Page 227

OLDER ADULT LEARNING: INSTRUMENTAL AND EXPRESSIVE CATEGORIES

Roger Hiemstra

Department of Adult and Continuing Education

University of Nebraska-Lincoln

Educational Gerontology, 1: 227-236, 1976

Copyright © by Hemisphere Publishing Corporation, A Taylor and Francis Journal.

Reprinted by Permission of the Journal’s editor

Stylistically, the article has been converted to APA, 5^th Edition.

This article examines older adult preferences for instrumental vs expressive learning activities. Course title selections and actual learning activity information were analyzed. Interviews with 256 Nebraskans, 55 or older, (average age was 68.11) were obtained. Hypotheses and results were (a) predicted preference for instrumental learning was supported; (b) greater preference for instrumental learning by blue-collar workers and the less educated received partial support as no differences existed for the occupational category but individuals without college degrees preferred instrumental courses; and (c) predicted differences in learning activity received partial support as younger people, white-collar workers, college graduates, nonwhites, and married people were more involved with instrumental learning. It was concluded that more instrumental learning opportunities must be made available to older people.

INTRODUCTION

Much has been written about the older adult and learning. Early writings focused on the premise that learning needs and capabilities decline with age. However, recent research and discussion have centered on a changing theme: declines in learning abilities and interests are less than has been historically thought. Havighurst (1972) pointed out that learning is necessary throughout life because of the continuing needs of new developmental tasks.

Thus, a variety of stereotypes and myths about the elderly are being challenged and dispelled. McClusky (1974) suggested that the elderly, in general, are active, intelligent, and involved people who have positive feelings about themselves and their potential. The purpose of the research reported here was to secure a better understanding of older adults' learning needs and potential.

The following questions served as study guides: (a) What are perceived preferences relative to instrumental and expressive forms of education? (b) What instrumental and expressive form of learning are older adults actually engaging in? (c) What are the

Page 228

relationships among various demographic characteristics and the perceived preferences or actual learning endeavors?

Learning needs and preferences among older people have been described in various ways. McClusky (1974) suggested five types of needs: coping, expression, contribution, influence, and transcendence. Examples of corresponding adult education programs might include (a) adult basic education for coping needs, (b) hobby courses for expressive needs, (c) leadership training for contributive needs, (d) community action education for influence needs, and (e) the study of literature or philosophy for transcendence needs.

Havighurst (1964) and Londoner (1971) described educational programs and course offerings in terms of instrumental and expressive categories of learning. Instrumental is defined as learning activities designed for effective mastery of old-age challenges and includes education on such topics as health, income, legal affairs, and adjusting to changing relationships with others. Expressive is defined as educational experiences that increase the enjoyment of life, serve to expand horizons, provide fairly immediate gratification, or facilitate opportunities for self-expression. Examples of expressive learning activities include hobby and craft instruction, travel experiences, music or art appreciation, and literature study.

Hiemstra (1972) investigated the educational needs and interests of older people. Instrumental activities were perceived by older people as more important than expressive activities. Subsequent researchers (Goodrow, 1974; Marcus, 1975; Whately, 1974) have found general support for the instrumental preference.

Dichotomizing educational opportunity into either/or categories has some drawbacks as noted by DeCrow (1974, p. 59). However, the fairly broad categories enable program planners and administrators to distinguish quite readily among learning activities according to the instrumental and expressive categories. Consequently, for purposes of this research it is hypothesized that older persons show preference for instrumental learning activities more frequently than for expressive activities. Preference is examined both by course title selection and actual learning activity.

Hiemstra (1973) also examined instrumental vs. expressive preferences in terms of various biographical and demographic characteristics. There were no significant differences in terms of age, sex, and urban vs. rural categories. However, significance testing did reveal that white-collar workers were less likely than blue-collar workers and college graduates were less likely than non-college graduates to report instrumental course preferences.

Page 229

Thus, in this study it is hypothesized that preferences for instrumental learning are in directions identical to those described. Significant relationships relative to additional demographic characteristics are reported to indicate direction for subsequent research.

Tough (1971) developed a means for determining the actual amount of learning activity undertaken by an adult in a year. Although his research focused on adult learners of all ages, subsequent reporting (Coolican, 1974) showed that the approach is applicable to various age groups and types of individuals. However, there have been no reports of comparisons between the amount of learning and preferences for instrumental or expressive forms of learning. Thus, directional predictions in terms of demographic characteristics are difficult to make. Utilizing Tough's methodology for discovering actual learning activity, it is hypothesized that differences in the amount of learning exist when instrumental and expressive categories are compared with several demographic variables. Any differences shown will facilitate directional predictions for subsequent research.

