Marketing Implications Associated with Non-Returning Students in University Certificate Programs

This study examines the attitudes and future plans of students who discontinued their studies in university certificate programs. These non-returning students were separated into three groups: opt-outs, stop-outs, and drop-outs. Two separate surveys were undertaken; one survey consisted of students enrolled in a certificate program offered by the University of Saskatchewan, and the other consisted of students enrolled in several certificate programs offered by the University of Manitoba. The study examined three primary questions: What proportion of non-returning students commenced their certificate studies with the intention of completing the program and earning the certificate? What proportion of non-returning students intend to return to complete their  program? What attitudes do these students have toward the certificate program in which they had enrolled? Several findings of special significance are reported. First, approximately one-quarter of the students in this study indicated that they entered their certificate program with no intention of completing the program. Second, a significant proportion of the students in this study indicated an intention to return and complete their certificate program. Finally, respondents indicated a very high level of satisfaction with their university experience despite the discontinuation of their studies. The implications of these findings for the marketing of certificate programs are examined and suggestions for further research are offered.

Revue canadienne de l'éducation permanente universitaire Vol. 31, No 2, automne 2005 1. What proportion of non-returning students commenced their certificate studies with the intention of completing the program and earning the certificate?
2. What proportion of nonreturning students intend to return to complete their program?
3. What attitudes do these students have toward the certificate program in which they had enrolled?
Several findings of special significance are reported. First, approximately one-quarter of the students in this study indicated that they entered their certificate program with no intention of completing the program. Second, a significant proportion of the students in this study indicated an intention to return and complete their certificate program. Finally, respondents indicated a very high level of satisfaction with their university experience despite the discontinuation of their studies. The implications of these findings for the marketing of certificate programs are examined and suggestions for further research are offered.

