Page 37


Adult Learning as an Iceberg: Establishing the Knowledge Based on Self-Direction

As was stressed in Chapter One, self-direction in learning is not a new idea. Yet, while self-direction or self-education has been held as an ideal for adult education by various authors over the past several decades (e.g., Lindeman, 1926; Bryson, 1936; Snedden, 1930), only since the early 1970s have adult education researchers actively and systematically directed their efforts toward this area. This may be due, in part, to the relative newness of adult education as a field of study in North America. While some authors (e.g., Boshier & Pickard, 1979; Boyd & Apps, 1980) have described adult education as a "discipline," the idea of adult education as an area of academic inquiry is nonetheless quite recent. In fact, much of the seminal thinking about the study of the adult education field can be traced to the publication of Adult Education: Outlines of an Emerging Field of University Study (Jensen, Liveright, & Hallenbeck, 1964). Further, a content analysis of Adult Education (Dickinson & Rusnell, 1971) revealed that during the mid-to-late 1960s, the number of descriptive/opinion-oriented articles declined while the number of original research articles increased. We suggest that self-direction in learning, therefore, has actually been one of the first major areas within the field to undergo systematic and sustained inquiry. Indeed, Beder (1985) supported this view by suggesting that, along with the area of participation, self-direction has been the only other research area to be extensively examined within adult education.

Another possible reason for the relatively recent emphasis on self-direction as a research area is the inherent difficulty of studying learners outside of educational institutions. Since so many self-directed learning efforts take place outside of institutions, it can be quite difficult to examine self-directedness in a holistic way that considers social, cultural, political, and psychological dimensions of the concept. As will be demonstrated in this chapter, the knowledge base of self-direction in adult learning has evolved largely through the efforts of researchers to identify numerous, often creative,

Page 38

strategies for answering questions about the frequency and nature of self-directed learning.

While efforts to study self-direction in learning have escalated since the early 1970s, one individual who seems to have played a key role in laying the groundwork for this research as far back as the early 1960s is Cyril Houle. In addition to serving as major professor for Malcolm Knowles and Allen Tough, two of the seminal contributors to current thought on self-direction in learning, Houle is author of The Inquiring Mind, published in 1961. For this study, sometimes credited with sparking the current interest in self-direction, Houle interviewed 22 adult learning participants in Milwaukee, Wisconsin, during 1960. Using an informal interview schedule, he concluded that these individuals fell within three subgroups with regard to their reasons for engaging in a continuing education activity. The first group, which Houle called goal-oriented, consisted of individuals who viewed their participation as a means to some other end, such as career advancement or change. A second group, the activity-oriented, chose to participate in continuing education primarily for fellowship, or the opportunity to meet new people. Finally, a group that Houle referred to as the learning-oriented, was comprised of individuals who participated for the sake of personal enjoyment; these individuals viewed learning as an end in and of itself. As will be discussed later in the chapter, it is this latter orientation that led Tough to pursue an interest in what he initially referred to as "adult self-teachers."

The focus of this and the following two chapters will be a review and critical analysis of existing research related to self-direction in learning. In this chapter, we will introduce what we believe to be the three major directions that research in this area has taken to date. We will then highlight the first of these areas. In the following two chapters, we will look at each of the other two major research directions and will offer an assessment of the current state of this research area. Subsequent chapters will apply this knowledge to examine ways in which adult educators can utilize these findings in order to improve practice.


Despite the relatively recent emergence of self-direction in learning as a research direction for adult education, the picture resulting from this body of knowledge seems rather bright. Unlike many of the researchable areas within the field that have been largely unexplored, deliberate efforts have been made to bridge theory, research, and practice in self-direction. As a result, it has been possible to apply information gained from research both to improve practice and to further expand research and theory-building efforts. For

Page 39

example, in the 20 issues of Adult Education Quarterly appearing during the five year period between Fall, 1984 and Summer, 1989, a total of 16 articles dealt directly with self-direction in learning, while several other articles were related in a secondary way. Also, during the 1985 Adult Education Research Conference, over 11% of the paper sessions were on the topic of self-direction.

Nonetheless, while we are encouraged by the picture presented in this and the following two chapters, we believe that the future of adult education research in general depends on continuing research efforts. To this end, self-direction in learning can serve as a "miner's canary," something of a gauge for measuring the ability of adult education to develop, sustain, and utilize a substantial knowledge base. Beder (1985) has suggested that while self-directed learning is one of the two relatively well developed research areas in adult education, it is doubtful that this knowledge has been utilized to guide practice in a major way. While, as we have stated above, research has often been used to improve practice, Beder's point is well taken, and should be kept in mind when implementing self-directed learning in the future.