METHOD

Subjects

Data collection involved interviewing 256 people, 55 and older, in a field setting. All subjects lived in Nebraska and were selected randomly from voter registration cards in two communities and 18 rural townships, from the rolls of two residential complexes built especially for the elderly, and from the rolls of a Mexican-American community center. There are built-in biases to voter registration cards. In addition, participants in programs for special groups or residents of homes or apartments built especially for the older person will be somewhat self-selecting. Consequently, one study limitation is that a totally random selection of subjects probably was not accomplished. A chi-square comparison of the study's demographic data with U.S. Census data for Nebraska did reveal that there were no significant differences in comparisons on sex, marital status, and occupation. However, the study population was somewhat older, more urban, more educated and had more nonwhites than a truly random sample of people 55 and older in Nebraska.

Of the interviewees, 41% were male, 89% were white American, 88% were middle class, 63% were married and 25% were widowed, 32% were high school graduates or less, 44% were or had been white-collar workers, and 40% were between the ages of 55 and 64. The average age was 68.11 years.

Page 230

Procedure

Trained interviewers asked the participants to indicate in which of 32 courses (course titles with no interpretation were given) they would enroll given the opportunity and if there were no preventative constraints. The course titles were chosen from an initial pool of 75 titles gleaned through literature searching and by securing course catalogs from five institutions offering educational programs to the elderly. The subjects were told nothing relative to the instrumental or expressive categories. The interviewers also used an extensive probing technique to uncover the subjects' amount of annual learning in terms of number of hours and number of different projects (Tough, 1971, p. 17). (A learning project is defined as at least seven hours of related learning within a six-month period.)

Construct validation involved the use of a panel of judges. Each panel member (a teacher in gerontology, an administrator of gerontology programs, and an adult education-Cooperative Extension researcher), working independently of the other two, was given the initial list of 75 courses and definitions of the terms instrumental and expressive. A course title was included on the interview schedule if all three judges agreed as to whether it were instrumental or expressive.

Concurrent validity was assessed by obtaining the information from interviewees as to their participation in learning activities during the year prior to the interview. The various activities were later classified as either expressive or instrumental in nature by the researcher and one independent judge. A correlation of the number of course preferences in a category to the number of corresponding actual learning projects for all individuals revealed that:

r _instrumental = .254

r_expressive = .347

Both correlation coefficients are significant at the .001 level and beyond.

A telephone follow up of one person chosen randomly from each interviewer's group of subjects was carried out approximately one month after each interview to check the reliability of both interviewer and instrument. There were no observed differences or discrepancies. In addition, the total sample was split randomly into two groups. The groups were then compared by chi-square on the

Page 231

total number of expressive and the total number of instrumental course selections. No significant differences were found:

	Instrumental	Expressive
Group 1	629	402
Group 2	615	439
Totals	1244	841	χ^2 = 1.42

Data for the first two hypotheses were analyzed with the chi-square test; the .05 level of confidence was utilized to determine significant differences. Whenever a direction was predicted, the one-tailed test of significance was utilized (Siegel, 1956, p. 13; Tuckman, 1972, p. 378). The third hypothesis was examined by the t-test for significance in the differences between means of two groups. The number of learning projects was assumed to represent an interval scale so that the t-test could be used (Kerlinger, 1967, p. 427).

RESULTS

Table 1 includes the preference information and chi-square values pertaining to the first hypothesis. A significant preference for instrumental learning was found both in course selections and in actual learning projects. Approximately 60% of all course selections and learning projects were instrumental in nature, thus supporting the hypothesis.

The second hypothesis predicted directions in the comparisons of instrumental or expressive preferences with various demographic characteristics. There is some support for the hypothesis (see Table 2)

TABLE 1. Instrumental and Expressive Learning Preferences

Preference category	Actual no.	Expected no.	χ^2value^a
Course title selection
Instrumental	1244	1042.5	77.89^*
Expressive	0841	1042.5
Actual learning projects
Instrumental	421	346	32.51^*
Expressive	271	346

^aAssuming a null hypothesis of no difference, 50% of the total number of course selections or learning projects could be expected in both categories.

^*p < .001.