INTRODUCTION
One of the ongoing challenges facing university continuing education units is the effective marketing of their programs. Faced with the pressures of institutional expectations for revenue generation and the growing pressure of competition from other providers, there is a need for continuous evaluation of the effectiveness of our marketing efforts (Craven & DuHamel, 2000;Fong, 2001).
At the same time, a disturbing number of students frequently discontinue their studies, especially in certificate programs, which consist of numerous courses that must be completed over an extended time period. Even successful efforts at marketing may be insufficient if we are unable to retain enough students in our programs. Many questions need to be answered. For example, do we know why students are motivated to enrol in our programs in the first place? Do we know why some students choose to discontinue their studies prior to completion? Is it appropriate to consider all non-returning students to be "drop-outs"? Is it possible that some students enrol for reasons other than to complete the program and earn a certificate? In fact, these students might be "opt-outs," rather than drop-outs, because they may consider their academic goals to have been satisfied, even though they did not complete their program. Might this be a significant factor in the case of non-returning students enrolled in certificate programs?
Moreover, is it appropriate to regard all non-returning students who started their studies with the intention of completing them to be drop-outs? Could a significant number of them merely be engaged in a hiatus from their studies and be planning to return? Such students might be "stop-outs," and it would be worthwhile to encourage their return. However, many university continuing education units eliminate non-returning students from their mailing lists after a defined period of non-attendance. In light of the costs involved, this practice is understandable, but are there hidden costs associated with this practice? Is an opportunity being lost to encourage the return of students who have discontinued their studies?
There is another reason to be concerned about the significant number of students who discontinue their certificate programs: the influence of word of mouth on the opinions of prospective students. We do have some evidence that those who participate in certificate programs offered by university continuing education units are generally satisfied with their programs (Hutton, 1997). What is uncertain, however, is the attitude of the students who discontinue their certificate studies. Do they hold positive or negative views toward their program? If they have negative views, might they discourage potential students from enrolling in certificate programs? Martin and Moore (1991) concluded that positive word-of-mouth promotion has a significant impact on the recruitment of new students. Accordingly, the attitudes of students who discontinue their studies deserve examination. Athanassopoulos, Gounaris, and Stathakopoulos (2001) contrasted defensive marketing strategies with offensive marketing strategies. Whereas offensive strategies seek to recruit new customers, defensive strategies are aimed at retaining existing customers. They concluded that it is more costly to recruit new customers than to retain existing ones. Moreover, they observed that loyal customers are likely to engage in favourable word-of-mouth responses. What are the implications of these ideas for the marketing of university certificate programs? Do we invest enough in our defensive marketing efforts aimed at non-returning students in certificate programs?
This study presents the results of two separate, but related surveys that sought to provide some answers to these questions. Interestingly, the two surveys were planned and conducted completely independently of each other. It was not until the results of the first survey were presented at a recent CAUCE conference that this author learned of the second study, which had been undertaken at the University of Manitoba. Perhaps of even greater interest is the surprising level of similarity in their results.
The first survey, undertaken at the University of Saskatchewan, was specifically concerned with students enrolled in the Certificate in Adult and Continuing Education (CACE) program. The CACE program is an independent-study (correspondence) program offered by the University of Saskatchewan in collaboration with the University of Manitoba, the University of Alberta, and the University of Victoria. The program originated in 1988 and has been remarkably successful. To date, it has had more than 1,000 graduates, and the combined annual enrolment for the four partner institutions has exceeded 1,200 registrants, attracting students from across Canada, as well as from other countries. However, like many independent study programs, it suffers from a high drop-out rate (see, e.g., Bates & Poole, 2003;Garrison, 1989;Holmberg, 1995;Moore & Kearsley, 1996). At the University of Saskatchewan, there has been an apparent drop-out rate of approximately 52%. That is, approximately 52% of the students who registered for at least one CACE course over the past five years discontinued their studies prior to completing the program. A non-returning rate of this magnitude places a heavy burden on recruitment efforts to ensure sufficient participation and, thereby, sufficient revenues to make the program viable. But is this drop-out rate real or apparent? Is it possible that some of these students intend to return to complete their program after a period of non-participation? This is one of the central questions examined in this study.
The second survey, conducted at the University of Manitoba, involved students registered in four different certificate programs: the Certificate in Applied Counseling (CAC); the Certificate in Adult and Continuing Education (CACE); the Certificate in Management and Administration (CMA); and the Certificate in Human Resource Management (CHRM). All of these programs are offered through on-campus, face-to-face instruction, with the exception of the CACE program, which is available in both face-to-face and distance-education delivery modes. Completion rates for these four programs were examined by looking at a seven-year comparison (from 1990/91 to 1996/97) of the number of new admissions with the number of graduates. The resulting estimates were 26% completion for CAC, 33% for CACE, 60% for CMA, and 62% for CHRM. Although these rates vary considerably, the low rates associated with the CAC and CACE programs are of particular concern. When fewer than 50% of students are completing their programs, it is important to examine the reasons behind this behaviour.
Based on these data, this study set out to answer three questions: 1. What reasons led these students to enter a certificate program? In particular, what percentage of non-returning students entered a certificate program with the intention of completing the program and being awarded a certificate?
2. What percentage of non-returning students who originally intended to complete the program still intend to complete it?
3. What are the overall impressions of non-returning students toward the certificate program in which they registered? In particular, do they hold positive, negative, or neutral attitudes toward the program? Do their attitudes differ depending upon whether or not they intend to return to complete the program?

