«SPEAKING OF FAITH: PUBLIC RELATIONS PRACTICE AMONG RELIGION COMMUNICATORS IN THE UNITED STATES Committee: Dominic L. Lasorsa, Co-Supervisor Ronald ...»
Alphas lower than 0.60 prompted questions about whether the responses were reliably measuring the same concept. Low inter-item correlations showed that RCC members did not agree that the four statements about the public information model (correlations for all items less than 0.30), the two-way asymmetrical model (correlations for two items less than 0.30) and the press agentry/publicity model (correlations for two items less than 0.40) were describing a specific way to practice public relations. Only the four statements about the two-way symmetrical model had acceptable (.40) inter-item correlations (Leech, Barrett & Morgan, 2008). Those results indicated that RCC members saw those statements going together. Responses from faith group leaders showed more pronounced disconnects among statements describing each model. For leaders, only the four statements about the press agentry/publicity model represented a unified description of public relations practice (all inter-item correlation greater than 0.40). Inter-item correlations for statements about the other three models were generally less than 0.40.
Because of the erratic ways religion communicators and faith group leaders connected the 16 statements, an exploratory principal axis factor analysis was run with varimax rotation. This factor analysis looked for new, unobserved statistical relationships among responses from RCC members and their leaders. The goal was to see how religion communicators or faith group leaders might group the 16 items describing the four models differently from the original factoring reported by Grunig and Grunig (1992).
Such groupings might yield more reliable scales for the faith-community sample.
Furthermore, factors might help explain whether religion communicators and leaders understood public relations functions differently from their secular counterparts.
Data from communicators generally satisfied assumptions about independent sampling, normality, linear relationships between variable pairs and variables being correlated at moderate rates. But for leaders the Kaiser-Meyer-Olkin measure of sampling adequacy showed that data did not have enough items for factors.
Table 4 shows factor loadings for RCC members. The analysis identified four factors. These factors remixed 12 of the 16 measures and dropped four, three from the public information model and one from the press agentry/publicity model. The omitted
+ The success of a communication program can be determined from the number of people who attend an event or who use products or services (a measure of the press agentry/publicity model).
+ In communication, nearly everyone is so busy writing news stories or producing publications that there is no time to do research (a measure of the public information model).
+ Communication is more a neutral disseminator of information than an advocate for the organization or a mediator between management and publics (a measure of the public information model).
+ Keeping a clipping file is about the only way to determine the success of communication (a measure of the public information model).
After rotation, the four factors accounted for 37.04% of variance (13.43 in the first, 8.59 in the second, 7.85 in the third and 7.18 in the fourth).
Factor 1 showed a strong emphasis on advance research as a basis for planning communication efforts and follow-up research as a basis for evaluating results— particularly attitude changes. This factor focused on how an organization could learn about publics so it could successfully influence them. This research factor combined
three two-way asymmetrical measures with one two-way symmetrical measure:
+ Before beginning a communication program, one should do research to determine public attitudes toward the organization and how they might be changed (a two-way asymmetrical measure).
+ Before starting a communication program, survey or informal research should be done to find out how much management and our publics understand each other (a two-way symmetrical measure).
+ Before starting a communication program, one should look at attitude surveys to make sure the organization and its policies are described in ways our publics would be most likely to accept (a two-way asymmetrical measure).
+ After completing a communication program, research should be done to determine how effective it has been in changing people’s attitudes (a two-way asymmetrical measure).
Organizations used communication to seek publicity. Publicity was the purpose of public relations. The goal of public relations was to get publics to behave the way the organization wants.
Factor 3 emphasized mutual influence and included key dynamics of Church-Sect Theory. The focus was on using communication to build mutual understanding, mediate conflicts between organizations and publics, and let publics influence the organization as much as it influences them. Factor 3 combined three measures from the two-way symmetrical model.
Factor 4 emphasized using media relations to get positive news coverage. The factor combined a measure from the public information model with one from the press agentry/publicity model. The focus was on getting accurate, positive information about the organization into the news and avoiding bad publicity.
The four factors did give additional insights into how RCC members understood public relations functions. Religion communicators did not link statements about public relations practices the way Excellence scholars did. That indicated RCC members did not always agree with the way each model described its approach to public relations. RCC members saw little need for tasks associated with the public information model. Religion communicators appeared to consider research (a key component of both two-way models) and publicity more tools of persuasion than relationship building. Nevertheless, RCC members did recognize three mutual-influence statements from the two-way symmetrical model as one approach to public relations.
But the four factors did not solve the scale problem. Only Factors 1 and 4 had truly acceptable reliability coefficients (Cronbach’s alpha=0.75 for Factor 1, 0.71 for Factor 4). But with only two elements, Factor 4 would not provide a useable scale (Jaccard & Wan, 1996). Coefficients were 0.59 for both Factors 2 and 3—just below the
0.60 acceptable level (Leech, Barrett & Morgan, 2008). Deleting items from either scale did not improve the alphas for Factor 2 or 3. Inter-item correlations ran at or just below the generally acceptable 0.40 level (Leech, Barrett & Morgan, 2008). As a unit, therefore, these four factors did not provide any better basis for further data analysis of RQ 3 than the original models.
RQ4 asked how well religion communicators could predict what their supervisors thought about public relations practices in the four models of public relations. Columns 2 and 3 of Table 3 show that RCC members were off the mark in most cases.
