Society for Industrial and Organizational Psychology Conference (SIOP)

Paper Presented at SIOP 2002 in Miller (Chair) Symposium, Cultivating a Diverse Workforce: Pitfalls and Potential Treasure 

Corey E. Miller, Ph.D., Anupama Narayan, M.Ph., Mark Palumbo, Esteban Tristan

Doverspike, Taylor, and Arthur (2000) reviewed the literature concerning research looking for antecedents to commitment to affirmative action programs. They suggested that although research has been done concerning demographics like race and sex, racism and sexism, Individualism and Collectivism, and liberalism and conservatism, there is a lack of research concerning Personality. Doverspike et al. (2000) went on to propose a theory incorporating personality. They hypothesized that an individual with a "Cosmopolitan Personality" is likely to be committed to diversity. The cosmopolitan personality would have a culturally open personality, or high on Openness to Experience, but would also have a multicultural knowledge structure as a result of exposure to diverse cultures. The present research attempted to test this hypothesis.

MS Powerpoint Slides

Paper Presented at SIOP 2002 in McFarland (Chair) Symposium, Applicant Faking: New Perspectives on an Old Issue

Corey E. Miller, Ph.D., and Esteban Tristan

MS Powerpoint Slides w/ Text in Notes 

Panel Discussion Conducted at SIOP 2001 and Chaired by Corey E. Miller

Conclusions drawn from research concerning faking of personality selection measures range from faking has no effect to faking destroys validity. Underlying the debate is the measurement of social desirability.Panelists will draw upon their experience as researchers, practitioners, and test developers in discussing misconceptions and future research needs.
 

Corey E. Miller, Wright State University, Chair

Gerald V. Barrett, Barrett & Associates, Panelist

Robert T. Hogan, University of Tulsa, Panelist

Leaetta M. Hough, The Dunnette Group, Ltd., Panelist

Douglas N. Jackson, University of Western Ontario, Panelist

Deniz S. Ones, University of Minnesota, Panelist

Summary

APA

The Effectiveness of a Case-law Based Sexual Harassment Prevention Training Program
 

Corey E. Miller
 

Paper Presentation at APA 2001

Summary
 

International Personnel Management Association

Award Address (Dissertation Award Winner for 2001): The Coachability and Fakability of Personality Selection Tests

Paper in MS Word Format

IO-OB Student Conference

Expanding the Definition of Faking Beyond Social Desirability: The Case for Job Desirability

Esteban Tristan and Corey E. Miller, Ph.D.

Paper Presented at the IO-OB Conference 2001

Abstract

Early research on faking suggests that faking consists primarily of social desirability (Edwards, 1957).However, recent literature shows that faking may be mostly comprised of a bias specifically related to the particular job an individual is applying for, or “job desirability” (Kluger, Russell, & Reilly, 1991; Kluger & Colella, 1993).

Xiaofang Zeng and Corey E. Miller, Ph.D.

To be Presented at IO-OB Conference 2002

There has been an explosion of interest in the construct of emotional intelligence in both the scientific literature as well as the general media. There are many different conceptualizations of emotional intelligence ranging from conceptualizations of emotional intelligence as a type of intelligence to conceptualizations of emotional intelligence as a new personality construct. Several researchers have suggested that emotional intelligence is an important predictor of job performance. The research has not to date shed much light on where emotional intelligence would lie in the nomological net. This study explored the relationship of emotional intelligence to other personality variables.

Participants were administered two measures of Emotional Intelligence, Schutte’s Emotional Intelligence Scale the Trait Meta-Mood Scale (TMMS) and items from Goldberg’s International Personality Item Pool (http://ipip.ori.org/ipip) which measured the Big Five Personality Dimensions. If emotional intelligence is a construct that is distinct from the Big Five, one would expect low correlations with the Big Five scales and an inability to measure emotional intelligence using the IPIP items. Results revealed significant correlations between the Big Five personality measures and the measures of Emotional Intelligence. Further analyses revealed that the construct of emotional intelligence could be measured with items from the Big Five measures. Thus, the results suggest the Emotional Intelligence is a construct that is strongly related to the Big Five measures of personality, and is unlikely to be a construct that occupies a place on the nomological net distinct from Big Five measures of personality. Implications of this finding for the literature as well as future directions for research will be discussed.