REVIEW OF RELATED LITERATURE
There is a substantial body of research associated with the phenomenon of "drop-out" from post-secondary educational programs. Moreover, it has been widely acknowledged that this level of drop-out is of considerable significance (see Barefoot, 2004;Manski, 1989;Miller, 1991;Moore & Keasley, 1996;Murtaugh, Burns, & Schuster, 1999;Polinsky, 2003). A number of published studies have reported that post-secondary educational programs suffer from a high rate of student attrition-in both traditional face-to-face instruction and distance education programs. For example, Burley, Butner, and Cejda (2001) reported that approximately 50% of students who enter undergraduate degree programs actually complete those programs. A similar result was reported by Desjardins, Ahlburg, and McCall (2002). Summers (2003) reported that community college attrition rates have consistently been found to be high and typically are higher than four-year degree programs. Conklin (1993) found that less than 10% of community college students earn a degree or certificate. Similarly, a study by Dennison, Forrester, and Jones Revue canadienne de l'éducation permanente universitaire Vol. 31, No 2, automne 2005(1982 noted that students enrolled in degree-credit studies at the University of British Columbia and the University of Victoria demonstrated a "modest rate of degree completion" (p. 53). As a result, a great deal of research has been carried out to identify the factors that contribute to drop-out and to investigate ways to reduce it. One of the major challenges associated with this area of research has been the recognition that drop-out behaviour in post-secondary studies is a complex phenomenon that typically results from multiple causes (see Bonham & Luckie, 1993a;Burley et al., 2001;Desjardins et al., 2002;Summers, 2003). Woodley (2004) observed that two general approaches have characterized research into the phenomenon of drop-out. The first involves surveying students who have discontinued their studies to ask them the reasons for their decision. For example, Grayson (1997) surveyed students who had enrolled for first-year studies at Atkinson College but did not return for a second year.
Of particular interest to the present study was his conclusion that students who had enrolled in their studies with the desire to complete a degree were far more likely to persist than those for whom degree completion was not a priority. The second approach to studying drop-out utilizes statistical analysis methods to investigate the relationship between certain student characteristics and persistence/drop-out. Much of this research has been influenced by models developed by Spady (1971), Tinto (1975), and Bean and Metzner (1985). Woodley (2004) described one study that investigated the statistical relationships of 110 predictor variables to drop-out behaviour.
Unfortunately, there are three problems associated with much of the research on drop-out behaviour in post-secondary education.
1. Much of this research has focused on traditional-age, full-time, oncampus undergraduate students. Thus, are the results of these studies generalizable to part-time adult learners? A number of reports have proposed that they are not (see, e.g., Bean & Metzner, 1985;Horn & Carroll, 1996;McGivney, 2004;Miller, 1991;Murtaugh et al., 1999;O'Toole, Stratton, & Wetzel, 2003;Rovai, 2003;Webb, 1989). Accordingly, they are of questionable value in understanding drop-out among adult learners in part-time study, especially those studying by means of distance education.
2. Most of the research has focused on drop-out from undergraduate degree studies. Regrettably, there is a dearth of research on participation in university certificate programs (Long, 1992;Robinson, 1991). It may be that students enrol in certificate programs for the same reasons they enrol in degree studies. Similarly, they may drop-out of certificate programs for the same reasons they drop-out of degree studies. But there is little, if any, evidence to substantiate these assumptions. 3. There appears to be an implicit assumption underlying much of this research that all students who enrol in post-secondary studies do so to earn a credential and, therefore, drop-out can be equated with failure. For example, synonyms that have been used for drop-out include wastage, attrition, and student mortality (Woodley, 2004). However, there is substantial evidence that students enrol in post-secondary education for multiple reasons and that completion of a credential is only one of those reasons. For example, Grayson (1997) reported that approximately 20% of the students in his study indicated they had enrolled without any intention of completing a degree. Manski (1989) proposed that enrolment in post-secondary education should be considered an "experiment" in which students do not know whether completion will be feasible or desirable. McGivney (2004) suggested that adult learners may well take courses out of personal interest, without any interest in or intention of completing a degree. Nonetheless, she observed that such students were still classified as "non-completers," even though they may have satisfied their educational goals. Sewall (1984) reported that approximately 50% of the adult students in his study indicated they had enrolled in undergraduate studies for the purpose of completing a degree. Whiteley (2002) found that 17% of students in her study did not plan to complete their degree studies. Therefore, it is important to recognize that some students may commence their studies with no intention of earning a credential or may decide during the course of their studies that they can achieve their personal goals without completing the entire program. Consistent with this view, Polinsky (2003) proposed that there is both positive and negative attrition. Positive attrition occurs when students leave an educational program prior to its completion but having satisfied their personal goals. Negative attrition occurs when students both fail to complete their program and satisfy their educational goals. This suggests that we need a better understanding of the nature of positive attrition, including the proportion of non-returning students whose behaviour is consistent with this concept.
It is also important to recognize that students may "stop-out" from their studies, with the intention of returning to complete their program. One of the earliest published references to stop-out is included in a study by Pascarella, Duby, Miller, and Rasher (1981). Over the past 25 years, growing attention has been paid to the stop-out phenomenon. For example, Bonham and Luckie (1993a;1993b) proposed three categories of non-returning students: drop-outs, opt-outs, and stop-outs. They defined "drop-outs" as students who had not achieved their academic goals and no longer intended to complete their program, "opt-outs" as students who had achieved their academic goals, despite not completing their studies, and "stop-outs" as Revue canadienne de l'éducation permanente universitaire Vol. 31, No 2, automne 2005 students who had not yet achieved their academic goals and intended to complete their program. They conducted a telephone survey of non-returning students and reported that 73% of those surveyed were stop-outs. In other words, almost three-quarters of them intended to return to complete their studies. However, it is important to note that Bonham and Luckie investigated students' declared intentions to return. Other investigators (e.g., Burley et al., 2001) have defined stop-outs as students who actually re-registered. Interestingly, Burley et al. (2001) found that 60% of the non-returning students in their study were stop-outs and that a number of them had actually discontinued their studies for several academic terms before returning. Thus, the periods of stopping-out can extend over a significant period of time for some students. Desjardins et al. (2002) reported that 61% of the students in their study stopped-out at least once during the course of their degree studies. In light of such studies, there has been a growing recognition that stop-out behaviour is significant and that those who discontinue their studies should not automatically be assumed to have dropped-out.