In trying to make sense of the research base on self-direction in learning, it may be helpful to classify the various types of studies that have contributed to the development of this research base. Caffarella and O'Donnell (1988) have offered one such scheme. According to this view, research can be classified along five categories:

  1. Nature of the philosophical position--conceptual perspectives on the process of self-directed learning;
  2. Verification studies--descriptive investigations of adults learning projects;
  3. Nature of the method of self-directed learning--questions of "how" people plan and implement learning projects;
  4. Nature of the individual learner--questions of who participates in self-directed learning and for what purposes; and
  5. Policy questions--pertaining to the educator, institutions, and society. (p. 40)

This classification is based on the content of research studies. Further, it emphasizes both empirical research "as well as serious conceptual articles" (Caffarella & O'Donnell, 1988, p. 39).

An alternative classification scheme is based on the type of research paradigm under which an investigation falls. Several years ago, we presented a classification scheme consisting of three major categories or "streams" of inquiry: descriptive learning projects investigations, research involving measurement of self-directed learning levels, and qualitative studies (Brockett, Hiemstra, & Penland, 1982). In our view, this "three streams" model still serves as an

Page 40

appropriate classification scheme, for the vast majority of studies on self-direction still fit within one of these categories.

The first major branch of inquiry grew out of the methodology utilized initially by Tough in his work, The Adult's Learning Projects (1971, 1979). This line of research, since replicated by numerous researchers with a wide range of adult populations, provides convincing descriptive evidence of the frequency of self-planning by adult learners.

A second major branch of research has essentially defined self-directedness as a personological, or personality-related variable. In this approach, written instruments are used to determine the extent to which individuals possess qualities associated with self-directedness. This branch of inquiry has led to a number of studies linking self-directedness with such concepts as creativity (Torrance & Mourad, 1978), self-concept (Sabbaghian, 1980), motivational orientation (Reynolds, 1986), life satisfaction (Brockett, 1985a), and intellectual development (Shaw, 1987). Also, several researchers have used instruments designed to measure self-directedness in a diagnostic capacity within the classroom setting (e.g., Caffarella, 1983; Kasworm, 1983; Savoie, 1980; Six, 1987; Six & Hiemstra, 1987). This branch of inquiry has been responsible for moving our understanding of self-direction in learning from basic description toward greater understanding of how such variables may be related to one another and, to a lesser degree, where changes in one variable may cause a change in the other. Chapter Four will highlight this branch of research.

A third approach to studying self-direction in learning has employed qualitative methods such as observation and interviewing in order to develop models that can help to explain the meanings and contexts of self-direction in learning during adulthood (e.g., Brookfield, 1981; Gibbons, et al., 1980; Spear & Mocker, 1984). This branch of research (which will serve as the focus for Chapter Five) has helped adult educators in theory-building, which, in turn, has influenced the ways in which researchers and practitioners now understand self-direction in adult learning.

It is important to bear in mind that while these three streams of research have evolved in a somewhat sequential manner, they are not distinct stages of research. In other words, newer methodologies have not replaced previous approaches. Rather, each stream of inquiry continues to serve an important role in addressing specific types of research questions relative to self-direction in adult learning. The remainder of this chapter will offer a closer look at the first major stream of research by focusing on selected studies that reflect the learning projects paradigm.

Page 41


The Work of Allen Tough

In 1965, Allen Tough completed his doctoral dissertation at the University of Chicago on teaching tasks performed by "adult self-teachers." He found that, while self-teaching implies a degree of independence or autonomy, the learning that occurs through self-teaching does not generally take place in isolation (Tough, 1966). Those individuals who engage in self-teaching are highly likely to seek the assistance of others, such as close friends and relatives, librarians, subject-matter experts, and fellow learners.

Tough, in this study, focused on individuals engaged in a self-teaching project. As an outgrowth of this work, he began to raise questions about the extent to which self-teaching is a part of the total range of an individual's learning activities. Subsequently, during 1970, Tough and several colleagues at the Ontario Institute for Studies in Education (Toronto) interviewed 66 adults in an attempt to examine "the highly deliberate learning efforts" of adults. Particular emphasis was placed on the "planning and deciding" aspects of the learning project.

