- Hebl, M. R., Foster, J. B., Mannix, L. M., & Dovidio, J. F. (2002). Formal and Interpersonal
Discrimination: A Field Study of Bias toward Homosexual Applicants. Pers Soc Psychol
Bull., 28, 815-825.
2.
The study consisted of the following ten independent variables: job availability, job callback, word count, interaction length, coded negativity, perceived negativity, stigma, perceived employer interest, permission to use bathroom, and permissions to complete application. In order to test the individual independent variables, the authors made use of a one-way MANOVA. A univariate analyses was also conducted on the same as a follow-up. On the other hand, correlation between specific independent variables was done via the Ficher’s Z transformation, which acted as an evaluation of the construct validity of the variables.
3.
The dependent variables of the study were perceived employer interest (by the applicants), perceived negativity (by the applicants), and coded negativity (as evaluated by the independent raters). In each of these three dependent variables, the use was made of a likert-scale for purposes of analysis. Thereafter, principle components analysis on the individual items of the likert-scale were performed on each of the three aspects of the dependent variables.
Participation to this study was on a voluntary basis. The participants included eight female and eight male graduate and undergraduate students drawn from a university in Texas. The researcher required each of 16 participants to write application letters for jobs, addressed to six different stores. While it was expected that a total of 96 applications would be made, five of the target stores had been closed and therefore the researcher ended up with 91 different stores. However, during the data compilation stage, the researchers discovered that there were irregularities with seven of the entries as participants saw labels “printed on their hats in store mirrors or glass doors” (Hebl et al., 2002, p. 817). Therefore, the ensuing results of the research study were based on the remaining 84 interactions.
The authors of the current study sought to test three key hypotheses. First, based on the proposal that understanding stigma entails learning the attributions, behaviors, and attitudes of the target and the stigmatizer, the researchers hypothesized that where there is a high likelihood for potential stigmatize to remain guarded about expressing formal and overt types of discrimination, they could also unknowingly manifest biases in subtle ways. With regard to the victim of stigmatization, however, such interpersonal and overtly subtle expressions of bias could be interpreted as cues for exceedingly overtly and formal forms of bias.
Secondly, the authors also assessed the interpersonal behaviors of future employers towards potential job applicants who presented themselves as either heterosexual, or gay/lesbian. The authors hypothesized that even as potential could manage to limit and control possible formal discrimination against lesbian and gay applicants on the issue of interpersonal discrimination, there is the likelihood that they could be less motivated to, able to, and attuned to control it. Based on this hypothesis, the authors made a prediction that there was a higher chance for stigmatized applicants to have their interactions terminated by employers faster than those of non-stigmatized applicants.
Finally, based on the findings of considerable research that such subtle interpersonal cues as nonverbal and paraverbal behavior have a powerful role to play in an individual’s attributions, impressions, and expectations, Hebl et al. (2010) arrived at another hypothesis; under stigmatized roles, confederates were more likely to perceive as being treated in a negative manner by potential employers than when under stigmatized roles.
Results
Formal discrimination
The overall MANOVA (one-way multivariate analysis of variance) was conducted to determine the level of displayed formal discrimination among employers, in which 72 cases of all available variables were included. The MANOVA analyses failed to yield “a significant stigma main effect for measures of formal discrimination” (Hebl et al., 2010, p. 820). A similar pattern of results was also reflected by separate analyses conducted on individual variables. When asked by applicants if there was a job vacancy, the affirmative response from employers was more likely to be in the affirmative (56%), in the case of applicants in non-stigmatized condition, than in the case of gay condition (43%). More non-stigmatized applicants (71%) were likely to receive permission from potential employers to complete a job application compared with stigmatized applicants (60%). Non-stigmatized applicants were slightly more likely to receive a job callback (19%), in comparison with stigmatized applicants (12%). Stigmatized applicants were almost equally likely to get permission to use the washroom (51%) as their non-stigmatized counterparts (56%). Therefore, on formal discrimination measures, employers were not seen to respond more positively to non-stigmatized applicants more than they did with stigmatized applicants.
