Enrolment and Retention in Adult Basic Education Programs Some Theoretical Implications of a National Study Follow-up Study
Ellen Long,
ABC CANADA Literacy Foundation, Canada
Abstract: Using the findings of a national follow-up study (n=338), this paper discusses the factors associated with nonenrolment and dropping out of literacy and upgrading programs. In addition to illuminating the complexity of low enrolment and high drop out rates, the paper moves toward a more integrated understanding of the structural and agency-related explanations for these patterns.
Introduction
Despite the need for higher basic literacy levels in the general adult population (Statistics Canada, 1996) and increased outreach efforts, recruitment and retention of adult basic education learners remain major challenges. It is a paradox that people with lower levels of formal education are the least likely to seek or receive further education and training (Livingstone, 1999). Indeed, only a small fraction (estimates range from five to ten percent) of eligible adults have ever enrolled in a literacy or upgrading program, and among those who do enrol, attrition rates are high (Quigley, 1997). And yet we have little information about the experience of people who attempt to access literacy or upgrading services or who enrol and then drop out.
The value of systematic follow-up, both to improve practice and to influence public policy, has long been recognized. However, the current structure and resource-base of the literacy field means that service provision has taken primacy over service evaluation. When follow-up is done, it is typically with people who have either completed or dropped out of programs. Little is therefore known about the full range of possibilities that can occur once individuals make initial contact with a literacy group. The number of people who call but who do not enrol is unknown, for example; and the reasons why someone would call but not sign up is only a matter of speculation.
This paper presents the results of a large-scale, systematic study that provides an unprecedented opportunity to look, on a national level, at the conditions that promote or deter successful participation in adult basic education. This was accomplished by conducting 338 telephone interviews with potential learners who had contacted one of fifty-five literacy groups across Canada.
This study aims to make a theoretical contribution by exploring and articulating a model of participation that encompasses multiple, interacting factors, where they have previously been seen as uni-dimensional and static. This will have the result of helping to close the theoretical gap in the false dichotomy between structure and agency that characterizes a great deal of sociological thought (Hays, 1994). Theoretically based explanations of participation will move the literacy field further along in being able to predict and, therefore, to influence enrolment and drop out rates. More theoretically sound models can form the basis for effective policy.
Literature critique, theoretical directions and research questions
Research on patterns of participation in adult literacy and upgrading programs is underdeveloped. Overall, empirical and theoretical research of any type is scarce and much of it is poor in quality by scholarly standards (NLS, 1996). There is an abundance of small, non-random samples that hinder effective demographic explorations and make it difficult to compare and quantify findings. Most studies are descriptive and anecdotal, rather than analytical and with notable exceptions (Beder, 1990; Quigley, 1990), lack explicit theoretical foundation. These problems go beyond the participation literature and, in fact, characterize the whole literacy field, which is in its theoretical infancy.
Though improving, this small body of literature exhibits theoretical and methodological constraints that have led to a conceptually fragmented, weak foundation for reflecting on patterns of enrolment and retention. Part of this weakness results from the fact that due to the difficulty in reaching certain segments of the population, particularly people who do not enrol at all, much of the research that informs current thought about nonenrolment has been formulated by studying the wrong types of people (Quigley, 1997).
A defining characteristic of the participation literature (and of popular understanding) is nicely reflected in the well-known metaphor about a group of philosophers who were trying to discover, in pitch darkness, the form of an elephant. Each describes the creature's nature in terms of the aspect examined-its trunk (snake-like), tusk (spear-like), ear (fan-like), leg (tree-like), side (wall-like), tail (rope-like), and so on. In the end, each of the obtained perspectives is only partly correct, but these partial explanations are treated as though they accounted for the whole picture. Alternatively, if perspectives are broader, factors are presented as unrelated. In thinking about patterns of enrolment, this problem in the literature makes it difficult to comprehend that multiple factors may be operating simultaneously though differentially, depending on the person and the situation, and that the factors themselves may be interacting with each other.
