L 5224 General Principles of English Law (E) 21 $160 V (C)=Chinese r (E)=English L=Language, V=Vocational, A=Academic, E=Enrichment 116 Appendix 2 Analys is a f ter 1st quarter , n=773 cr ad at sg Pearson's Corre la t ion Age .02 .06 -.08 .08 Education .00 - .05 .00 . 1 4** Duration • .02 - .06 -.15** .30** Fee - . 1 3** .00 .27** .04 One-way Occupat ion N . S . N.S . N.S . F=5.27 p <.01 i Nature F = 8 .39 N.S. F=32.83 F=14.54 P < .01 p <.01 p <.01 T-Test Language t = 3 .59 N.S . t=-22.55 t=-2.00 P < .01 p <.01 p <.05 Sex N . S . N.S . N.S . t=2.15 p <.05 Pearson's Corre la t ion cr 1 .00 ad - . 1 2** 1 ,00 at - .21** .02 1 .00 sg - .05 -.01 -.14** 1 .00 * p <.01, ** 2 <• 001 cr age educ dura fee ad at sg Mean 0.71 2.36 2.29 1 1 .55 124.98 20.64 97 .61 0.73 S.D. 0.26 1 .39 0.59 3.46 23.54 27.77 94 .03 0.10 1 1 7 Appendix 3 Analys is af ter 2nd quarter , n=773 cr ad at sg Pearson's Corre la t ion Age - .06 .05 -.08 .10* Educat ion .03 - .06 .00 • m -j 4 * * Duration .05 - .07 -.15** .28** Fee - .07 .00 .27** .02 One-way Occupat ion N .S . N.S. F=2.18 F = 6.20 p <.05 P <.01 Nature F = 7 .46 N.S . F=35.87 F = 19.26 P < .01 p <.01 P <.01 T-Test Language t = 2 .97 N.S . t=23.86 N.S. P < .01 p <.01 Sex N • S. N.S . N.S . N.S . Pearson's Corre la t ion cr 1 .00 ad - .11** 1 .00 at - . 20** .02 1 .00 sg - .03 .02 -.13 1 loo * p_ <.01 , ** p_ <. 001 cr age educ dura fee ad at sg Mean 0.52 2.36 2.29 1 1 . 55 124.98 20.40 98 .40 0.73 S.D. 0.30 1 .39 0.59 3. 46 23.54 28.34 90 .70 0.09 118 Appendix 4 Analys is af ter 3rd quarter , n=773 cr ad at sg Pearson's Corre la t ion Age .07 .04 - .07 .10* Educat ion .06 - .07 .01 . 1 4** Duration .02 - .11* -.15** .28** Fee - .04 -.01 . 27** .02 One-way Occupation N .S . N .S . F=2.31 F = 6 .19 p <.05 P < .01 Nature F = 9 .93 F = 3 .27 F=36.76 F = 21 .77 P < .01 P <.05 p <.01 P < .01 T-Test Language N .S . N .S . t=-24.52 N • S. p <.01 Sex N .s. N.S . N.S . N .S . Pearson's Corre la t ion cr 1 .00 ad - . 1 2** 1 .00 at. - . 1 6** .03 1 .00 sg - .02 -.01 -.14** 1 .00 * p <.01, ** p <. 001 cr age educ dura fees ad at sg Mean 0.42 2.36 2.29 11. 55 124.98 20.17 97 .84 0.73 S.D. 0.30 1 .39 0.59 3. 46 23.54 28.47 88 .51 0.09 119 Appendix 5 Analys i s for those who handed in 12 or more  assignments, n=40 cr ad at sg Pearson's Corre lat ion Age . 1 2 .04 -.32 - .07 Educat ion -.36- -.31 - .09 - .26 Duration -.67** -.51** - .47* .74** Fee -.49** -.40* -.73** .70** One-way Occupation F=3.87 N.S. N.S. F=2 .43 p <.01 P < .05 Nature F=4.14 F = 3.68 F=12.20 F=1 9 .50 p <.05 P <.05 p <.01 P < .01 T-Test Language t=3.75 N.S. N.S. t = -3 .86 p <.01 P < .01 Sex N.S. N.S . t=3.56 N .S . p <.01 Pearson's Corre la t ion cr 1 .00 ad .37* 1 .00 at .34 .39* 1 .00 sg -.55** - .47* -.30 1 .00 * 2 < « 0 1 , ** p <. 001 cr age educ dura fee ad at sg Mean 0.85 3.00 2.50 16. 98 149.25 2.39 45 .29 0.78 S.D. 0.16 1 .41 0.82 3. 98 16.11 33.82 36 .84 0.11 120 Appendix 6 Analysis for those who handed in 10 or more  assignments, n=8l cr ad at sg Pearson's Correlation Age .01 .23 -.21 - .09 Education - .17 -.22 .00 . 