Page 232

TABLE 2. Comparison of Various Demographic Characteristics with Instrumental or Expressive Learning Preferences

Characteristic	Instrumental preference^a	Expressive preference^a	χ^2
Age
55-64	067 (054)^b	024 (020)	.73
65 and older	091 (062)	044 (041)	(2.57)
Gender
Female	083 (059)	048 (041)	5.64^**
Male	075 (057)	020 (020)	(3.71)^*
Location^c
Urban	082 (060)	049 (041)	7.12^***
Rural	076 (056)	019 (020)	(3.31)^*
Occupation^d
Blue collar	085 (053)	036 (027)	.00
White collar	072 (063)	032 (033)	(.01)
Education
College graduate	028 (023)	018 (017)	1.68
Less than college graduate	129 (093)	050 (043)	(1.18)
Race
White American	133 (100)	066 (053)	6.32^**
Other	025 (016)	002 (008)	(.01)
Marital Status^e
Married	107 (088)	035 (031)	4.70^*
Not married	051 (028)	033 (030)	(10.27)^***

^aInstrumental or expressive preferences were determined by tabulating an individual’s total number of selections or projects in each category. If the subject’s total for instrumental was larger than the total for expressive, the label of instrumental preferences was given (vice versa for expressive preferences). Not included in the computations for this table were 30 individuals who had chosen an equal number of instrumental and expressive courses (79 individuals had an equal number of instrumental and expressive learning projects). Totals are not always equal between characteristics because of some nonresponses.

^bNumbers in parentheses represent the data on learning projects. See footnote a for an explanation of how the numbers were derived.

^cUrban residents were from Lincoln, Nebraska. Rural residents were from rural areas and small towns in Nebraska.

^dHomemakers were included in the “Blue collar” classification.

^eSubjects in the “Not married” classification were never married, widowed, divorced, or separated.

^*p < .05.

^**p < .01.

^***p < .005.

Page 233

because individuals without college degrees were more likely to pick instrumental courses than were college graduates. However, there was no significant difference in occupational classification. In addition, males and rural residents were significantly more likely to pick instrumental topics. Nonwhite and married people were significantly more likely to make instrumental choices. Although complete support for the hypothesis was not found, the new information should be useful for subsequent research.

When the information on learning projects is compared by chi-square with the demographic characteristics (see Table 2--the figures in parentheses), the results are almost identical to the course-selection data. Only the "race" characteristic differed, as no significant difference was found for the learning-projects comparison.

The third hypothesis suggested that differences exist in preferences for instrumental or expressive learning when actual learning activity and the demographic characteristics are compared. No directional predictions are made. Table 3 contains t-test information and shows that the hypothesis can be supported somewhat. Only the gender characteristic and the rural or urban status showed no significant differences in the testing for instrumental preferences. The two age groups, the occupational classification, the race categories, and the marital status characteristics showed no significant differences in the t-test on expressive preferences.

The findings do supply some directional information for subsequent research. Table 3 shows that younger subjects, white-collar workers, college graduates, nonwhites, and married people are more likely to be engaged in instrumental activities. Females, urban residents, and college graduates are more likely to be engaged in expressive forms of learning.

DISCUSSION

Participation by older adults in formal adult education courses and programs has been relatively limited according to census and other enrollment data. Older people have not participated because of a variety of obstacles; in addition, educational program planners and administrators often believe that older people are not interested in formal programs and, thus, do not encourage them to participate (Mason, 1974, pp. 71-72). The findings presented suggest that older individuals not only have a variety of educational preferences but also are engaging in a variety of learning projects each year. Therefore, if education is a basic right of all people of all ages

Page 234

TABLE 3. Comparison of Various Demographic Characteristics with the Number of Annual Instrumental or Expressive Learning Projects

Characteristic	Number of people	Instrumental mean	Instrumental St. dev.	t-test Values	Expressive mean	Expressive St. dev.	t-test Values
Age
55-64	101	2.03	1.76		1.03	1.21
65 and older	155	1.38	1.33	3.29^**	1.08	1.26	-.30
Gender
Female	105	1.58	1.45		1.23	1.30
Male	151	1.74	1.66	-.85	0.82	1.10	2.69^**
Location^a
Urban	145	1.64	1.58		1.21	1.28
Rural	111	1.65	1.49	-.04	0.86	1.15	2.35^*
Occupation^b
Blue collar	142	1.46	1.38		0.96	1.18
White collar	113	1.88	1.71	-2.17^*	1.16	1.29	-1.24
Education
College graduate	050	2.14	1.83		1.44	1.22
Less than college graduate	205	1.53	1.44	2.53^*	0.96	1.23	2.49^*
Race
White American	227	1.56