Tough defined a learning project as "a series of related episodes, adding up to at least seven hours" where "more than half of the person's total motivation is to gain and retain certain fairly clear knowledge and skill, or to produce some other lasting change in himself" (p. 7). Similarly, an episode was defined as "a period of time devoted to a cluster or sequence of similar or related activities, which are not interrupted much by other activities" (p. 7). Examples of episodes might include reading a newspaper or a chapter of a book, visiting a museum, or attending a class. An episode had a definite beginning and concluding time, and to be classified as an episode, the information had to be retained for at least two days after the learning activity took place. This qualification excluded activities in which learning was intended to serve an immediate purpose, such as being able to assemble a piece of household furniture--an ability quickly forgotten. A learning project, then, was considered to be the total of all episodes that a person had undertaken in order to gain (and retain) some specific knowledge or skill.

Tough was very specific in stating that a learning project must encompass a minimum of seven hours. There were two major reasons for setting this seven hour minimum. First, this period of time is approximately equivalent to a traditional working day (excluding breaks and "down time"), which Tough thought a considerable investment of time for a single learning project. Second, seven hours was found to work well in interviews with subjects because it eliminated very brief activities but not major learning efforts. Additionally, this seven hours must have taken place within a six-month

Page 42

period. It is important to emphasize that the "seven hours within a six-month period" qualification was intended as a minimum expenditure of time. In reality, Tough found that the majority of learning projects uncovered by his survey team far exceeded this minimum.

Thus, the 66 subjects were interviewed about their involvement in learning projects over the previous year. These individuals were drawn from the following seven populations:

  1. Blue-collar workers;
  2. Women in lower level white-collar positions;
  3. Men in lower level white-collar positions;
  4. Beginning elementary school teachers;
  5. Municipal politicians;
  6. Social science professors; and
  7. Upper-middle-class women with preschool children.

To obtain data, a highly structured interview process was employed. Interviewers were trained in how to use the interview format and in how to identify learning projects. Thus, interviewees could uncover many different kinds of efforts while eliminating "borderline learning," such as brief episodes and activities in which learning was not the major reason for undertaking the effort. While the interview procedure was generally structured, there were certain points in the process where interviewers would share with subjects various "probe sheets" that prompted a recall of learning projects from a wide range of areas (e.g., child-rearing, gardening, music, collecting, and job-related knowledge and skills).

It was found that the "typical" adult had been involved in about eight different learning projects during the year prior to the interviews. The mean number of projects was 8.3; the median was eight. Of the 66 persons interviewed, all but one reported having been involved in at least one learning project. The highest median number of projects were completed by the social science professors (11.5) and elementary school teachers (9.0), while blue collar workers and municipal politicians had the lowest mean number of projects (5.5 and 7.0 projects, respectively). The mean number of hours spent on an individual learning project was 104 while the median was 81, a figure well above the minimum criterion set by Tough. Further, less than 1% of all learning projects were undertaken in order to gain credit. Why do adults make the choice to undertake deliberate learning projects? According to the individuals interviewed in the Tough survey, the major reasons included preparation for a job and maintenance of job skills, solving a specific problem or task on the job, gaining knowledge and skill about some aspect of one's personal development or responsibilities within the home, and learning out of curiosity or interest in a topic or as a leisure pursuit. Tough stressed that

Page 43

most learning projects are motivated by "some anticipated use or application" of what has been learned. This finding is quite consistent with other surveys of reasons for adult learning participation (e.g., Aslanian & Brickell, 1980; Cross, 1981; Boshier & Collins, 1985).

Probably the most important finding to emerge from Tough's study--certainly with regard to our current discussion--pertains to the question of who assumes responsibility for planning learning projects. By far the majority of projects identified in the Tough study (68%) were planned primarily by the individual learners themselves. Another 12% were planned by a group or its leader/instructor (e.g., formal classes), 8% were planned primarily by another person in a one-to-one situation (e.g., tutorial), and a nonhuman resource (e.g., a programmed instruction manual) served as the planner in only 3% of the projects. In the remainder of projects, no single type of planner could be clearly identified.

It is this finding about individual planning preferences that lies at the heart of the current emphasis on self-direction research. While self-direction has long been assumed to be a major goal of adult education, it was not until Tough's investigation that the impact of this preference for individual responsibility in planning was made apparent. In fact, Tough used the analogy of an iceberg to describe adult learning. Only a very small portion of each (i.e., an iceberg and learning) is clearly visible, while the rest lies beneath the surface. The point here is that the vast majority of what adults learn is not easily observed as through rates of participation in formal adult education programs.