What is clear is that the factors influencing participation are complex, numerous, and much debated. The literature contains two principal types of explanations for nonenrolment in literacy programs: those focussing on structural reasons, essentially arising from the multitude of social, political or economic forces that shape people's lives; and those emphasizing human agency, the ways in which attitudes direct people's behaviour. A great deal of the literature characterizes people with low literacy skills as overwhelmed and passive in the face of structural barriers or stigma. Individuals are seen as passive and fearful; basically, the assumption is that people do not participate in literacy education because they are hiding their low literacy skills out of a sense of shame and stigma.
In contrast to the field of sociology, which has had an historical bias toward structure (Hays, 1994), the literacy field has been long biased toward agency, particularly in the form of "attitudes." In an attempt to rectify the bias toward passive and decontextualized attitudinal explanations, some researchers have focused exclusively on either structural barriers or have forcefully asserted ideological, agency-related explanations (Fingeret, 1983; Quigley, 1990). Even studies that try to combine structural and agency-related factors often end by stating the primacy of one over the other, in the absence of convincing empirical evidence (Thomas, 1990).
The consequences of focussing on one or the other of these partial factors can be quite problematic because to speak only about structural factors portrays people as passive victims with limited ability to think about or influence the direction of their lives. To focus only on agency-related factors risks either romanticizing people's behaviour or choices, or "blaming the victim" by not appreciating the context in which attitudes are formed and decisions made. On a theoretical level, it suggests that there is no relationship between issues of agency and structure; instead, people's actions are seen as either habitually patterned or floating freely above or outside of structural influences, and material forces are seen as existing outside of the creation and ongoing transformative influence of human beings. (As discussed in Hays, 1994).
The structure-agency debate has a long history in sociological thought, with many sociologists now realizing that the concepts are not only "slippery and contested" (Hays, 1994, p. 57) but in need of being theoretically integrated. If the ideal of integrating concepts of structure and agency is not yet fully unrealised within the field of sociology, it is not even on the distant horizon within the field of literacy. Trying to capture and articulate the interconnected complexity of both structural and agency-related factors is critical to this study because unique patterns of enrolment can not be truly understood outside of the social, cultural and economic contexts of people's lives, or outside of broad questions about programs themselves.
The contradictory and fragmented nature of the participation literature lead me to design a survey that was structured enough to explore certain themes that emerged from the literature, but was flexible enough to reveal a wealth of unexpected information. So for example, the survey included a number of open-ended survey questions, to ensure accurate data in some strategic areas, rather than categories constructed from potentially or incomplete or inapplicable research.
The current study attempts to make a significant step toward the ultimate goal of developing clearer theoretical models and integrated explanations of participation by exploring the following research questions:
Methods used to study enrolment and retention
This study involved doing telephone follow up with a large sample of Canadians, six-to-eight-months after their initial call to a literacy group. Help with sampling methods was enlisted through Statistics Canada. From a pool of 314 groups working in cooperation with ABC CANADA's LEARN campaign, a random sample of 60, stratified by province and territory, was drawn. The sample included community-based providers, school-boards, community colleges, phone referral lines, government literacy agencies, literacy networks, and urban and rural groups. Of the 60 groups, 55 (92 percent) agreed to ask enquiring callers in January and February of 1999 for permission to be called back.
We were able to reach 338 people (a response rate of 67%). Because the findings of this study do not reflect the experiences of both the people reluctant to discuss their situation and the most transient portion of the overall sample, enrolment figures may be overestimated, and drop out rates underestimated.
The data were analysed using SPSS; in classifying the survey data, I resisted the temptation to borrow terms uncritically from the broader literature on general adult education relating to colleges, universities, and continuing and distance education. These terms (e.g., Cross 1981) assume a well developed delivery infrastructure, and a participant population with relatively high degrees of disposable income. Research in the literacy field can assume neither of these things. The methodology, along with the sample size of this study allows us to readily see the full range of factors that could operate in respondents' decisions to enroll or drop out of programs. These factors were post-coded as program/policy-related, socioeconomic-circumstantial (structural) or cognitive-emotive (agency-related).