1 9 Durat ion - .67** -.37** -.26* .59** Fee - .53** -.07 -.29* .39** One-way Occupat ion F = 3 .55 N.S. N.S. N .S. P < .01 Nature F = 5 . 1 2 N.S. F=14.52 F = 21 .91 P < .01 p <.01 P < .01 T-Test Language t = 5 .01 t = 2.05 t=-5.37 t = -2 .54 P < .01 P <.05 p <.0l P < .05 Sex N .S. N.S. N.S. N .S. Pearson's Correlation cr 1 .00 ad .21 1 .00 at - .01 .09 1 .00 sg - .37** -.17 -.28* 1 .00 * £ <.01r ** £ <. 001 cr age educ dura fee ad at sg Mean 0.84 2.89 2.41 14. 94 137.04 10.82 63 .57 0.75 S.D. 0.16 1 .82 0.74 4. 04 23.4 38.65 53 .16 0.10 121 Appendix 7 Result of analysis for those who handed in 8 or more  assignments, n =111 cr ad at sg Pearson's Correlation Age .04 . 1 4 -.23* .00 Education - .07 -.23* -.08 .22* Durat ion - .50** -.37** -.28* .54** Fee - .43** -.11 -.05 .25* One-way Occupat ion F = 3 .38 N.S. N.S. F=2.13 P < .01 p <.05 Nature N .S. N.S. F=17.49 F=25.00 p <.01 p <.01 T-Test Language t = 5 .96 t = 2.23 t=-7.84 N.S. P < .01 P <• 05 p <.01 Sex N .S. N.S. N.S. N.S. Pearson's Correlation cr 1 .00 ad . 16 1 .00 at - .18 .07 1 .00 sg - .24* -.17 -.40** 1 .00 * 2 <.01, ** p <. 001 cr age educ dura fee ad at sg Mean 0.81 2.70 2.33 14. 08 133.6 13.15 75 .23 0.74 S.D. 0.16 1.71 0.67 3. 92 23.9 34.57 59 .83 0.10 1 22 Appendix 8 Analysis for those who handed in 6 or more  assignments, n=20l cr ad at sg Pearson's Correlation Age 0.13 0.13 -0. 1-1 -0.01 Educat ion 0.01 -0.19* -0.02 0.17* Durat ion -.40** -.26** -.1 6 .35** Fee -0.24 -0.08 0.08 0.07 One-way Occupat ion F=2.96 N.S. N.S. F=2.10 p <.01 p <.05 / Nature N.S. N.S. F=11.97 F=22.05 p <.01 p <.01 T-Test Language t=2.41 N.S. t=-12.51 N.S. p <.05 p <.01 Sex N.S. N.S. N.S. N.S. Pearson's Correlation cr 1 . 00 ad • 10 1 .00 at • 07 .08 1 .00 sg • 1 0 -.14 -.28** 1 .00 * £ <.01 , * * 2 <• 001 Mean S.D. cr 0.75 0.20 age 2.56 1 .56 educ 2.32 0.63 dura 12.37 4.10 fee 125.35 25.70 13 28 ad .72 78 .84 64 at .48 .77 0 0 sg .73 .09 123 Appendix 9 Analys is for those who handed in 6 or fewer  assignments, n=640 cr ad at sg Pearson's Corre lat ion Age .01 .04 -.04 . 1 2* Education .04 .05 .03 . 1 4** Durat ion -.44** -.03 -.13** .24** Fee - .25** .03 .33** -.04 One-way Occupat ion N.S . N.S . F=2.22 F = 4.88 p <.05 P <. 01 Nature F=5.95 N.S. F=26.36 F = ;7.84 p <.01 p <.01 P <.01 T-Test Language N.S. N.S . t=-23.83 N.S. p <.01 Sex N.S. N.S . N.S . N.S. Pearson's Corre la t ion cr 1 .00 ad - .05 1 .00 at - .07 .01 1 .00 sg - .07 .01 -.14** 1 .00 * 2 < « 0 1 , ** g <. 001 cr age educ dura fee ad at sg Mean 0.27 2.30 2.27 11. 14 124.00 21.32 101 .28 0.73 S.D. 0.20 1 .30 0.57 3. 19 23.09 28.56 90 .21 0.09 1 24 Appendix 10 Analysis for those who handed in 4 or fewer  assignments, n=539 cr ad at sg Pearson's Correlat ion Age - .08 .04 -.05 . 1 5** Educat ion .03 -.05 .03 . 1 6** Duration - .41** -.04 -. 17** .26** Fee - .31** .03 .34** - .01 One-way Occupat ion N .S. F = 2.45 F=2.87 F= :4 .80 P <.05 p <.01 P < .01 Nature F = 4 .70 N.S. F=27.13 F= •1 .65 P < .01 p <.01 P < .01 T-Test Language N .S. N.S. t=-21.86 N .S. p <.01 Sex N . s . N.S. N.S. N . s . Pearson's Correlation cr 1 .00 ad - .04 1 .00 at .12 .00 1 .00 sg - .11* .03 -.13* 1 .00 * p_ <.01 , ** £ <. 