Survey #1 -The University of Saskatchewan
The survey process began by examining the academic records of all students registered in the CACE program at the University of Saskatchewan during the 1997/98 to 2001/02 academic years. Of these, 86 students who had registered for at least one course in the program during this period did not register for any CACE course in the 2002/03 academic year. This represented a non-returning rate of 52%. These 86 students constituted the initial survey sample, and attempts were made to contact them to invite them to participate in it. However, many of them had moved, and there was no way of contacting them, and a number of other students declined to be interviewed. In the end, 26 of the original 86 students were contacted and agreed to participate in the survey, for a participation rate of 30%.
Survey participants were interviewed by telephone, and a standard interview protocol was employed. Based upon their responses, participants were sorted into three categories: drop-outs, opt-outs, and stop-outs. Drop-outs were those who indicated they had entered the program with the intention of completing the CACE certificate but had since decided not to complete the program, even though they had not achieved their educational goals. Optouts were of two types: the first type consisted of those who had enrolled in the program for some reason other than to complete the program and graduate. In other words, these students had opted-out of the program from the outset. The second type consisted of those who had enrolled with the intention of completing the program and graduating but who felt that their goals had been met, even though they had not completed the program. In other words, they had changed their original goal and no longer intended to complete the program. Stop-outs were those who had entered the CACE program intending to complete it and were still intending to do so.

Survey #2 -The University of Manitoba
The subjects selected for this survey were students who had enrolled in one of four certificate programs but had not re-registered for at least one full academic year prior to the 1999/2000 academic year. These "non-returning" students were of two types: those who had exceeded the allowable time limits for completion of the program (classified as "inactives") and those who were still within the allowable time limits (classified as "actives"). Inactive students were sent a survey questionnaire, whereas active students were interviewed by telephone, with an interview protocol based upon the survey questionnaire. Students who were mailed the questionnaire and did not respond to the first invitation to participate were sent a follow-up request. A total of 82 students agreed to participate out of the 263 students invited to do so, representing a response rate of 31%. Of the 178 students contacted by telephone, 109 agreed to participate in the survey, a response rate of 61%. Accordingly, 191 students agreed to participate in the survey out of the 441 who were invited to do so, which represented an overall response rate of 43%.
Respondents were sorted into three categories: stop-outs, opt-outs, and drop-outs. The criteria used to categorize students in this study were identical to those used in the University of Saskatchewan survey.
Because these students were enrolled in four separate certificate programs, Chi-square analyses were done to determine whether the respondents in the various certificate programs differed significantly in either their non-returning category (i.e., drop-out, opt-out, or stop-out) or their assessment of their academic experience in the program. No significant differences were found on either variable. Therefore, it was concluded that it was appropriate to ignore the independent variable of the different certificate programs and treat all respondents as being from a single population.