Communicators generally underestimated leader support for the two-way symmetrical and two-way asymmetrical models and overestimated agreement with the press agentry/publicity and public information models. Predictions were significantly different from reality for the public information model (t = +4.45, df = 6, p.01). The correlation of the 16 predictions by communicators to the 16 statements by leaders—while not significant at the p.05 level—was negative (r=-.73). Correlation of predictions and what leaders said for the press agentry/publicity model were also negative (r=-.39, p=ns).
Communicators thought faith group leaders would consider attendance at events the primary indicator of communication success, publicity the purpose of public relations and getting good news into the press the primary objective. Top executives generally disagreed with all those ideas.
RQ5 asked how the expertise of RCC members to practice each public relations model compared to secular practitioners. Table 5 shows no significant expertise differences in mean responses at the p.05 level between religion and secular communicators. In fact, reports of expertise needed to practice the two-way symmetrical (r=.95) and two-way asymmetrical (r=.97) models correlated solidly at the p.05 level.
Expertise to practice the public information model—which had the highest composite means for both groups—approached significance (r=.93, p=.07).
RCC members and secular practitioners both claimed their departments had the highest expertise to perform the journalistic tasks related to the public information model.
Both groups saw their departments having average capabilities for publicity work. Both groups said their organizations were less prepared than an average department to handle the research-based interactive tasks associated with both two-way models. Raw mean numbers indicated that religion communicators rated their departments better prepared to handle most public information, two-way symmetrical and two-way asymmetrical tasks than secular communicators judged their organizations to be.
Cronbach’s alpha reliability coefficients for all scales in the RCC sample closely mirrored statistics reported by Grunig, Grunig and Dozier (2002). Alphas were 0.76 for the RCC study and 0.75 for the secular results for the public information model, 0.84 for
RQ6 asked how levels of communication managerial and technical expertise in departments where RCC members worked compared to those of their secular counterparts. Table 6 shows no significant difference between the groups for either scale at the p.05 level. Response from both groups to all 16 statements (r=.82, p.001) and the eight manager items (r=.97, p.001) correlated solidly. Correlation of the eight technician responses (r=.65) was not significant at the p.05 level. Nevertheless, raw mean numbers showed that religion communicators claimed their departments had average manager expertise and above-average technical expertise. Secular communicators claimed above average departmental expertise for both areas. The two groups ranked their departmental manager skills in the same order. Religion communicators rated their departmental expertise for taking pictures and writing ads higher than secular practitioners did.
Cronbach’s alpha reliability coefficients for the RCC sample generally corresponded to statistics reported by Grunig, Grunig and Dozier (2002). Alphas for the manager scale were 0.83 for both groups. Alphas for the technician scale were 0.78 for the RCC study and
0.80 for the secular results.
relations. Grunig, Grunig and Dozier (2002) considered formal education, professional association memberships and familiarity with trade publications all indicators of public relations knowledge. Four cross tabulations were used to check these relationships of knowledge to practice. Chi-square tests of that analysis showed no consistently significant relationships at the p.05 level between measures of knowledge and preferences for the four public relations models. Of the 288 cross tabulations done to investigate RQ 7, only 14 were statistically significant at the p.05 level. Highest level of
education was significantly associated with:
+ The two-way symmetrical public relations measure saying that the purpose of communication is to change management attitudes and behaviors as much
education beyond the bachelor’s degree (94% of the 108 responses) were less likely to agree with that statement than others. Communicators with
statement (91% for those with some graduate training disagreed; 80% for those with master’s degrees, and 71% for those with doctorates disagreed)
Formal training in communication, journalism, advertising, public relations or
marketing was significantly associated with:
+ The press agentry/publicity measure saying public relations essentially means publicity. RCC members with little or no formal education in communication were most likely to agree with this statement (45% of the 110 responses). Fifty-eight percent of those with no formal training, 20% of those with some continuing education or 50% of those with undergraduate coursework but not a major in a communication discipline agreed. More than a quarter of RCC members with graduate training
+ The public information measure of expertise saying that the practitioner’s office understood the news values of journalists. RCC members with at least a bachelor’s degree in a communication discipline (65% of the 110 responses) were more likely to rate their expertise for this statement as better than average (35% for those with bachelor’s degrees in communication, 31% for those with some graduate credit, 39% for those with a master’s degree) or tops in the field (57% for those with bachelor’s degrees, 39% for those with some graduate credit, 39% for those with master’s degrees) (p.05).
+ The two-way asymmetrical measure of expertise saying the practitioner’s office could get publics to behave the way the organization wanted. RCC members with graduate training in a communication discipline (32% of the 114 responses) were more likely to rate their expertise for this statement as better than average (54% for those with some graduate training, 40% for those with a master’s degree) (p.01).
+ The two-way asymmetrical measure of expertise saying the practitioner’s office
attitude theories (33% for those with some graduate training, 46% with master’s degrees and all with doctorates saying “none”) (p.05).
+ The two-way symmetrical measure of expertise saying the practitioner’s office could use conflict-resolutions theories. RCC members with at least a
bachelor’s degrees, 39% with some graduate education and 43% with master’s degrees) as low expertise (45% with bachelor’s degrees, 33% with some graduate education and 46% with master’s degrees) (p.05).
+ The two-way symmetrical measure of expertise saying the practitioner’s office could help management understand public opinion. RCC members with at