More recently, Tough (1982) has expanded his focus from the notion of adults learning projects to the broader concept of intentional changes. Here, Tough looked at the ways people undertake major personal change in various aspects of their lives. Using a format similar to the learning projects interview protocol and via a team of interviewers, Tough found that 75% of the changes reported by interviewees fell into four areas: job, career, and training; human relationships, emotions, and self-perception; enjoyable activities; and changes in residence location. With regard to taking responsibility for intentional changes, Tough observed that on the average, "the person assumes about 70% of the responsibility for all the subtasks involved in choosing the change, planning the strategy, and implementing the change" (1982, p. 52).

While the intentional changes approach does not seem to have been embraced as widely as the learning projects paradigm (e.g., Caffarella, 1983a; Sisco, 1983), there have been a few efforts to build on Tough's initial study. Moore (1986), for example, found that most of the prison inmates in his survey undertook several important changes in a year, with self-directed learning serving as an important foundational aspect of the change. Lundgren

Page 44

(1988) interviewed 46 Type II diabetics in Montana and found that at least 85% had made at least one health-related change in the previous five years. Self-planned learning projects were an important part of the change process for many of the individuals interviewed by Lundgren, and reading was viewed as the primary learning activity. From these studies, it can be seen that while the thrust of the intentional changes paradigm is different from Tough's earlier work, the link to self-planned learning nonetheless remains.

With the publication of The Adult's Learning Projects in 1971, numerous researchers began to adopt the methodology used by Tough in conducting additional studies with different segments of the adult population. In the following sections, we discuss several of these replication studies. Our intent in these sections is to provide a feel for the diversity and flavor that have characterized the follow-up work to Tough's study. In reviewing these replications, it should be noted that despite considerable variation in both the total number of learning projects and in the total percentage of self-planned projects, the findings from the original Tough investigation are largely substantiated.

Mothers With Preschool Children

Johnstone and Rivera (1965), in their study of participation in adult education, found that mothers with young children had especially low rates of participation in institutional adult education programs. Of the seven groups studied by Tough, mothers with preschool-age children completed the third smallest number of projects during the study period. In one of the earliest learning projects replications conducted by someone other than one of Tough's associates, Coolican (1975) explored the total range of learning activities (institutional as well as noninstitutional) of a sample of 48 mothers in Syracuse, New York, whose oldest children had not yet entered school.

Data were obtained through the use of a two-part survey instrument. One part of the instrument was a questionnaire designed to obtain demographic information about the interviewee. The other instrument was an interview schedule adapted from the one used by Tough. All interviews were conducted by the researcher and took place in the subjects' homes. Each interview began with the researcher striving to develop a relaxed tone. After rapport was established, the structured interview began. On the average, interviews lasted 85 minutes.

Coolican found that the subjects in her sample had conducted an average of 5.8 projects per year with a mean length of 43 hours for each project. It was found that 66% of these projects were self-planned, a figure very similar to that of Tough. Nearly half of the projects were concerned with

Page 45

home and family issues. Other major areas of interest included hobbies and recreation (18%) and personal development (11%). With the demands of caring for small children and running a household, Coolican found that these women typically were only able to spend short amounts of time on subjects in which they were interested.

Several possible limitations of the study identified by Coolican serve to point out some of the potential constraints of the learning projects methodology. First, since the data were obtained from a rather homogeneous sample, results cannot be generalized to a more heterogeneous population. Second, there may be limitations in using the interview approach, especially with regard to the recollection of learning projects that had taken place several months prior to the interview. Third, because of an unwillingness of subjects to reveal certain kinds of information or an inability to verbalize responses, there may be instances where the responses of subjects do not accurately reflect these actual experience. Fourth, no attempts were made to verify the accuracy of interviewees' statements. Fifth, the sample was based entirely on willing participants and, thus, there is a possibility that those who chose to respond were more oriented to or interested in learning than those who did not participate in the study. These limiting factors do not detract from the importance of Coolican's findings. Rather, they serve to articulate some of the very real concerns that need to be taken into account when interpreting the findings of studies using the learning projects methodology.

Rural and Urban Adults

In the earliest large-scale study comparing the learning projects of adults within a particular geographical area, Peters and Gordon (1974) interviewed a total of 475 persons residing in Knoxville, Tennessee and a nearby rural county. At the outset, 15 interviewers received nine hours of formal training and were involved in several days of related activities such as readings, role playing, and practice interviews. The interviews consisted of the basic structure developed by Tough and lasted an average of one hour.