My main research interests are in Personnel Psychology. My research program explores the legal context of Personnel Psychology, and how legal, social, and political initiatives play out in terms of diversity management, with particular emphasis of the design and content of Personnel Selection Programs in such a climate. I have begun a program of research on the use of personality (or non-cognitive) measures in Personnel Selection. Specifically, my program of research is looking at controlling faking or response distortion, more specifically at exploring the construct of job desirability. I am interested in this area of research because the use of non-cognitive measures is driven by an attempt to reduce adverse impact, or allow an employer to increase the diversity of their workforce. This theme of legal issues in Personnel Psychology is also evident in my research on exploring the antecedents of commitment to diversity in organizations, my research on the effectiveness of a sexual harassment prevention training program, and my work on the application of stereotype theory in employment litigation. In addition, I also am collaborating on a series of papers investigating an analysis of reliability of performance appraisals.

Investigating the Use of Personality Measures in Personnel Selection

My program of research investigating faking of personality measures in personnel selection grew out of my dissertation research, which won the International Personnel Management Association's (IPMA) 2001 Dissertation competition. The main thrust of this research was examining the effect coaching might have on personality measures used for selection. Coaching programs are common for safety forces jobs (Ryan, Ployhart, Greguras, and Schmit, 1998; Zickar & Drasgow, 1996). In fact, there have even been occasions where I/O Psychologists who lose bids to develop the tests for municipalities develop training programs to thwart the efforts of the consulting firms that develop the tests. The Department of Justice has taken legal action against companies that offer training on how to beat civil service tests. It has been documented that these companies receive over 100,000 calls a month. The test coaching industry is a large industry, and I fear our field will have problems similar to the Educational Testing Service's problems with test coaching companies.

My dissertation showed that a lecture based on the Big Five personality factors, which is commonly presented in Introductory Psychology classes, enabled individuals to raise their score dramatically on the measures. In fact, it is quite likely that the only individuals hired would be the applicants that had this information. The literature review of my dissertation revealed that the use of social desirability measures to partial out faking variance is virtually doomed to failure. The literature shows that Social Desirability measures tap construct relevant variance to at least some extent, as evidence by their correlations with criteria such as training performance (Ones, Viswesvaran, & Reiss, 1996). Thus, they will not function as suppressor variables as correcting personality measures for social desirability will partial out construct relevant variance, not just construct irrelevant variance.
 

In my Dissertation I also included a preliminary study on job desirability. My first graduate student (Esteban Tristan) presented this data set at the 2001 IO/OB student conference. We have collected a large sample of follow-up data that I am now in the process of analyzing, and have submitted as part of a SIOP symposium. In this research I develop a measure of job desirability. For example, an item on a social desirability scale might be "I always declare everything at customs." If people answered "True", the inference would be that they are trying to present an overly favorable impression. In contrast, an item from my job desirability scale is "I have read the Ohio state Constitution Article concerning physical ability standards for police officers." The implication would be that someone who answered "True" to this item is lying, as no such Article of the Ohio State Constitution exists.

The data set I have collected reveals that Job Desirability is a better predictor of faking than social desirability (Miller & Tristan, under review). Although social desirability scales (the BIDR, the Marlowe-Crowne) experienced a standardized mean difference (commonly known as an effect size) of about 1.5 Standard Deviations, the Job Desirability scale experienced a standardized mean difference of over 2.5 standard deviations. A hierarchical regression revealed that the Job Desirability scale accounted for unique variance over and above two parallel forms of the Marlowe-Crowne, and the BIDR. Thus, researchers must begin using job desirability scales, as social desirability scales do not capture all faking.
 