Survey #1 -The University of Saskatchewan
Respondents were asked to indicate their educational goals upon first enrolling in the CACE program. Four response choices were provided: (1) to complete the CACE certificate, (2) to complete one or more courses of particular interest, (3) to prepare for new career opportunities, (4) to develop skills to do my job better. Respondents were allowed to select more than one Revue canadienne de l'éducation permanente universitaire Vol. 31, No 2, automne 2005 response. A total of 19 students (73%) indicated that one of their goals was to complete the CACE certificate. Of these students, 11 (58%) indicated they were still intending to complete the program.
Respondents were sorted into three categories: stop-outs, opt-outs, and drop-outs. Of the 26 respondents, there were 11 (42%) stop-outs, 8 (31%) opt-outs, and 7 (27%) drop-outs. Of the 8 opt-outs, all but one indicated they had entered the program with no intention of completing it. In other words, only one respondent had commenced the program intending to completing it but had subsequently changed academic goals.
Respondents were asked to assess their academic experience in the program by selecting one of the following responses: excellent, very good, good, poor, unsatisfactory. Of the 26 respondents, 3 provided no response. Of the remaining 23 respondents, 11 (48%) indicated their experience was either excellent or very good; 9 (39%) indicated their experience was good; and 3 (13%) indicated their experience was either poor or unsatisfactory.
In addition, respondents were asked whether they would recommend the CACE program to other students. Of the 23 students who responded to this question, 22 (96%) indicated they would recommend the program to other students. An examination of the academic experience ratings by respondent categories produced no significant interaction effects. That is, there were no significant differences in academic experience ratings between drop-outs, opt-outs, and stop-outs.

Survey #2 -The University of Manitoba
Respondents were asked to indicate their primary goal upon entering their program. Four response choices were provided: (1) to take interesting courses regardless of what program they belonged to; (2) to complete the program and graduate; (3) to take one or more courses before deciding whether or not to complete the program; (4) to take one or more courses but not complete the program. A total of 145 respondents (76%) indicated that their primary goal had been to complete the program and graduate. Of these, 47 (32%) indicated they still intended to complete their program.
Respondents were sorted into three categories: stop-outs, opt-outs, and drop-outs. Of the 191 respondents, 34 provided insufficient information to permit assignment to a category. Of the remaining 157 respondents, there were 47 (30%) stop-outs, 89 (57%) opt-outs, and 21 (13%) drop-outs. Of the 89 opt-outs, 44 (49%) had entered the program with no intention of completing it, and the remaining 45 had intended to complete it but had subsequently changed their academic goals.
Respondents were asked to assess their academic experience in the program by selecting one of the following responses: definitely satisfac- tory, satisfactory, mixed, unsatisfactory, definitely unsatisfactory. Of the 191 respondents, 67 provided no response. Of the remaining 124 respondents, 108 (87%) indicated their experience was either definitely satisfactory or satisfactory; 14 (11%) indicated their experience was mixed; and 2 (2%) indicated their experience was either unsatisfactory or definitely unsatisfactory. An examination of the academic experience ratings by respondent categories produced no significant interaction effects. Unfortunately, a significant number of those in the drop-out category did not respond to the question regarding their assessment of their academic experience. Thus, a further analysis was done of drop-out students' responses to a related question. Students who had indicated they did not plan to return to their program were asked whether dissatisfaction with the program was an important factor in their decision. They were invited to indicate which of the following responses was appropriate: very important; somewhat important; not very important; not at all important. Two students did not provide a response. Of the remaining 19 drop-out students, 5 (26%) indicated that dissatisfaction with the program was an important factor in their decision not to return. The remaining 14 (74%) respondents indicated it was either not very important or not at all important. Accordingly, it was concluded that the majority of drop-outs were not discontinuing their studies because of dissatisfaction with their program, but for other reasons.