The 475 persons who were interviewed for the study ranged in age from 18 to 90, with a mean age of 41; the majority (61%) were male. Fifty-four percent reported having less than a high school education. Seventeen percent were high school graduates while the remaining 29% held either undergraduate or graduate degrees.

Peters and Gordon found that their subjects had conducted an average of 3.7 learning projects in the year prior to the study. This figure was higher for the urban subjects (4.1 projects) than for the rural interviewees (3.1 projects). Half of the subjects reported four or more learning projects during the study period; however, nearly 12% conducted no projects during this time.

Page 46

The number of projects for females and single people was slightly higher than for men and those who were married. Looking at age groups, persons in the 35-39 age group conducted the largest number of projects (4.8) while those in the 55+ age group conducted the fewest projects (3.7). It is interesting to note the difference between urban and rural subjects in the older group. While interviewees in the 55+ age group who lived in the city averaged 5.4 projects--somewhat higher than the average for the total sample--those who lived in rural areas averaged only three projects.

Consistent with the findings of Tough (1979) and Coolican (1975), learning projects conducted by persons in the Peters and Gordon sample were overwhelmingly self-planned. Sixty-six percent of these projects, in fact, were planned by the individual learner. Unlike Tough's findings, however, 12.6% of the Peters and Gordon sample were enrolled in formal courses, a finding attributed by the authors to the relatively high number of college graduates found within the sample. This figure is consistent with data from the National Center for Education Statistics, which found that in 1981, 12.8% of all adults participated in an adult education course of some type (Kay, 1982). Other findings reported by Peters and Gordon dealt with utilization of resources, obstacles to participation, and content areas studied by learners in the sample. Resources most frequently used included books, magazines, tools and raw materials, and other individuals. The obstacles most often perceived by learners were lack of time, money, motivation, or education, plus family conflicts. The content of the majority of projects focused on topics related to employment or recreational pursuits.

In a follow-up analysis of the Peters and Gordon data, Brasfield (1984) focused specifically on the link between learning projects activity and educational attainment. In this analysis, Brasfield found that those adults with more formal education reported a greater number of learning projects than those with less formal education. However, she also found that the highest proportion of self-planned projects were found among those with fewer years of formal education. While the figures for self-planning among those with high school diplomas/some college and those with at least a bachelor's degree were 72% and 74% respectively, the self-planning figure for those with 0-12 years of education was 86%. This is a major departure from previous and subsequent research supporting a link between self-direction and educational attainment. We believe that the findings lend empirical support to the idea, which will be stressed further in the next chapter, that self-direction holds much potential for serving learners traditionally considered to be "hard-to-reach."

Page 47

Older Adults

For the most part, learning projects research has been based on general adult samples, not focusing upon any specific age group within the population. Yet, in studies of formal adult education participation, an almost universally consistent finding has been that participation declines drastically among those persons age 55 and above (e.g., Kay, 1982). However, at least four studies employing the learning projects methodology have painted a somewhat more optimistic picture of the older adult as a potential learning participant.

The first study in North America to focus on the learning projects of older adults was Hiemstra's 1975 study in Nebraska. He examined 256 adults, at least 55 years of age, selected randomly from the voter registration cards in two communities and 18 rural townships, and from the membership list of a Hispanic community center. The sample for this study was predominantly white (88.7%) and middle-class blue or white collar (87.8%). Slightly more than three-quarters of the subjects (75.4%) lived in their own homes, and 63.3% were married while 25.4% were widowed. In terms of previous education, 66.3% were at least high school graduates and nearly 20% had graduated from college. The mean age of the sample was 68.1 years of age (Hiemstra, 1975, 1976a).

Using the learning projects interview schedule, seven trained interviewers spent between one and two hours with each of the interviewees. Hiemstra found that of the 256 individuals interviewed, 214 or 83.5% reported conducting one or more learning projects in the previous year. An average of 325 hours per year was spent on a mean of 3.3 projects. And, as with other learning projects studies, it was also found that the majority of projects (55%) were self-planned. The primary planner used by subjects in this study is shown in Table 3.1.