Findings
Less than half of the respondents who inquired about literacy or upgrading services enrolled in a program. Program/policy-related (PPR) factors were the main factors driving nonenrolment for the largest percentage of callers (43 percent). In rank order, these factors include not being called back by a program, long waiting lists, inconvenient class locations and times, wrong content or teaching structure, and having to pay for the program or tutor.
Socioeconomic-circumstantial (SEC) barriers were reported by the next largest group of those who did not enrol (30 percent). Contrary to popular understanding, cognitive-emotive (CE) reasons (e.g., fear) were least likely (15 percent) to be cited as the main factor for not enrolling. While it is important to look at main reasons for nonenrolment, most respondents readily cited a complex constellation of contributing reasons that reflects both structural (PPR and SEC) and agency-related (CE) factors in roughly equal proportion.
Of those respondents who enrolled, one-third had dropped out by the six-to-eight-month mark. Socioeconomic-circumstantial factors were the main reason reported by more than half of those who dropped out of programs. In rank order, job-related pressures, money problems, and childcare conflicts were the primary factors. Another quarter cited program/policy-related factors as their most important reason for dropping out. These factors include wrong program level, content or teaching structure, inconvenient location or program cancellation. Few people (6 percent) cited cognitive-emotive factors as the main reason for dropping out. As with nonenrolment, those who drop out cite multiple reasons, reflecting wide range of structural and agency-related factors.
While it is important to look at main reasons, most respondents cited a complex constellation of reasons for nonenrolment and dropping out, which reflects PPR, SEC and CE factors. This suggests the necessity to develop complex explanatory models of multifaceted strategies that address a wide range of factors. Most factors had a high degree of demographic variability.
Discussion
What empirical contribution does this study make toward the possibility of integrating structure and agency within the participation literature? First, in a clear departure from previous studies, the methodology of this study facilitated the emergence of a broad framework for understanding nonenrolment by revealing a wide variety of explanatory factors without privileging either structural factors or agency-related factors, in the absence of empirical evidence. Asking open-ended questions about nonenrolment to a large sample of respondents, allowed a complex picture to emerge, which included both types of factors. Had the study only probed factors suggested by the existing literature (e.g., inconvenient location, childcare, fear) it may have reproduced a narrow set of methodologically induced findings. Instead, twenty-five distinct reasons were cited by callers, and these were later classified into the three factor areas of PPR, SEC and CE.
In many cases, the reasons for nonenrolment given by respondents confirmed expectations. However, an interesting finding of this study was the appearance of reasons not yet discussed in the literature. New categories included "didn't hear back" and "unhelpful/unknowledgeable contact." The emergence of these categories can be partially traced to the fact that this study spoke with people who had phoned literacy referral lines but had not enrolled in a program. These reasons- quite stunning in their import- never would have emerged in studies using samples of people who have already enrolled, which is the most typically studied group.
The strong explanatory emphasis in this study on PPR factors would not have been possible had I just stayed within the participation literature, which underemphasizes PPR factors in favour of CE and SEC explanations (Livingstone, 1999). An emphasis on PPR factors is particularly important in looking at literacy participation because the literacy services themselves are profoundly uneven and underdeveloped.
Another interesting outcome of this study is that some of the reasons for nonenrolment that are widely accepted in the literature seemed much less salient in this study, most specifically some of the cognitive-emotive reasons such as dislike of school (Thomas, 1990). Overall, the findings of this study suggest that the literacy field has some way to go in understanding the full range of reasons that might influence enrolment. Quigley (1997) argues that people who have never contacted a literacy group may be quite distinct from those who participate in programs. Given the emergence of new categories and the low relevancy of others, this study suggests that people who call but do not enrol may be yet another distinct group.