Limitations of the Study
A number of limitations to this study must be acknowledged. First, there is the risk that self-reports may not be completely honest. A number of studies have warned that students who discontinue their studies may be inclined to offer socially acceptable reasons for their choices rather than reveal their actual reasons (see McKeown, MacDonell, & Bowman, 1993;Walker, 1999;Woodley, 2004). In the case of the present study, this might result in nonreturning students stating they had never intended to complete their certificate program but had entered the program for other reasons. This might be a face-saving way to explain their failure to complete the program. In such an event, they would have been classified as opt-outs rather than drop-outs. Accordingly, the relatively low level of drop-outs in the two surveys (27% for the University of Saskatchewan survey and 13% for the University of Manitoba survey) may actually be an underestimate of the true drop-out rate.
Second, like the study by Bonham and Luckie (1993b), students in this study were classified as stop-outs based upon their declared intention to complete their studies. Some other studies, such as Burley et al., (2001 defined stop-outs as those who actually returned to their studies after a period of non-attendance. As noted in the first limitation, students may offer more self-flattering responses, and although they have declared an intention to return to their studies, they may not be serious about it. Accordingly, there is the possibility that some respondents in this study who were classified as stop-outs will actually turn out to be drop-outs.
Third, the sample size is modest, especially in the case of the University of Saskatchewan survey. Nonetheless, it is important to note that the two surveys produced remarkably similar results.
In light of these limitations, the following observations and conclusions are offered.

Reasons for Entering Certificate Program Studies
In both studies, a significant number of students entered a certificate program with no intention of completing the program and earning a certificate. This was true for 27% of the respondents in the University of Saskatchewan survey and 24% of those in the University of Manitoba survey. This finding is somewhat surprising as it suggests that fully one-quarter of students who enrol in a certificate program can be expected to discontinue their studies even if every student were fully satisfied with their academic experience. This result should be explored in further research to determine whether the results of the present study are atypical.

Do Students Intend to Complete Their Certificate Program?
A high proportion of respondents indicated an intention to return to complete their program. In the University of Saskatchewan survey, 58% of the non-returning students who had started the program with the intention of completing it still intended to do so. For the University of Manitoba survey, the somewhat more modest result (32%) still represented a significant portion of the non-returning students. Of course, it is not certain if they will follow through on their intention.
Follow-up analyses were conducted at the University of Saskatchewan in the 2003/04 and 2004/05 academic years to determine how many of the 86 non-returnees identified in the study did re-register for the program. Six of them returned for the 2003/04 academic year and another three for the 2004/05 academic year. Accordingly, while 58% of the 19 certificate-seeking respondents indicated an intention to return (the declared stop-outs), only 10% of the 86 non-returnees have actually returned to date. It is also interesting to note the length of the periods of non-attendance for these nine students. One had a hiatus of two years, five had a hiatus of three years, and the remaining two had periods of non-attendance of five and seven years. It is remarkable that they returned to their studies after such extended absences. Indeed, most certificate programs have restrictions upon how long students may take to complete these programs.
Most university continuing education units undertake regular "pruning" of their mailing lists. The results of this study suggest that this should be done with care-especially for students in certificate programs. Indeed, additional efforts to retain these students are worthwhile if even a minority of non-returning students can be encouraged to return. These results should be especially interesting to those involved in marketing certificate programs, in general, and distance education certificate programs, in particular, as they suggest that special efforts to recruit non-returning students may well be productive. Newsletters and mail outs to current and prospective students should be sent to non-returning students as well. In fact, the data reported here suggest that a higher "payoff " could be realized from targeting nonreturning students rather than prospective students. That is, a higher proportion of program registrants might be derived from promotional efforts directed at non-returning students than from a like number of prospective students who have had no direct association with the program.

