The Effect of Union Protest Behavior on Attitudes Toward Unions: An Experimental Analysis Mel E. Schnake Valdosta State University Michael P. Dumler Illinois State University K. Nathan Moates Valdosta State University Student attitudes toward unions were measured at the beginning of a semester. Approximately 90 days later, students were shown a video of union protestors shouting down teachers who had called a press conference to announce they had filed a lawsuit against the union. Following the video, student attitudes toward unions were measured again. The results suggest that union behavior perceived as negative does have a deleterious effect on observer attitudes toward unions. This effect occurs even for observers whose parents were union members. Individual attitudes toward unions have been shown to be a central factor in union organizing (Barling, Fullagar & Kelloway, 1992). Specifically, attitudes toward unions have been related to voting for union representation in certification elections (Brett, 1980; DeCotiis & LeLouarn, 1981; Getman, Goldberg & Herman, 1976; Montgomery, 1989), participation in union activities (McShane, 1986), and affective commitment to a labor union (Fullagar, McCoy & Schull, 1992). Individual beliefs and attitudes play a central role in the model of union voting decisions developed by Summers, Betton and DeCotiis (1986). Several studies have produced support for a sequence of prounion attitudes beginning with an intent to join, followed by joining or voting to join a union (Davy & Shipper, 1993; Deshpande, 1992; Haberfeld, 1995; Hemmasi & Graf, 1993; Premack & Hunter, 1988). Kelloway, Barling, Fullagar and Lalibert (1995) suggest that attitudes toward unions may serve as a perceptual filter through which work and union experiences are interpreted. While individual attitudes toward unions have been related to a number of outcome variables, recently researchers have begun examining the origins of these attitudes. Barling, Kelloway and Bremermann (1991) found that parental socialization and individual work beliefs are significant predictors of union attitudes. Specifically, they produced evidence that the effects of perceptions of parents’ union participation were mediated by perceptions of parents’ general union attitudes. Hester and Fuller (1999) found support for this mediation effect. Kelloway and Newton (1996) found that parents’ general union attitudes, parents’ union participation, and parents’ job satisfaction impact students’ attitudes toward unions. Chang (2003) found that race and class affect union attitudes directly and indirectly through negative economic events. Both specific attitudes toward unions (i.e., union 90 American Journal of Management Vol. 16(2) 2016 instrumentality) and general attitudes toward unions have been established as predictors of willingness to join a union (Barling, et al., 1992). Houghton (2000) presents a “fundamental attitudinal model of union voting intentions,” presented in Figure 1, in which family socialization factors and prior work and union experience (including job dissatisfaction) impact general union attitudes. These attitudes are also affected by personality factors indirectly through work beliefs. “Miscellaneous contextual factors” (including the influence of campaigns for and against union representation, and financial and economic factors) impact perceived union instrumentality. Getman, Goldberg and Herman (1976) found that campaigns both for and against union representation do impact employee voting intentions. Houghton’s model shows miscellaneous contextual factors affecting willingness to join/vote for a union indirectly through perceived union instrumentality. Finally, general union attitudes also impact perceived union instrumentality. Perceived union instrumentality perceptions directly affect willingness to join/vote for a union. FIGURE 1 HOUGHTON’S FUNDAMENTAL ATTITUDINAL MODEL OF UNION VOTING BEHAVIOR A potential modification of this model, given previous research findings is a direct link between general union attitudes and willingness to join/vote for a union, in addition to the indirect link to willingness to join/vote for a union through perceived union instrumentality. That is, general union attitudes may affect willingness to join/vote for a union both directly and indirectly through perceived union instrumentality (Barling, et al., 1992). Another potential modification to this model and one which is directly related to the current study, is a direct effect between some contextual variables and general union attitudes. Some contextual variables, such as organizing campaigns, may affect employees’ perceptions of union instrumentality. During organizing campaigns, unions often provide information about what union representation can do for employees, while management provides information about the downside of union representation. This type of information is more directly related to perceived union instrumentality. However, other union behaviors, such as political activism, informational picketing, lobbying, and protesting anti-union press conferences, may affect observers’ general union attitudes. Personality Factors Work Beliefs General Union Attitudes Willingness to Join/Vote for A Union Family Socialization Factors Prior Work and Union Experience (e.g., Job Dissatisfaction) Miscellaneous Contextual Factors Perceived Union Instrumentality American Journal of Management Vol. 16(2) 2016 91 One potential influence on attitudes toward unions which has not yet been examined fully