Beyond Traditional County Programs

Contemporary concerns in personal and family health, public issues, and new federal priorities have led family living agents to question the extent to which traditional assumptions about program topics and audiences are valid indicators of future interest in, and need for, Extension programs.1 Data directly from potential clientele can help avoid serving only the needs of select groups with traditional programs delivered by accepted methods. To obtain such data, family living agents in a multicounty area in Pennsylvania conducted a needs assessment and surveyed the population at large through a random phone survey.

For the needs assessment, the agents included 20 program topics they were currently teaching or had interest in developing. Some were traditional topics that had been taught for many years. Others included contemporary topics reflecting recent clientele interest and federal priorities.

Interest in future Extension programs was measured by the need a respondent expressed for information in each program topic.2 Included in the survey were questions on age, gender, and employment status as these variables could influence the extent of a respondent's interest in a particular program topic. Respondents who said they needed information on a topic were asked how they'd prefer to get this information.

To gain the widest representation of opinion from the residents in four counties,3 a group of 26 Extension volunteers conducted the telephone survey in 1986. They used random digit dialing4 and the Hagen Collier method to screen adults within households.5

The study targeted residents in a four-county rural area in central Pennsylvania with a population of 184,000. Most communities have a population of about 1,500. There are a few small manufacturing firms that employ several hundred individuals. The farms in the surrounding area are dairy or part-time operations. Two larger communities have populations around 12,000.

Volunteer interviewers participated in a day-long training session to build confidence and ease in using the questionnaire. The training focused on the use of consistent procedures and back-up techniques for problem interviews.6 A follow-up telephone call by a different volunteer to five percent of the sample validated that the interviews had taken place.

Seventy-four percent of the people contacted by phone agreed to be interviewed, yielding a sample of 540 residents. Seventy-eight percent of the sample were non-Extension users. In theory, in 95 out of 100 cases the results based on such a sample in the four-county area will differ by no more than four percentage points in either direction from what could have been obtained by interviewing all telephone users.7 Comparison of the sample data with the census suggests no reason to conclude that the sample is anything but representative of the population in the four-county area.8

Assumptions About Gender

The family living agents traditionally planned programs with the assumption that their major audience was female. The results of this survey undermined this assumption. In 16 out of 20 program topics, no statistical difference existed in the amount of interest that was shown between males and females (see Figure 1).

Assumptions About Age

Based on the success of their previous experiences, the family living agents assumed that the audience wanting more information would be middle-age and older women. This assumption wasn't supported by the survey data. In 15 out of 20 topics, young women (20-34 years old) were significantly more interested than middle-age women (35-54 years old), and in all 20 topics, older women (55 years and over) expressed the least interest. The findings on 10 program topics in Figure 2 show the trend reflected in all programs.9

Men of different age groups appeared more homogeneous than women. In general (in 15 out of 20 program topics), younger men were just as likely to express a need for information as middle-age and older men.

Assumptions About Employment Status

In planning programs in the past, the family living agents assumed that more nonworking women would express a need for information in family living topics than employed women. The survey found this wasn't true. Full-time and part-time employed women expressed a significantly greater need for information in 17 out of 20 topics. Women who worked part-time showed as much and sometimes an even greater interest than employed women in almost all programs. The findings in 10 program topics in Figure 3 show the trend in all programs.

The agents assumed that working men weren't typically interested in family living programs because of the many other demands on their time. The survey found, however, that employed men have a significantly greater interest in 16 out of 20 topics than nonworking men.10

The findings above suggest that there were many more people interested in family living programs than the agents imagined before the survey. Some groups stood out and warrant attention from Extension. Men, employed women (especially part-time women), and younger people (age 20-34 years) should be potential target audiences for future programming.

Findings About Content

Although the above results suggest that the potential need for programming is great, limited resources required the four participating home economists to set priorities among programs. To facilitate the selection of programs for development and implementation, the family living agents established a criterion for selection - if there was a strong interest, that is, if 50% or more of any target audience in this study (age, gender, or employment group) expressed a need for information on a program topic, that topic would be selected for delivery.