Students' Overall Attitudes Toward Their Certificate Programs
It is interesting to note the very positive attitudes that students in this study held toward the CACE program. Overall, 87% of those in the University of Saskatchewan survey reported they would assess their academic experience in the program as good, very good, or excellent. In addition, 96% indicated they would recommend the CACE program to other students. Similarly, 87% of the University of Manitoba survey respondents indicated their academic experience in the program was either satisfactory or definitely satisfactory. This is a remarkably positive finding. One might expect to find this level of satisfaction expressed by students who are re-registering in their program of study. Conversely, in the case of non-returning students, one might expect to find more critical views being expressed toward their academic experience. In fact, a similar result was reported by Conklin (1993), who found that more than 90% of the non-returning students in her study would recommend their college to friends. This is particularly noteworthy since we know that word-of-mouth is a significant factor in the recruitment of adult learners. Former students can be the best advocates and promoters of our programs. Accordingly, it would be wise to utilize the good opinion of non-returning students to help promote our programs, even if they do not plan to complete their own studies. Efforts to keep them informed of new program develop- It is also worth noting that the University of Saskatchewan interviewer reported that a significant number of interviewees expressed appreciation for the interest being demonstrated in their plans and their attitudes toward the program. There appeared to be a "public relations" benefit arising from the contact. A similar finding was reported by Bonham and Luckie (1993b). This suggests that such follow-up contacts with non-returning students might be worth doing on a periodic basis.
The results of the present study suggest that, by and large, non-returning students tend to have positive views of their program, despite having discontinued their studies.

Suggestions for Practice
A number of suggestions for practice also emerged from the study results. First, surveys of non-returning students should be done on a regular basis. Many students who were contacted in the telephone surveys expressed appreciation for the interest being shown by the educational institution. Thus, these surveys can be used to identify ways in which certificate programs can be improved to better respond to the needs of their target audience. In addition, there is a public relations benefit to be gained by such contact.
Second, non-returning students should be kept on mailing lists for program brochures, newsletters, and other promotional materials to keep them informed of developments in their programs of study. They should not be removed from these mailing lists for at least five years, since they may be encouraged to serve as advocates and recruiters for the program, even if they do not plan to return to the program themselves. In the case of students who indicate they still plan to return to their studies, they should be kept on mailing lists for perhaps as long as seven years. The cost associated with doing so should not be measured only against the financial benefits of those students who actually return to the program. Rather, it should be recognized that non-returning students may contribute as program advocates.
Finally, these results should encourage continuing education practitioners to reflect upon some commonly held assumptions. For example, not all persons who register for certificate programs are doing so with the objective of earning a credential, and dropping-out is not necessarily a bad thing. In addition, perhaps we should be open to examining ways of "packaging" our certificate programs to target audiences who are interested in some parts of the program but not the whole package.

Recommendations for Further Research
The results of the present study are the basis of several recommendations for further research. First, additional studies should be undertaken to establish whether the findings reported here can be considered broadly representative of the experiences of other university continuing education units insofar as certificate programs are concerned.
Second, research should be undertaken to determine how many applicants to certificate programs have been significantly influenced in their decision to register for the program as a consequence of the positive experience of previous students.
Third, further research should examine how best to influence our current and former (graduate and non-returnee) students in certificate programs to serve as advocates and recruiters for our programs. Is it enough to keep them on our mailing lists or do we need to consider supplementary strategies to influence their behaviour?
Finally, further research is needed to examine whether different reasons are associated with the decision to drop-out compared with the decision to stop-out. This study found no such differences, but that may be a consequence of the relatively small sample sizes. By contrast, Malloch and Montgomery (1996) reported that students who were unlikely to return to their studies were more likely to indicate that the reasons for their decision were under the control of the institution (such as level of tuition) than to indicate reasons that were not under the institution's control (such as availability of financial support from employers). Accordingly, it is important for future researchers to separately analyze the responses of stop-outs from drop-outs. If we homogenize their responses, we may find that tuition levels, for example, are not a significant concern. However, if we look separately at the responses of each group, it may be that tuition levels are not a concern for stop-outs but are a definite concern for drop-outs. Without such differential analyses, we might fail to identify important factors contributing to drop-out.

A Final Word
The results of this study have significant implications for university continuing education units-especially insofar as the marketing of certificate programs is concerned. Hopefully, this study will be a useful guide to practitioners and will promote additional research relating to the attitudes and behaviours of non-returning students.