Running Head: RESEARCH STUDY ON PERSUASIVE EFFECTS OF COMMUNICATION

Research Study on Persuasive Effects of Communication

University of Phoenix

Research Study on Persuasive Effects of Communication

Respondents completed public opinion surveys regarding preferred options for local government, conducted by the Local Government Commission for England and Wales. The surveys included a briefing to ensure that the respondents were adequately informed. A detailed analysis of the first results from Durham and Cleveland counties indicated that the social characteristics of the respondents who changed their preferences after the briefing were only marginally atypical compared with those of the other respondents, but that district of residence was an unexpectedly important factor. A change in preferred option did not seem to be related in any simple manner to the information that was provided in the briefing. The findings from this field research are discussed in the context of experimental studies of persuasive communication, with an emphasis on communication coherence and complexity.

Research Purpose

SOCIAL PSYCHOLOGISTS HAVE MADE a significant contribution to researchers' knowledge about the effects of persuasive communications. Numerous carefully controlled and well-reported experimental studies have pinpointed the effects of communicator variables, respondent variables, communication content, and setting variables on attitudes (summarized in Hovland, Janis, & Kelly, 1953). A powerful body of laboratory-based research has established the effects of fear-arousing appeals versus non-fear-arousing appeals, primacy versus recency effects, one-sided versus two-sided communications, and conclusion-drawing versus non-conclusion-drawing communications; indeed, experimental variables that might affect attitudes have been thoroughly researched (Allen, 1991; Chaiken, 1979; Chaiken & Eagly, 1976; Eagly & Chaiken, 1975; Eagly & Himmelfarb, 1978; Eagly & Warren. 1976).
The limitations of these early studies offset any real advances in research on persuasion, however (Eagly, 1992), and resulted in a burgeoning but fragmented research literature. This situation remained unchanged until the advent of cognitive response theory (Petty & Cacioppo, 1981, 1984, 1986) and heuristic approaches to persuasion (Chaiken, 1980; Eagly & Chaiken, 1984). Subsequently, researchers have placed more emphasis on the cognitive processes that are used in the active consideration of communication content, personal strategies used to assess the significance of a communication by reference to variables established in earlier experimental research and to idiosyncratic rules of thumb, metaphors, analogies, and other characteristic heuristic devices (Chaiken & Maheswaran, 1994). To some extent, these developments reversed the earlier tendency of experimental studies on persuasive communication toward accumulative fragmentalism (Kelly, 1955).
The case study investigated in the present research was an instance in which something akin to a loosely controlled large-scale experiment was conducted on behalf of a government agency. The methodology permitted a measure of control through careful sampling and the standardization of the data collection procedure. The kinds of data that were being Collected allowed the effects of an important communication to be monitored unusually closely, although this was by no means the main purpose of the present research.
Beginning in the summer of 1993, the Local Government Commission for England and Wales (LGC) conducted a series of public opinion surveys about proposed options for the reorganization of local government across the nation. The proposal to introduce an element of briefing into these surveys, so that the public would be adequately informed about the details of each of the options, was a source of considerable interest and controversy, because it was felt that the form and content of this briefing would radically affect residents' responses. Indeed, strong protests about the exact nature of the proposed briefing were made to the LGC by the Association of District Councils and the Association of County Councils.
The LGC planned to ask representative samples of residents in each area across the nation to indicate their preferred choice from a number of options regarding the reform of local government. After residents had indicated their preferred options, they would be briefed about the details and the cost of each option and subsequently asked to indicate their preferred options again. It was expected that preferences might change in view of the new information.