001 cr age educ dura fee • ad at sg Mean 0.21 2.28 2.27 1 1 . 33 124.81 21 .77 104 .79 0.73 S.D. 0.11 1 .32 0.58 3. 17 22.81 29.27 92 .46 0.10 125 Appendix 11 Analys i s for those who handed in 2 or fewer  assignments, n=380 cr ad at sg Pearson's Corre la t ion -Age .05 .02 -.04 . 1 1 Educat ion .01 -.04 .06 . 20** Duration -.60** .02 - . 16** .23** Fee -.38** .04 . 38** -.04 One-way Occupat ion N.S. N.S . N.S . F =  3.50 P <.01 Nature F= = 6.7.1 N.S . F=14.39 N.S. P <.01 p <.01 T-Test Language N.S. N.S . t=-17.86 N.S. p <.01 Sex N.S. N.S . N.S . N.S. Pearson's Corre la t ion cr 1 .00 ad -.40 1 .00 at .03 -.01 1 .00 sg - .07 .05 - .15 1 .00 * p <.01, ** 2 <-001 cr age educ dura fee ad at sg Mean 0.16 2.37 2.26 11.10 124.71 22.98 110.4 0.73 S.D. 0.06 1.40 0.55 2.92 22.36 31.74 97.97 0.10 1 26 Appendix 12 Correlations between completion and age By number of assignments submitted 6 or more .13 8 or more .04 10 or more .01 12 or more .12 6 or less .01 4 or less -.08 2 or less .05 By time into the course 1st quarter .02 2nd quarter -.06 3rd quarter .07 4th quarter .08* * p <.01 , ** p_ < . 00 1 Appendix 13 Correlations between completion and education By number of assignments submitted 6 or more .01 8 or more -.07 10 or more -.17 12 or more -.36 6 or less .04 4 or less .03 2 or less .01 By time into the course 1st quarter .00 2nd quarter .03 3rd quarter .06 4th quarter .04 * 2 <•0 1 / ** 2 <.001 127 Appendix 14 Correlations between average deviation and age By number of assignments submitted 6 or more . 1 3 8 or more . 1 4 10 or more .23 12 or more .04 6 or less .04 4 or less .04 2 or less .02 By time into the course 1st quarter .06 2nd quarter .05 3rd quarter .04 4th quarter .05* * p <.01, ** p <.001 Appendix 15 Correlations between average deviation and education By number of assignments submitted 6 or more -.19* 8 or more -.23* 10 or more -.22 12 or more -.31 6 or less .05 4 or less -.05 2 or less -.04 By time into the course 1st quarter -.05 2nd quarter • -.06 3rd quarter -.07 4th quarter -.08 * 2 < « 0 1 r ** £ <-001 128 Appendix 16 Correlations between average turnaround and age By number of assignments submitted 6 or more -.11 8 or more -.23* 10 or more -.21 12 or more -.32 6 or less -.04 4 or less -.04 2 or less -.04 By time into the course 1st quarter -.08 2nd quarter - .08 3rd quarter -.07 4th quarter -.07 * 2 < « 0 1 > ** E <.001 Appendix 17 Correlations between average turnaround and education By number of assignments submitted 6 or more -.02 8 or more -.08 10 or more .00 12 or more -.09 6 or less .03 4 or less .03 2 or less .06 By time into the course-1st quarter .00 2nd quarter .00 3rd quarter .01 4th quarter .10 * 2 < « 0 1 > ** 2 <.001 129 Appendix 18 Correlations between standardised grades and age By number of assignments submitted 6 or more -.01 8 or more .00 •1 0 or more -.09 12 or more -.07 6 or less '.03 4 or less .03 2 or less .06 By time into the course 1st quarter .00 2nd quarter .00 3rd quarter .01 4th quarter , .10 * p < . 