is the behavior of unions themselves. Union organizers and pro-union activists engage in a variety of collective (i.e., concerted) behaviors such as picketing, marches and demonstrations, and press conferences. The behavior of union activists may impact the attitudes of observers. Getman, et al. (1976) provide some evidence that organizing campaign behavior affects union voting intentions. The effects of other forms of union behavior on general attitudes toward unions have not been extensively examined. On September 22, 2005 a group of California teachers, supported by the National Right to Work Foundation, held a news conference in Sacramento, California to announce a “free speech” lawsuit. The teachers claimed that the California Teachers Association had begun withholding $60 a year from their checks without notifying them in advance. They further argued that this money was being used for political purposes for which they should not be required to pay. Three teachers opposed to the increase in dues, an official from the Right to Work Foundation and California State Senator Tom McClintock spoke during the press conference. As they spoke, a group of California Teachers Association (union) members attempted to shout down the speakers by loudly chanting “Shame on you.” Senator McClintock said “This is an example of the kind of intimidation, bullying and thuggery that our public school teachers are enduring from the union every day.” Thus, some observers clearly viewed the behavior of the union activists as inappropriate. A video of this press conference can be viewed (in two parts) at: http://www.youtube.com/watch?v=0Cglp-9sjp4 http://www.youtube.com/watch?v=qfrRCUO7AWw&NR=1 VICARIOUS LEARNING OF ATTITUDES Psychological research suggests that attitudes may be changed by changing information and beliefs, by changing behavior through reinforcement, or by social influence (e.g., influence of parents, peers, important others) (Zimbardo & Leippe, 1991). Thus, to the extent that a press conference is informational, it may lead to attitude change. Attitudes develop not only as a result of direct experience with stimuli, but also through the observation of others’ reactions to relevant stimuli (Bandura, 1965; Berger, 1962; Berger, 1980). Bandura (1977) notes that there is evidence that attitudes are often acquired through the observation of the emotional reactions of other persons exposed to aversive stimuli. According to behavioral theory, an attitude is a habit between a stimulus and an emotional response (Doop, 1947; Kanekar, 1975; Radtke, 1967; Staats, 1968). Thus, an attitudinal response is a conditioned emotional response acquired through classical conditioning. A model’s positive or negative emotional response to a stimulus leads to a similar emotion in an observer (Kanekar, 1976). Thus, observing an event, such as a press conference, where they key participants (the presenters) are exposed to a continuous aversive stimulus may classically condition observers of the event. Should the key participants of the event express displeasure, observers may develop a similar attitude toward the aversive stimulus. This effect is likely to be magnified to the extent that observers identify with the key participants (i.e., models) (Kanfer, Karoly & Newman, 1974). Identity is influenced by perceived model- observer similarity and/or models with whom observers empathize (e.g., small children, teachers, the disadvantaged). The purpose of this study is to examine the effects of observing union members’ protesting an anti- union press conference on observer union attitudes. METHOD A twelve item questionnaire containing a general union attitude scale (e.g., “Unions are an important, positive force in our society.” “Unions are necessary to balance the power and authority of management.”) and a specific union attitude (i.e., union instrumentality) scale (e.g., “Unions are 92 American Journal of Management Vol. 16(2) 2016 http://www.youtube.com/watch?v=0Cglp-9sjp4� http://www.youtube.com/watch?v=qfrRCUO7AWw&NR=1� instrumental in improving working conditions and pay for employees.” “Without unions, employees would have little job security.”) was presented to senior level business students at a state university in the Midwest. In addition to the union attitude scale, age, gender, nationality and both students’ and their parents’ experience/membership in unions were collected. A cover story suggested that the researchers were attempting to validate a new questionnaire. Student ID numbers were collected in order to match these questionnaires with the second questionnaire. 154 of 165 students completed usable questionnaires as a result of the first questionnaire administration (eleven questionnaires were omitted due to missing data). The average age of the respondents was 22.5 years and 53% were female. Only one respondent was currently a union member and less than 5% had ever been members of a union. Twenty-four percent reported that their parents had ever been members of a union. Toward the end of the semester (approximately 90 days after the first questionnaire administration), students in the same classes were briefly told of the California Teachers Association press conference. They were told that the union had increased teachers’ union dues without first notifying the teachers and that some of the teachers were filing a lawsuit to recover this money with the help of the National Right to Work Organization. They were also told that the reason the union raised dues was to support a campaign against two propositions which the union opposed that Governor Schwarzenegger had placed on the ballot. The students then viewed the video and responded to a second questionnaire containing the same union attitude scale. 