In the area of food, almost all target audiences regardless of age, gender, and employment status showed a strong interest in four out of the five food topics (Food and Disease, Food and Drugs, Food Safety, More for Food Dollar). It's significant that strong interest in most food topics wasn't restricted to what many home economists consider their traditional audience - nonworking women either middle-age or older.

Eight contemporary topics - Health for Public at Large, Water Quality, Managing Stress, Communication Skills, Role Changes, Personal Development, Balancing Home/Work, Retirement Planning - exhibited strong interest from selected target audiences. Women, young and/or middle-age and employed, showed a strong interest in all the topics. Employed men, and often young men, showed a strong interest in the first four topics.

In some program topics, strong interest was restricted to young audiences. Young women were the target audience for five traditional topics - Budgeting, More for Clothing Dollar, Food Preservation, Parenting Skills, Housing Costs/Concerns. Young men were the target audience for two topics - Housing Costs/Concerns and Small Business Development.

For the remaining two traditional topics of Clothing Construction and Wardrobe Selection/Care, none of the target audiences showed strong interest. This finding was a surprise to the family living agents who expected a strong interest from some groups of women.

The results of the survey demonstrate that many more family living topics were of interest to the public than the family agents assumed before the survey. In a few topics, where family living agents felt assured there would be strong interest, none was expressed. Although strong interest came from all segments of the population in four food topics, the expressed need for most topics came from selected segments of population.

Traditional Extension program delivery has relied on meetings and face-to-face teaching. Survey respondents expressed a strong preference for newsletters as the preferred method of delivery. In all program areas, interest surged well over 50% of the respondents. The next most preferred methods were meetings (13%) and learn-at-home programs (9%).

The findings of the survey have serious implications for future decisions about family living programming. The population of the four rural counties expressed a need for information in a broad range of family living topics, both traditional and contemporary. Only a few topics were of interest to the traditional family living audience. Some topics were of interest to the widest spectrum of the population, most topics were of interest to selected groups, and a few were of little interest to anyone. Therefore, to respond to the strong, expressed needs of the people, family living agents in these four rural counties need to branch out with a wider array of programs, but ones that are tailored to specific audiences. Some programs need to be dropped due to the lack of interest.

The survey evidence questioned assumptions family living agents had about appropriate program content, audiences, and delivery methods for a rural area in the late 1980s. The challenge facing family living agents is to break out of a traditional focus and move forward to develop programs and methods to meet the expressed needs of a broader base of people in their counties. By systematically surveying the citizens of a county, family living agents can develop valid indicators of future interest in programs, meet the real educational needs of the people, and greatly affect the future success of Extension.

1. Orville G. Bentley and James H. Anderson, FY 1987 Priorities for Research, Extension and Higher Education (Washington, D.C.: Joint Council on Food and Agriculture Sciences 1985), pp. 1-5.

2. Respondents were asked, "For each topic I read, tell me if you need information on it, do not need information, or if you are not sure."

3. The 1980 U.S. Census indicated that the number of households without telephones in these counties is low, between 3% and 5.8%.

4. James H. Frey, Survey Research by Telephone (Beverly Hills, California: Sage Publications, 1983).

5. Dan E. Hagen and Charlotte Meir Collier, "Must Respondent Selection Procedures for Telephone Surveys Be Invasive?" Public Opinion Quarterly, XLVII (No. 4, 1983), 547-56.

6. Within the survey, there was little missing data. Response to each question was nearly complete: 99% or better.

7. The error for sampling smaller groups like men, women, or different age groups is larger; for men or for women + 6%. In addition, changing the wording or context of some questions or the practical difficulties of conducting the interview could have introduced other sources of bias.

8. Comparison of the sample with the 1980 U.S. Census reveals no statistical differences on age distribution for either men or women. This is important since recent research has shown that older people, especially the very old, are reluctant to participate in telephone surveys. See, A. Reula Herzog, William L. Rodgers, and Richard A. Kulka, "Interviewing Older Adults: A Comparison of Telephone and Face to Face Modalities," Public Opinion Quarterly, XLVII (No. 4, 1983), 405-18.

9. Findings on the remaining program topics in this figure and those following may be obtained from the authors.

10. Ibid.