Table 3.1 Different Types of Primary Planner in Learning Projects
Type of Primary Planner Percent
A Group or its Leader/Instructor 20.45%
One Person in a One-to-One Situation                       10.30%
Material/Nonhuman Resource 03.95%
The Learner Him or Herself 55.15%
Mixed (No Dominant Type of Planner) 10.16%

With regard to various sub-groupings, there were some important differences. For example, a higher proportion of people 65 and older reported no learning projects in the year before the interview than did younger

Page 48

individuals. However, there were no significant differences across age groups between those who participated in more self-planned projects and those participating in fewer projects of this type. Finally, when asked to name the primary subject matter of their learning projects, there were no significant differences among the two age groups in terms of preferences for self-fulfillment subjects (Hiemstra, 1985b).

From this study, we believe there is evidence that older adults often are active learners and that most of this activity is reflected through the self-planned learning projects in which such individuals are engaged. This study suggests that "educators must learn how to remove their institutional blinders and recognize all the self-directed, independent learning going on and needed outside institutional structures" (Hiemstra, 1976b, p. 337).

A second study using the learning projects approach with older subjects was reported by Ralston (1981), who was interested in comparing black and non-black groups. Ralston interviewed 110 randomly-selected persons age 65 or older in Champaign-Urbana, Illinois. Among the findings of this study, Ralston found that her interviewees conducted a mean of 2.45 projects. She noted that "respondents who were white, white-collar, and had higher educational levels were involved in significantly more learning projects than other subgroups" (p. 237). Ralston did not seek to identify the primary planner of projects so it is not possible to determine the extent to which these projects were self-planned.

Hassan (1982) studied self-directed learning among a sample of 77 adults residing in Ames, Iowa. Of this sample, more than 37% were over 55 years of age. The 29 people in this older age group conducted a mean of 9.7 projects per year, a figure that was not significantly different from that of the 48 younger people. The Hassan study is discussed further in Chapter Four.

In a fourth study involving older adults, Estrin (1986) examined the relationship between life satisfaction and learning participation among older women living in two subsidized senior housing developments in southern Rhode Island. Of the 87 women in this study, 54 were widowed, the mean age was 72.5, and the mean educational level was 10.3 years. Estrin found a significant positive relationship between life satisfaction and learning activity, both in terms of number of projects and number of hours spent engaged in learning. She concluded that participation in learning can be viewed as a strategy for enhancing life satisfaction in the later years. This conclusion is further supported through research (Brockett, 1985a) that will be addressed in the next chapter.

Page 49

A U. S. National Sample

Probably the most comprehensive learning projects study to date was conducted by Penland (1977, 1978, 1979). For the most part, learning projects studies have focused on self-planned activities within specific groups; thus, the findings have not been generalizable to a larger population. By obtaining data from a U.S. national probability sample, Penland (1978) attempted to overcome this limiting factor by conducting a study that would address what he felt to be a "lack of a systematic behavioral foundation for a professional helping relationship" (p. 1).

The study sample was drawn from the total U.S. population of persons aged 18 or older, in a manner similar to the way in which public opinion polls are conducted. Since the study was conducted under different circumstances than other learning projects studies, the interview format was modified somewhat from the structured interview schedule developed by Tough (Penland, 1979). The instrument used by Penland was developed so that interviews could be completed in one hour and so those subjects not involved in learning could be accommodated "more efficiently" (Penland, 1978). For purposes of the study, Penland defined a continuing self-learner as "an individual (usually adult) who plans and designs an independent learning project," and independent self-designed learning was defined as "the overt behavioral evidence of a sequence of information processing episodes" (p.3).

The interview schedule used by Penland looked at three different sets of questions. The first set sought information about the "patterns and purposes" of adult learning activities. These questions explored the frequency and nature of learning projects. A second group of questions explored the use of resources as a part of the learning process. In addition, several questions focused on obtaining demographic information about the subjects.

Penland found that 78.9% of the sample perceived themselves to be "continuing learners" and that 76.1% had conducted at least one self-planned learning project in the year prior to the study. The average number of projects per person was 3.3 with a range of from one to 18. A mean of 155.8 hours was spent on each project. Penland attributed much of this figure to the involvement of the learner in planning aspects of activities that might otherwise be assumed by someone or something else.

The three reasons cited by subjects for preferring self- planned learning include the desire to learn what they choose and at their own pace, to maintain flexibility in their learning activities, and to structure their own project. According to Penland, the widespread emphasis on self-planned learning can be viewed as evidence to support the centrality of individualism in United States society.