121 students completed usable questionnaires as a result of the second questionnaire administration resulting in a total of 121 matched questionnaires. Students were then debriefed as to the true purpose of the study. RESULTS The twelve item union attitude scale was factor analyzed using both the pre-video data and the post- video data. Before analysis three negatively worded items (e.g., “Most unions are corrupt.”) were recoded. A principal components exploratory factory analysis of the pre-video data resulted in three interpretable factors with eigenvalues greater than 1.0, explaining 65.43% of the variance. Four items loaded cleanly on factor one and were interpreted to be a general union attitude dimension (e.g., Unions are an important, positive force in our society). Three items loaded on a second factor which was interpreted to be union instrumentality (e.g., “Without unions, employees would not have a voice with management” and “Without unions employees would have little job security”). Three items loaded on a third factor which was interpreted to be union image (e.g., “The country would be much better off without unions”, and “Most unions are corrupt”). This factor is similar to the “big labor” public image noted by Kochan (1979). The other two scale items exhibited split loadings on two or more factors and were dropped from further analysis. Using the same procedure on the union attitude scale administered after students were shown the video, the same three factors emerged, explaining 64.75% of the variance. Table 1 presents a correlation matrix and reliability estimates for the study variables. TABLE 1 CORRELATION MATRIX AND SCALE CHARACTERISTICS Pre-video Data 1 2 3 1. GenAtt .75 .51** .52** 2. Instr .79 .38** 3. Image .72 Coefficient alpha reliability estimates appear boldfaced on the diagonal. ** p> .01 n=156 American Journal of Management Vol. 16(2) 2016 93 Post-video Data 1 2 3 4. GenAtt .75 .59** .64** 5. Instr .80 .27** 6. Image .73 Coefficient alpha reliability estimates appear boldfaced on the diagonal. ** p> .01 n=121 Pre-video and post-video union attitude scale responses were compared (t-test) to determine whether viewing the union members’ behavior during the press conference had any impact on observers’ union attitudes. The results of this analysis appear in Table 2. TABLE 2 RESULTS OF T-TEST OF PRE-VIDEO AND POST-VIDEO UNION ATTITUDE SCALES PreGenAtt(3.46(/PostGenAtt(3.19) 4.33** PreInstr(2.72)/PostInstr(2.36) 5.26** PreImage(3.23)/PostImage(3.01) 3.75** ** p < .01 * p > .05 n = 112 (cases were excluded analysis by analysis). Means of each measure appear in parentheses. In each case, the post-video measure was significantly lower than the pre-video measure. Respondents’ general attitude toward unions and perceptions of union instrumentality were negatively affected by viewing the union members protesting an anti-union press conference. The Image measure contained items which were recoded (e.g., “Most unions are corrupt”, “Unions are too powerful”, and “The country would be better off without unions”) before analysis. Thus, the significant decline in the mean of this measure indicates that observers’ perceptions of unions became more negative after viewing the video. One important influence on union attitudes identified in previous research is parents’ participation/ membership in unions. To assess the impact of this influence, the pre and post-video union attitude measures were submitted to an analysis of variance comparing respondents whose parents had been union members to respondents who reported their parents had never been union members. The results of this analysis appear in Table 3. Measures of all three union attitude measures were significantly different at the .01 level before respondents viewed the press conference video. After viewing the video, only one of the union attitude measures was still statistically significant and the level of significance dropped to .05. 94 American Journal of Management Vol. 16(2) 2016 TABLE 3 ANALYSIS OF VARIANCE OF UNION ATTITUDES BY PARENTAL UNION MEMBERSHIP Df F Sig PreGenAtt 151 12.46 .001 PreInstr 150 5.87 .017 PreImage 151 14.83 .001 PostGenAtt 110 2.02 .16 PostInstr 110 .10 .75 PostImage 110 4.06 .05 Means PreGenAtt/Parents Members 3.78 PreGenAtt/Parents Not Members 3.35 PreInstr/Parents Members 3.05 PreInstr/Parents Not Members 2.67 PreImage/Parents Members 3.35 PreImage/Parents Not Members 3.14 PostGenAtt/Parents Members 3.36 PostGenAtt/Parents Not Members 3.14 PostInstr/Parents Members 2.41 PostInstr/Parents Not Members 2.35 PostImage/Parents Members 3.27 PostImage/Parents Not Members 2.93 American Journal of Management Vol. 16(2) 2016 95 DISCUSSION This research examined the effect of union members’ behavior during a press conference where bargaining unit members opposed a recent union action. The union members chanted “Shame on you” repeatedly during the press conference making it difficult for the audience to hear what the participants were saying. Respondents’ general attitudes toward unions, perceived union instrumentality and perceptions of union image all significantly deteriorated after viewing the video of the press conference. The results clearly show that some union protest behavior may have a negative impact on the union attitudes of observers. While the union may believe it