I have also run a second set of analyses on this data set to determine the relative fakability of the different tests. I found that James's Conditional Reasoning Test of Achievement Motivation was resistant to faking, in fact experiencing slightly lower scores in the faking condition than the honest condition. The test is described to the participant as a reasoning test and that the must choose the answer that makes the most sense. For example, and individual might be presented with information suggesting that top executives are more likely to develop heart disease than the general population. An individual who chooses the response that not all executives get heart disease, so the job of top executive is still a highly desirable position, would get one point. An individual who respond that the job of top executive is not a desirable position would get one point. Although the test was resistant to faking, it had more than a half of a standard deviation of adverse impact. I believe the reading level of the passages might cause this. The reading level averaged above twelfth grade reading level. In the future I would like to design such a test with a lower reading level. In the study I could also test for the effects of stereotype threat.

A third set of analyses I have run with this data set concerns the construct validity of the social desirability measures. I have found that the measures are measuring different constructs in the honest and faking conditions as evidenced by the correlations. This finding is corroborated by recent research (Stark, Chernyshenko, Chan, Lee, & Drasgow, 2001). This finding has implications for studies that compare the effects of social desirability on the factor structure of personality measures in applicant and non-applicant samples (i.e. Ellingson, Sackett, & Hough, 1999; Ellingson, Smith, & Sackett, 2001; Smith, Hanges, & Dickson, 2001). Selecting individuals high on social desirability (or low) and comparing them in applicant and non-applicant samples is a questionable test of the effects of social desirability if social desirability measures are not measuring the same construct across the conditions.
 

A future study I have in mind is using a job desirability measure in the context of a biodata measure. I think that it would fit into such a context very well. Some global companies are moving to biodata, as it is a very useful tool to screen applicant over the Internet, enabling a global reach. A job desirability measure could be used to flag those individuals who the company should do a follow up to determine the truthfulness of their answers.

My second graduate student is beginning to collect data on an analysis of the emotional intelligence construct in the context of other personality variables and prediction of job performance (Miller & Zeng, to be presented at IO-OB 2002).
 

Diversity Management
 

I have two papers on the topic of diversity management that I am currently preparing for submission to a journal. The first is on the effectiveness of a case law based sexual harassment training program. Although there has been a great deal of research on sexual harassment in the recent past, the effectiveness of sexual harassment prevention training programs has not been demonstrated, and in fact much of the evidence shows they are not effective. However, my training program is effective. I use the work of Fitzgerald and associates to determine which behaviors women define as sexual harassment. I then found court cases in which an individual displayed this behavior, and then the company lost the case. I use a revised Sexual Experiences Questionnaire (SEQ) as an outcome measure. I found that men (as well as women) would be more likely to discipline individuals who display these sexually harassing behaviors. I am currently in the process of negotiating with two police departments to allow me to present my training program, and collect publishable data on its effectiveness. Although, I don't follow up explicitly planned out yet, I certainly see the opportunity to use these contacts to collect follow-up data, likely not limited to just sexual harassment.
 

I am also preparing a second manuscript for publication on diversity management. My graduate students and myself developed measures of Exposure to Diversity and Commitment to Diversity. We found that Exposure to Diversity and Openness to Experience predicted Commitment to Diversity. The interaction was significant for minorities, but not Caucasians. (For a full discussion see the enclosed SIOP submission: Miller, Narayan, Palumbo, & Tristan, under review). I have ideas for follow-up studies on this topic as well. Wright State University has a uniquely high population of individuals with disabilities. It is common for non-disabled students to interact with these disabled individuals, and they cannot help but interacting with them. I would like to do a comparison of the freshmen at Wright State and another university. I think that we could see the attitudes to individuals with disabilities change at Wright State, in contrast to a University with few individuals with disabilities. I plan to submit this idea as a grant proposal. I have also made some preliminary contacts with individuals in the military. I might have to opportunity to obtain a grant from the military to study this phenomenon with military recruits. I would think many of these recruits would be exposed to minorities for the first time, or at least the first time of any importance, which would be a great context to do a follow-up study.
 