Page 50

An important component of the Penland survey was the use of resources, particularly the library, in conjunction with self- planned learning projects. It was found that 73.2% of those persons who considered themselves to be continuing learners had deliberately looked up some piece of information during the seven days prior to the interview. However, only 14% of the sample stated that they used the library on a regular basis while another 60% never used the library for self-planned learning. Penland concluded that those persons who participate in lifelong learning activities need "shopping center" access to resources. This, he feels, requires the cooperation of adult educators, information brokers, and community development personnel. In Chapter Nine, we suggest some policies related to this need.

Low Levels of Self-Planning

The studies described above indicate a high degree of self- planning among learners from a variety of settings. Indeed, most learning projects studies have shown that a clear majority of projects are self-planned. However, some studies have found half or less of all projects to be planned primarily by the learners themselves.

Johnson (1973) interviewed 40 adults in Ft. Lauderdale, Florida, who had recently completed senior high school examinations. This group conducted an average of 14.4 learning projects, 50% of which were self-planned.

Miller and Botsman (1975), using a modified case study approach, examined the learning projects of nine Cooperative Extension agents in New York State. Subjects conducted an average of 12 projects. Forty percent of the projects were self-planned, while more than half of the projects consisted of workshops planned by others.

Umoren (1978) studied 50 individuals from two socioeconomic groups in Lincoln, Nebraska. In this sample, the 38 low income adults and 22 middle or high income individuals conducted an average of 4.7 learning projects, and about 40% of these projects were self-planned.

Field (1979) examined the learning activities of 86 low literacy attainment adults in the Brownstown, Jamaica, area. Of the 4.2 average number of projects conducted by these subjects, about 20% were self-planned and more than 50% were planned by group leaders. Field attributed these figures to a prevalence of literacy training and projects pertaining to religious activities, both of which tended to rely on group leaders.

Baghi (1979) looked at the learning project activity of 46 participants in adult basic education classes and learning centers near Des Moines, Iowa. These individuals reported an average of 6.6 projects per year, 57% of

Page 51

which were self-planned. In this sample black respondents reported a higher mean number of learning projects and of hours spent on learning projects activity than white interviewees, although the differences were not statistically significant.

While studies reporting low figures for self-planning are a clear minority of learning projects research, they make an important contribution to this research area in at least three ways. First, the findings indicate that such groups are, indeed, much more actively involved in adult learning activities than studies focusing on more "formal" participation typically indicate. This has clear implications for how educators of adults might work to reach traditionally "hard-to-reach" groups of adults (Brockett, 1983b). Second, these studies accentuate the need to use caution when espousing the value of self-planning for all adult learners. This concern is related to some of the myths presented in Chapter One. Finally, these findings can serve as an impetus for encouraging a critical look at both the contributions and potential limitations of the learning projects methodology.

Contributions and Limitations of Learning Projects Research

Has the learning projects methodology had a major impact on efforts to understand self-direction in adult learning? To address this question, it is necessary to assess both the contributions and limitations of this approach. Looking first at contributions, there are at least three ways in which learning projects research has benefitted the field of adult education. First, it has offered a method for studying the learning efforts of countless individuals who prefer to engage in learning activities outside of the formal educational institution setting. Because the methodology emphasizes learning that takes place both within and outside of institutions, it has provided a way of studying learning among individuals, such as older adults and those persons with little or no formal education, who have traditionally been considered "hard-to-reach."

A second, related contribution is that the learning projects methodology has served to redefine the meaning of adult education participation (Brockett, 1983b). For the most part, surveys of participation in adult education, such as those conducted by the National Center for Education Statistics (e.g., Kay, 1982) are based only on involvement in adult education courses. However, as the Tough study and subsequent replications indicate, courses comprise only a very small portion of all adult learning activity. The learning projects approach can provide data that are much more inclusive--and, as such, representative of the actual nature of adult learning--than data reported in most survey studies to date.

Still another contribution of learning projects research is that the approach

Page 52

has represented the first effort by adult education researchers to systematically study the concept of self-direction in learning. The learning projects approach seems to have served a "consciousness-raising" function for the adult education field, providing data to confirm that which was known intuitively for many years. If the extent to which the research approach has been replicated by other researchers can serve as testimonial, Tough's initial study would have to be considered one of the most significant pieces of research in all of North American adult education.

It is clear that the contributions of the learning projects methodology are quite substantial. At the same time, as with any other research approach, the paradigm has certain constraints that define and delimit the ways in which learning projects data may be interpreted. Brookfield (1981a), for example, has reviewed Tough's research and identified three limitations with the approach.