is acting to protest decisions or actions perceived to be harmful to the union, this very protest behavior may damage its public image. Perhaps an even stronger finding is the results involving whether observers’ parents were currently or had ever been union members. The results showed that before viewing the video, observers whose parents had been union members had significantly more positive attitudes toward unions than respondents whose parents had never been union members. However, after viewing the video, there was no significant difference between respondents’ general union attitudes and perceived instrumentality based on parental union membership. While a significant difference based on parental union membership remained after viewing the video, the level of statistical significance declined. The results have great practical significance for unions. They should take great care in the public behavior of union members, especially when that behavior may be recorded and broadcast in the public media or on the Internet by opposing groups. The results of this study show that aggressive, protesting behavior by union members may have a damaging effect on observers’ attitudes toward unions, even if those observers are predisposed to view unions positively as a result of family socialization. One limitation of the study is the nature of a student sample. However, respondents were all senior level business students, near graduation and entering the business world. Some of them may be faced with decisions about union membership as they begin their work careers. Previous research shows that family socialization is an important determinant of union attitudes even after work experience. Thus, the impact of this limitation may be minimal. REFERENCES Bandura, A. (1965) Vicarious processes: A case of no-trial learning. In L. Berkowitz (Ed.) Advances in experimental social psychology, Vol. 2. New York: Academic Press, pp. 1-55. Bandura, A. (1977). Social Learning Theory. New York: General Learning Press. Barling, J., C. Fullagar & E.K. Kelloway (1992) The union and its members: A psychological approach. New York: Oxford University Press. Berger, P.L. & T. Luckman (1980) The social construction of reality: A treatise in the sociology of knowledge. New York: Irvington Publishers. Berger, S.M. (1962) Conditioning through vicarious instigation. Psychological Review, 69: 450-466. Brett, J.M. (1980) Why employees want unions. Organizational Dynamics, 8: 47-59. Chang, T.F.H. (2003) A structural model of race, gender, class and attitudes toward labor unions. Social Science Journal, 40(2): 189-200. Davy, J.A. & F. Shipper (1993) Voting behavior in union certification elections: A longitudinal study. Academy of Management Journal, 36: 187-199. DeCotiis, T.A. & J. LeLouarn (1981) A predictive study of voting behavior in a representation election using union instrumentality and work perceptions, Organizational Behavior and Human Performance, 27: 103-110. Deshpande, S.P. & J. Fiorito (1989) Specific and general beliefs in union voting models. Academy of Management Journal, 32: 883-897. Doop, L.W. (1947) The behavior of attitudes. Psychological Review, 54: 135-155. Fullagar, C. D. McCoy, & C. Schull (1992) The socialization of union loyalty. Journal of Organizational Behavior, 13(1): 13-26. 96 American Journal of Management Vol. 16(2) 2016 Getman, J.G., S.B. Goldberg, & J.B. Herman (1976) Union representation elections: Law and reality. New York: Russell Sage. Hemmasi, M. & L.A. Graf (1993) Determinants of faculty voting behavior in union representation elections: A multivariate model. Journal of Management, 19: 19-32. Hester, K. & J.B. Fuller (1999) An extension of the family socialization model of union attitudes. The Journal of Social Psychology, 139(3): 396-398. Kanekar, S. (1973) Observational learning of attitudes. Journal of Social Psychology, 90: 231-237. Kanekar, S. (1976) Observational learning of attitudes: A behavioral analysis. European Journal of Social Psychology, 6(1): 5-24 Kanekar, S., P. Karoly & A. Newman (1974) Source of feedback, observational learning, and attitude change. Journal of Personality and Social Psychology, 29: 30-38. Kelloway, E.K. & T. Newton (1996) Preemployment predictors of union attitudes: The effects of parental union and work experiences. Canadian Journal of Behavior Science, 28: 113-120. Kochan, T.A. (1979) How Americans view labor unions. Monthly Labor Review, 102: 23-31. McShane, S. (1986) General union attitudes: A construct validation. Journal of Labor Research, 7: 403- 417. Montgomery, R. (1989) The influence of attitudes and normative pressures on voting decisions in a union certification election. Industrial & Labor Relations Review, 42(2): 262-279. Premack, S.L. & Hunter, J.L. (1988) Individual unionization decisions. Psychological Bulletin, 103: 223- 234. Radtke, R.C. (1967) Effects of verbally mediated drive on a motor response and evaluative ratings. Journal of Experimental Psychology, 73: 22-27. Staats, A.W. (1968) Social behaviorism and human motivation: Principles of the attitude-reinforcer- discriminative system. In A.G. Greenwald, T.C. Brock & T.M. Ostrom (Eds.) Psychological foundations of attitudes. New York: Academic Press, pp. 33-66. Summers, T.P., J.H. Betton & T.A. DeCotiis (1986) Voting for and against unions: A decision model. Academy of Management Review, 11(3): 643-655. Zimbardo, P. G., & Leippe, M. (1991). The psychology of attitude change and social influence. New York: McGraw-Hill. American Journal of Management Vol. 16(2) 2016 97