0 1 , ** p < . 0 0 1 Appendix 19 Correlations between standardised grade and education By number of assignments submitted 6 or more .17* 8 or more .22* 10 or more . 1 9 12 or more -.26 6 or less .14** 4 or less .16** 2 or less .20** By time into the course 1st quarter . 14** 2nd quarter .14** 3rd quarter .14** 4th quarter .14** * 2 < ' 0 1 ' * * E < . 0 0 1 1 30 Appendix 20 One-way analysis for completion and occupation By number of assignments submitted 6 or more F=2.96 P <.01 8 or more F=3.38 P <.01 10 or more F=3.55 P <.01 12 or more F=3.87 P <.01 6 or less N.S. 4 or less N.S. 2 or less N.S. By time into the course 1st quarter N.S. 2nd quarter N.S. 3rd quarter N.S. 4th quarter N.S. * 2 <'° 1 r ** 2 <-001 Appendix 21 One-way analysis for average deviation and occupation By number of assignments submitted 6 or more N.S. 8 or more N.S. 10 or more N.S. 12 or more N.S. 6 or less N.S. . 4 or less F=2.45 p <.05 2 or less N.S. By time into the course 1st quarter N.S. 2nd quarter N.S. 3rd quarter N.S. 4th quarter N.S. * 2 <« 0 1> ** 2 <•001 131 Appendix 22 One-way analys i s for standardised grade and occupation By number of assignments submitted 6 or more F=2.10 P <.05 8 or more F=2. 13 P <.05 10 or more N.S. 12 or more F=2.43 P <.05 6 or less F=4.88 P <. 01 4 or less F=4.80 P <.01 2 or less F=3.50 P <.01 By time into the course 1st quarter F=5.27 P <.01 2nd quarter F=6.20 P <.01 3rd quarter F = 6. 19 P <.01 4th quarter F=6.26 P <.01 * p <.01, ** 2 <-001 Appendix 23 T-tes t for completion and sex By number of assignments submitted 6 or more N.S. 8 or more N.S . 10 or more N.S . 12 or more N.S . 6 or less N.S . 4 or less N.S . 2 or less N.S. By time into the course 1st quarter N.S. 2nd quarter N.S. 3rd quarter N,S. 4th quarter N.S. * 2 < ' 0 1 > ** P. <• 001 132 Appendix 24 T-test for average deviation and sex By number of assignments submitted 6 or. more N.S. 8 or more N.S. 10 or more N.S. 12 or more N.S. 6 or less N.S. 4 or less N.S. 2 or less N.S. By time into the course 1st quarter N.S. 2nd quarter N.S. 3rd quarter N.S. 4th quarter N.S. * 2 <.01 , ** £ < . 001 Appendix 25 T-test for standardised grade and sex By number of assignments submitted 6 or more N.S. 8 or more N.S. 10 or more N.S. 12 or more N.S. 6 or less N.S. 4 or less N.S. 2 or less N.S. By time into the course 1st quarter t=2.15 p <.05 2nd quarter N.S. 3rd quarter N.S. 4th quarter N.S. * 2 < - 0 1 > ** E <-001 133 Appendix 26 Correlations between completion and duration By number of assignments submitted 6 or more -.40** 8 or more -.50** 10 or more -.67** 12 or more -.67** 6 or less -.44** 4 or less -.41** 2 or less -.60** By time into the course 1st quarter .02 2nd quarter .05 3rd quarter .02 4th quarter -.11** * 2 < - 0 1 f ** 2 <.001 Appendix 27 Correlations between completion and fee By number of assignments submitted 6 or more -.24 8 or more -.43** 10 or more -.53** 12 or more -.49** 6 or l e s s ' -.25** 4 or less -.31** 2 or less -.38** By time into the course 1st quarter -.13** 2nd quarter -.08 3rd quarter -.04 4th quarter -. 