Data Collection Methods

Field Research for Durham and Cleveland Counties
The sampling procedure for the LGC public opinion surveys was identical for the two county areas. In both areas, quota samples were secured so as to represent the demographic characteristics of each area, as determined by the 1991 national census. Following are the results for Durham County, with the equivalent figures for Cleveland County immediately afterward in parentheses. In Durham, a sample of 2,478 (1,235) was drawn from a total of 218 (105) sampling points: 49% (48%) were men and 51% (52%) were women. Age quotas were as follows: 18-24 years = 11% 13%); 25-34 years = 20% (21%); 35-44 years = 18% (20%); 45-54 years = 16% (14%); 55-64 years = 4% (14%); 65-74 years = 13% (13%); and 75+ years = 7% (6%). I used the employment status of the head of the household, asked during the interview, to establish the social composition of the sample: AB (professional/managerial) = 10% (10%); C1 (skilled manual/white-collar) = 21% (20%); C2 (semiskilled) = 24% (26%), D (unskilled) = 20% (20%); E (other/unemployed) = 24% (24%). After establishing contact with a respondent who fit the quota requirements, the interviewers met with the respondent at his or her home, as described previously, and discussed the information that is provided in Appendixes A and B. The two surveys were conducted concurrently during July and August of 1993.

Research Results

The LGC reported almost no change in the preferred options of the residents of Cleveland and Durham counties; the highest proportion of net change after the briefing in any district in the two counties was reported to be 1%. However, a case-by-case examination in a secondary analysis indicated that the figures for net overall change masked the extent of individual change. The results of this more detailed analysis, reported in this article, indicated that the 228 residents of Durham County who changed their original preferences actually constituted 9.2% of the overall sample of 2,478. One hundred ninety of the 1,235 respondents in Cleveland county, or 15.4 % of the overall sample, changed their preferences after briefing.
Durham
The LGC options for Durham, listed in Appendix A, varied considerably, from the status quo of a county council and eight local district councils (Option A), to other possibilities combining the existing local districts in different ways (Options B, C, and E), to the possibility of one area authority (Option D).
After being briefed about the probable effects of these options, the respondents were asked to indicate which options they preferred, taking into consideration the information that had been presented in the briefing. The interviewers had been trained to use maps to clarify the options, and the respondents reviewed a leaflet about the relative costs of the options.
A relatively large number (59) of the 228 Durham changers altered their original stated preference in favor of the existent system of local government to another preference, but none of the new preferences was clearly favored. Almost half these changers shifted to "don't know" as a result of the new information that was presented in the briefing. More than half (63%) of the respondents who originally preferred Option B (eight authorities) and about 40% of the respondents who originally preferred the other remaining options changed their preferences to the existent system of local government. Thirty-two respondents changed their preferences to Option D (one authority), 27 respondents changed their preferences to Option C (two authorities), 22 respondents changed their preferences to Option E (four authorities), and 11 respondents changed their preferences to Option B (eight authorities).
The net effect of the briefing on the Durham sample was to increase professed support for the existent two-tier system and to reduce support for the eight-authority option. The eight-authority option proposed a separate authority for each of the following areas in the district -- Chester-le Street, Darlington, Derwentside, Durham City, Easington, Sedgefield, Teesdale, and Wear Valley -- and, according to the briefing, was the option that would involve the least expensive start-up costs. The net effect of the briefing on the other options was almost neutral, with only small changes occurring.
Cleveland
The briefing for the 1,235 Cleveland residents contained the same type of information the briefing for the Durham sample did but was based on the estimated costs for six options instead of five (see Appendix B). Two of the six options differed only in respect to a relatively small boundary change. The range of estimated start-up costs and ongoing savings for the Cleveland options was much smaller than the comparable range for the Durham options, so no one option was clearly financially advantageous. The start-up costs were relatively modest in comparison with those in the Durham study.
One hundred ninety of the 1,235 respondents in the Cleveland sample changed their original preferences for local government reorganization after the briefing. The existent two-tier system (Option A) lost the support of 18 respondents but gained the support of 69 respondents, a net gain of 51. The first of the four authority-solutions (Option B, net gain = 5) and the one-authority solution (Option F, net gain = 10) also demonstrated small gains. Net losses were evident for the second of the four authority-solutions (Option C, net loss = 7), the two-authorities solution (Option D, net loss = 11), and the three-authorities solution (Option E, net loss = 7).
Seven residents changed their preferences from the one-authority unitary county solution (Option F) to the existent two-tier system (Option A), despite the fact that the briefing portrayed the one-authority solution as the best option financially. Given the financial disadvantages of Option A relative to the other options, in conjunction with the stated equivalence of the level of services, the shift in the preferences of 69 residents (51 net) to the existent two-tier system was unexpected.

Conclusion

References