Another way that this study contributes to a deeper conceptual understanding of nonenrolment is by demonstrating a way of beginning to look at the relative weight of factors and reasons. Whereas the existing literature presents factors either in rank order (Thomas, 1990) or in no particular order (Hoddinnott, 1998), this study asked a large sample of people to name the main reason they did not enrol, followed by as many contributing reasons as they wanted to add. This method allowed for a much more nuanced picture to emerge.
Overall, PPR reasons were the main factors driving nonenrolment for the largest percentage of callers (43 percent), followed by SEC reasons (30 percent) and, more distantly, by CE ( 15 percent) and "other" factors (11 percent). These findings have major implications for how the literacy field thinks about nonenrolment because they stand in sharp contrast to the common perception that potential learners are primarily deterred from participation because of fear and a sense of stigma. This perception is what led large numbers of participation researchers to focus on attitudinal factors to the exclusion of structural factors.
Still, while identifying main factors is important, in the end, composite reasons offer a more sophisticated picture of the multiple, interacting factors involved in nonenrolment. Although many reasons stand out-for example, "money worries" was by far the most-named factor, followed by "job-related reasons," "wrong content," and "worried or nervous about school"-there is nothing in this study to suggest that either structural or agency-related reasons are so dominant that the other can be ignored. The study found that similar percentages of people named factors that fall into each of the three factor areas (SEC, PPR, and CE), which clearly points to the need to put aside "either-or thinking" about structure and agency, and to develop a theoretical framework that integrates them. When this more complex picture is considered, the need for sophisticated theory and multifaceted solutions that can encompass a multiplicity of factors is readily apparent.
Although the measurement techniques used in this study ultimately help lay a firmer empirical foundation for theory-building than was formerly possible, the study still falls short of the ideal of being able to account for the specific interactions among multiple factors. A full-scale analysis of the simultaneity of factors was impossible because the sample size does not allow for factor analysis. Factor analysis would help provide a much clearer picture of which factors cluster together, thereby revealing their underlying dimension. Instead, we are still in the position of needing to speculate about certain relationships. While the study does take us many steps toward being able to do that by providing a stronger empirical and conceptual base. The main goal of this study-to integrate structural and agency-related explanations for nonenrolment-is not fully within grasp because many aspects of this integration remain to be demonstrated.
References
Beder. H. (1990). Reasons for Nonparticipation in ABE. Adult Education Quarterly, 40, (4):207-218.
Cross, K.P. (1981). Adults as learners: Increasing participation and facilitating learning. San Francisco: Jossey-Bass.
Fingeret, A. (1983). Social network: A new perspective on independence and illiterate adults. Adult Education Quarterly, 33, 133-145.
Hays, S. (1994). Structure and Agency and the Sticky Problem of Culture. Sociological Theory, 12, (1): 57-72.
Hoddinnott, S. (1998). Something to think about. Ottawa: ON. Ottawa Board of Education.
Livingstone, D.W. (1999). And the walls come tumbling down-Non-traditional learners in higher education. Paper presented at the International Symposium, University of British Columbia, Vancouver, B.C., August 16.
McGivney, V. (1990). Education's for other people: Access to education for non-participant adults. A research report. National Institute of Adult Continuing Education.
National Literacy Secretariat. (1996). Policy Conversation on Literacy Research. Ottawa: National Literacy Secretariat.
Quigley, Allan. (1997). Rethinking literacy education: The critical need for practice-based change. San Francisco: Jossey-Bass.
_________. (1990). Hidden Logic: Reproduction and Resistance in Adult Literacy and Adult Basic Education. Adult Education Quarterly, 40, (2). 103-115.
Statistics Canada. (1996). Reading the future: A portrait of literacy in Canada, Ottawa: ON. Statistics Canada, Human Resources Development Canada and the National Literacy Secretariat.
Thomas, A. (1990). The reluctant learner: A research report on nonparticipation
and dropout in literacy programs in British Columbia. Victoria, BC. BC: Ministry
of Advanced Education, Training and Technology.
[Home] [Search] [Conference Proceedings] [Back] [Top]