First, there are limitations in using the structured interview. While the structured interview format has the advantage that it can be replicated, Brookfield points out that the methodology can "run the risk of forcing the researcher's notion of what are admissible and appropriate substantive concerns" upon subjects without taking into consideration what they believe to be relevant (1981a, p. 115). In other words, the interview questions and the biases of the interviewer might prompt subjects to respond in a specific way.

Second, most of the subjects in Tough's 1971 survey were highly educated. Using the popular notion that education begets education, it could be argued that, since self-planning was clearly the predominant mode of learning in this study, it is not surprising for these learners to express a preference for this mode. Since then, however, several studies have investigated the learning projects conducted by adults of lower educational attainment than those in Tough's sample and have often found high degrees of self-planning (e.g., Brockett, 1983a, 1985c). Yet, as noted earlier, in the studies where the figures for self-planning were low, the samples tended to have a large number of "undereducated" adults. Thus, one could believe that a preference for self-planned learning will tend to be higher among those who have completed more formal schooling. But the relationship between self-directedness and previous education is not entirely clear and might, in fact, be related to other factors. In any case, this potential limitation is certainly a relevant concern when assessing the impact of self-directed learning research and, thus, is addressed further in Chapter Five.

Third, one might question the appropriateness of studying self-planned learning. Brookfield summarized this argument by stating that " . . . extended investigations into the nature of noninstitutionalized adult learning, while interesting, are inappropriate in the current economic climate.

Page 53

Instead of studying those adults who choose to ignore formal adult education provision and to devise their own learning schemes, we should be concentrating our efforts on retaining the loyalty and continuing participation of existing students as well as working to increase overall student numbers." (1981a, p. 117)

However, Brookfield is quick to negate this argument by stating that if, in fact, self-planning is as prevalent as the learning projects research suggests, adult education faces the challenge of identifying ways to support such efforts. It is this emphasis on supporting self-directed learners and their efforts that comprises the major focus of subsequent chapters in this book.

There are at least two other considerations that we believe are important when reviewing learning projects research. These have less to do with the methodology, per se, than with the way in which results could potentially be misinterpreted. When interpreting the results of learning projects studies, it is important to bear in mind that, while most researchers have used the Tough interview schedule as a point of departure, there is often considerable variation in the specific questions asked, the actual interview process, and the data analysis procedures. For example, most studies report figures for self-planning as a percentage of the total projects conducted by all subjects in the sample. However, Penland (1979) based his finding relative to self-planning on the percent of the sample who completed at least one self-planned project rather than the total percentage of projects that were planned by the learners themselves. This kind of variation is not a limitation in and of itself; however, those who utilize the findings need to be aware of these variations when interpreting data from these studies.

Another way in which learning projects data can be misinterpreted is that the approach is sometimes described as a qualitative research method. In reality, it is a quantitative descriptive survey approach. While interviewing is frequently associated with qualitative research, it is the way in which learning projects data are analyzed that places the method into a descriptive category. Fingeret (1982) points out that qualitative data are presented as quotations that have been taken from an interview or observation. She states that data are not the numbers of responses given to a particular question but, rather, are the actual words of the subjects themselves. In learning projects studies data are treated primarily through descriptive statistics.

Finally, Caffarella and O'Donnell (1988) have raised the concern that learning projects studies have "reached the point of dullness." They go on to offer the following suggestion: "We may need to put the element of surprise into future verification studies. If we know the answer in advance, do we really need to ask the question?

Page 54

The greater the surprise or astonishment with the findings of our self-directed learning research, the greater the new knowledge about the concept will be." (p. 47)

The implication here is that while learning projects research has been vital to our understanding of self-direction in adult learning, we have pretty much reached a point of saturation with this approach. The kinds of studies that will be reported in the next two chapters are indicative of a healthy evolution in the development of research on self-direction in adult learning.


In summary, we believe that the learning projects approach has made a landmark contribution to the adult education literature and has, both directly and indirectly, served as a major source of impetus for the development of a systematic body of knowledge in the area of self-directed adult learning. At the same time, we believe that it is important to interpret learning projects findings with a degree of caution and to recognize that the approach was designed to address the frequency and nature of learning projects activity, not to assess the quality of or reasons for such activities. These kinds of issues have begun to be addressed through other methodologies and will be explored in the next two chapters.


Go to the bibliography.

Return to the book's table of contents.

Return to the book's index.

Return to the first page.