12** * 2 < « 0 1 > ** 2 <-001 134 Appendix 28 Correlations between average deviation and duration By number of assignments submitted 6 or more -.26** 8 or more -. 37** 10 or more -.37** 12 or more -.51** 6 or less -.03 4 or less -.04 2 or less .02 By time into the course 1st quarter -.07 2nd quarter -.07 3rd quarter -.11* 4th quarter -.12** * 2 < - 0 1 r ** E <- 001 Appendix 29 Correlations between average deviation and fee By number of assignments submitted 6 or more -.08 8 or more -.11 10 or more -.07 12 or more -.40* 6 or less .03 4 or less .03 2 or less .04 By time into the course 1st quarter .00 2nd quarter .00 3rd quarter -.01 4th quarter -.01 * p_ <-01> ** E <•001 1 35 Appendix 30 Correlations between standardised grade and duration By number of assignments submitted 6 or more .35** 8 or more .54** 10 or more .59** 12 or more .74** 6 or less .26** 4 or less .26** 2 or less .23** By time into the course 1st quarter .30** 2nd quarter .28** 3rd quarter .28** 4th quarter .28** * p_ <.01 , ** p < .00 1 Appendix 31 Correlations between standardised grade and fee By number of assignments submitted 6 or more .07 8 or more .25** 10 or more .39** 12 or more .60** 6 or less -.04 4 or less -.01 2 or less -.04 By time into the course 1st quarter .04 2nd quarter .02 3rd quarter .02 4th quarter .01 * 2 <-°1# ** e <-oo1 136 Appendix 32 One-way analys i s for completion and nature of course . By number of assignments submitted 6 or more N.S . 8 or more N.S. 10 or more F=5.12 P <.01 12 or more F=4.14 P <.05 6 or less F=5.95 P <.01 4 or less F=4.70 P <.01 2 or less F=6.71 P <.01 By time into the course 1st quarter F=8.39 P <.01 2nd quarter F=7.46 P <.01 3rd quarter F=9.93 P <.01 4th quarter F=3.83 P <. 01 * p_ <.01 , ** 2 <. 001 Appendix 3 3 One-way analys i s for average deviat ion and nature of course By number of assignments submitted • 6 or more N.S. 8 or more N.S. 10 or more N.S. 12 or more F=3.68 p <.05 6 or less N.S. 4 or less N.S. 2 or less N.S. By time into the course 1 st quarter N.S. 2nd quarter N.S. 3rd quarter F=3.27 P <.05 4th quarter F=2.74 P <.05 * 2 < - 0 1 r ** E <.001 137 Appendix 34 One-way analys i s for average turnaround and nature of course By number of assignments submitted 6 or more F=11.97 P <.01 8 or more F=17.49 P <.01 10 or more F=14.52 P <.01 12 or more F=12.20 P <.01 6 or less F=26.36 P <.01 4 or less F = 27. 13 P <.01 2 or less F=14.39 P <.01 By time into the course 1st quarter F=32.83 P <.01 2nd quarter F=35.87 P <.01 3rd quarter F=36.76 P <.01 4th quarter F=37.27 P <.01 * 2 < « 0 1 < * * E <.001 Appendix 35 One-way analys i s for standardised grade and nature of course By number of assignments submitted 6 or more F=22.05 ' P <.01 8 or more F=25.00 P <.01 10 or more F=21.91 P <.01 12 or more F=19.50 P <.01 6 or less F=7.84 P <.01 4 or less F=7.65 P <.01 2 or less N.S. By time into the course 1 st quarter F=14.54 N.S. 2nd quarter F=19.26 N.S. 3rd quarter F=21 .77 P <.01 4th quarter F=22.36 P <.01 * 2 < ' 0 1 ' ** E <•001 138 Appendix 36 T-test for completion and language of instruction By number of assignments submitted 6 or more t=2.41 P <.05 8 or more t=5.96 P <.01 10 or more t=5.0l P <.01 12 or more t=3.75 P <.01 6 or less N.S. 4 or less N.S. 2 or less N.S. By time into the course 1st quarter t=3.59 P <.01 2nd quarter t=2.97 P <.01 3rd quarter N.S. 4th quarter N.S. * 2 <.01, ** 2 <.001 Appendix 37 One-way analysis for, average deviation and language By number of assignments submitted 6 or more N.S. 8 or more t=2.23 P <.05 10 or more t=2.05 P <.05 12 or more N.S. 6 or less N.S. 4 or less N.S. 2 or less N.S. By time into the course 1st quarter N.S. 2nd quarter N.S. 3rd quarter N.S. 4th quarter N.S. * 2 < « 0 1 i * * 2 <.001 139 Appendix 38 T - t e s t f o r average turnaround and language of i n s t r u c t i o n By number of assignments submitted 6 or more t=-12.51 P <. 01 8 or more t=-7.81 " P <.01 10 or more t=-5.37 P <. 01 12 or more N.S. 6 or l e s s t=-23.83 P <. 01 4 or l e s s t=-21.86 P <. 01 2 or l e s s t=-17.86 P <. 01 By time i n t o the course 1st q u a r t e r t=-22.55 P <.01 2nd q u a r t e r t = -23 .86 P <. 01 3rd q u a r t e r t=-24.52 P <. 01 4th q u a r t e r t=-25.05 P <• 01 * p <.01, ** p <.001 Appendix 39 T - t e s t f o r s t a n d a r d i s e d grade and language of i n s t r u c t i o n By number of assignments submitted 6 or more N.S. 8 or more N.S. 10 or more t=-2.54 P <.05 12 or more t = -3 .86 P <. 01 6 or l e s s N.S. 4 or l e s s N.S. 2 or l e s s N.S. By time i n t o the course 1st q u a r t e r t = -2 .00 P <.05 2nd q u a r t e r N.S. 3rd q u a r t e r N.S. 4th q u a r t e r N.S. * 2 < - 0 1 > * * E <•001 140 Appendix 40 Correlat ions between deviation and turnaround By number of assignments submitted 6 or more .08 8 or more .07 10 or more .09 12 or more .39* 6 or less .01 4 or less .00 2 or less -.01 By time into the course 1st quarter .02 2nd quarter .02 3rd quarter .03 4th quarter .03 * p <.01, ** p <.001 Appendix 41 Correlations between completion and grades By number of assignments submitted 6 or more -.10 8 or more -.24* 10 or more -.37** 12 or more -.55** 6 or less -.07 4 or less -.11* 2 or less -.07 By time into the course 1st quarter -.05 2nd quarter -.03 3rd quarter -.02 4th quarter -.04 * 2 <«0 1> ** 2 <- 001 141 Appendix 42 Correlations between deviation and grades By number of assignments submitted 6 or more -.14 8 or more -.18 10 or more -.17 12 or more -.47* 6 or less .02 4 or less .03 2 or less .05 By time into the course 1st quarter -.01 2nd quarter .02 3rd quarter -.01 4th quarter -.02 * 2 <- 0 1f ** 2 <.001 Appendix 43 Correlat ions between turnaround and grades By number of assignments submitted 6 or more -.28** 8 or more -.40** 10 or more -.28* 1 2 or more -.30 6 or less -.13* 4 or less -.13* 2 or less -.15 By time into the course 1st quarter -.14** 2nd quarter -.13 3rd quarter -. 14** 4th quarter -.16** * 2 < - 0 1 > * * 2 <-001 142 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0055760/manifest

Comment

Related Items