Interpersonal assessment of discrimination
Since it was not possible for the researchers to undertake word count and interaction length on all interactions, they used 61 cases to determine interpersonal discrimination. Accordingly, they conducted a one-way MANOVA on the following variables: word count, interaction length, coded negativity, and perceived negativity. Predictably, the findings revealed “a significant main effect for stigma, F (4, 56) = 3.38, p < .03, η2 = .19” (Hebl et al., 2010, p. 820). Univariate analyses conducted as a follow-up revealed that the effect of stigma was consistent across the measures. Stigmatized applicants were more likely to have fewer words spoken to them ((M = 169.45, SD = 111.05) relative to their non-applicants (M = 257.18, SD = 210.08). In the same way, when talking to stigmatized applicant, employers were likely to make use of shorter interaction length (M = 4 min., 5 sec.), as opposed to when interacting with non-stigmatized applicants SD = 3 min., 45 sec.) than with non-stigmatized applicants (M = 6 min., 23 sec.) (Hebl et al., 2010, p. 821).
In addition, an evaluation of coded negativity as completed by the study’s independent rates indicated that when dealing with stigmatized applicants, employers were more likely to manifest higher rates of coded negativity (M = 4.40, SD = 1.43) relative to when interacting with non-stigmatized applicants (M = 4.01, SD = 1.42). Perceived employer interest was negatively associated with perceived negativity [r (84) = –.50, p < .001]. There was also a significant correlation between job callback and perceived applicants’ negativity [r (84) = –.23, p < .05].
6.
Even as only a modest number of interactions were observed by the study’s applicants, they, nonetheless, managed to detect discrimination. Largely, employers were perceived to be overly nervous, hostile, less helpful and interest, when dealing with stigmatized applicants, more than non-stigmatized applicants. There was also a higher likelihood of employers avoiding eye contact in case the hat of the applicant stated “Gay and Proud”, and this is an indication of the obvious bias evidenced in the formal and interpersonal discrimination by potential employers. The above research findings are indicative of the diverse expectations and impressions by both targets and stigmatizers.
Internal validity is an indication of evidence that the outcome of the study being carried out is as a result of what was done to the study, that is, the process involved in the research. An internal validity of a study can be tested by determining the causal relationship between the variables. In the current study, internal validity was determined by testing the study’s hypotheses. On the other hand, external validity of a study entails causal or generalized inference as they apply to scientific research, often on experimental validity. Put simply, external validity refers to the level to which the results of a given study could be generalized to other people or situation. In the current study, this was achieved by using volunteers as the study’s participants.
- Evaluate the experimental realism and mundane realism of the experiment.
Experimental realism described the level to which an experimental manipulation has pro-actively involved study participants in the study. The goal is to ensure that the experiences of the study participants match the psychological states desired by the experimenters. It means therefore that the experimenters have to control for influences in the experiment. In the current study, experimental realism was achieved by first conducting a pilot study to determine what the actual study would look like. In contrast, mundane realism refers to the level to which the whole study or a specific activity in the study is aligned to the process or activity an individual would likely complete in their daily lives.
The experiment involved human subjects and as such, there was need to consider the ethical issues that are likely to affect the applicants and respondents. Accordingly, the researchers should be concerned about the welfare of the applicants. In this case, there is need to seek written consent from the applicants. In this case, the authors ought to inform them about what the research entails, what role they are expected to play, and that they are free to leave the study at any point. Moreover, they should be informed that they are not being coerced to partake in the study. More importantly, because the applicants are students, there is need for the researchers to seek consent from ethics board of their respective universities. On the other hand, the respondents to the study (storeowners) should also be informed about the study they are to partake in. Likewise, they should be free to participate in or leave the study at any point. In addition, the respondents should be informed that their respondents should only be used for purposes of informing the study.
I found this article to be interesting in a number of ways. First, the article has done well to shed light on the various levels of interpersonal and formal discrimination evident in the workplace. In addition, the study’s research findings also reveal that while discrimination in the workplace on the basis of one’s sexual orientation is illegal, it is still prevalent, albeit to varying levels. However, according to the research findings, it is quite difficult to document the type of discrimination that lesbian and gay are likely to undergo when applying for jobs, from potential employers.
References
Hebl, M. R., Foster, J. B., Mannix, L.M., & Dovidio, J. F. (2002). Formal and Interpersonal
Discrimination: A Field Study of Bias Toward Homosexual Applicants. Pers Soc Psychol
Bull., 28, 815-825.