Psychology and the Law
 

I also have three papers on the topic of Psychology and the Law. Williamson, Campion, Malos, Roehling, and Campion (1997) analyzed court cases and concluded that interview structure contributed to defensibility of interviews. We collected ratings from individuals that were blind to the intent of the research. The analyses show that the vast majority of the cases did not contain enough information to determine if structure contributed to legal defensibility of the interviews. We found that the best predictor was whether or not there was evidence of discriminatory intent in the statements of the interviewers. Possible follow-ups might include questionnaires directed at EEOC investigators to determine which business practices contribute to legal defensibility. After reviewing the cases involved in this analysis, I have concluded that the judge just simply does not include enough information to reliably determine which factors lead to her decision, if she could even determine that herself. Combined with the fact that less than 1% of cases actually go to trial (which could not be an adequate sample as those 1% typically have unusual features or they would be settled), I have concluded that examining court decisions is simply not a reliable base to base these conclusions on.
 

I have two papers on the topic of the application of stereotype research in the courtroom. My research shows that the stereotypes of women in management are not as negative as many research suggest, or expert witnesses claim. I plan to focus future research efforts at trying to discover strategies successful females managers have used, with the goal of offering constructive advice on things women in management can do to aid their career. In my opinion, the research is too focused on linking stereotypes and discrimination, and offers little in the way of practical advice women could actually use. My discussions with successful women have shown me that there are things that women in management can do, and I believe that research based on this perspective could be very beneficial.

Ellingson, J. E., Sackett, P. R., & Hough, L. M. (1999). Social desirability corrections in personality measurement: Issues of applicant comparison and construct validity. Journal of Applied Psychology, 84, 155-166.

Ellingson, J. E., Smith, D. B., & Sackett, P. R. (2001). Investigating the influence of social desirability on personality factor structure. Journal of Applied Psychology, 86, 122-133.

Miller, C. E. (2001, August). The effectiveness of a case-law based sexual harassment training program. Paper presented at the 109^th annual convention of the American Psychological Association, Washington, D. C.

Miller, C. E., & Barrett, G. V. (in progress, a). The susceptibility of personality selection tests to coaching and faking. Journal of Applied Psychology.

Miller, C. E., & Barrett, G. V. (in progress, b). Measurement Issues in Faking Research: A comment on the use of Social Desirability measures as suppressor variables. Journal of Applied Psychology.

Miller, C. E., Barrett, G. V., & Doverspike, D. (under review). An Evaluation of the Red Flag Words Technique in the Context of the Career Woman and Career Man Stereotypes. Law and Human Behavior.

Barrett, G. V., Miller, C. E., & Doverspike, D. (under review). The Application of Stereotype Research in the Courtroom: The Relative Importance of Individuating Information. Law and Human Behavior.

Miller, C. E. (under review). Chair. In C. E. Miller (Chair), Cultivating a diverse workforce: Pitfalls and potential treasure, symposium submitted to the 17^th annual meeting of the Society for Industrial and Organizational Psychology, San Diego, CA.

Miller, C. E., Narayan, A., Palumbo, M., & Tristan, E. (under review) An Examination of the Antecedents of Commitment to Diversity. In C. E. Miller (Chair), Cultivating a diverse workforce: Pitfalls and potential treasure, symposium submitted to the 17^th annual meeting of the Society for Industrial and Organizational Psychology, San Diego, CA.

Miller, C. E., & Tristan, E. (under review). Expanding the Definition of Faking Beyond Social Desirability: The Case for Job Desirability. In L. McFarland (Chair), New approaches to dealing with the faking of non-cognitive measures, symposium submitted to the 17^th annual meeting of the Society for Industrial and Organizational Psychology, San Diego, CA.

Miller, C. E., & Zeng, X., & (in progress). A comparison of emotional intelligence scales and traditional personality scales in a personnel selection context.

Ones, D. S., Viswesvaran, C., & Reiss, A. D. (1996). Role of social desirability in personality testing for personnel selection: The red-herring. Journal of Applied Psychology, 81, 660-679.

Ryan, A.M., Ployhart, R.E., Greguras, G.J., & Schmit, M.J.(1998).Test preparation programs in selection contexts: Self-selection and program effectiveness.Personnel Psychology, 51, 599-621.

Smith, D. B., Hanges, P. J., & Dickson, M. W. (2001). Personnel selection and the Five-factor Model: Reexamining the effects of Applicant's frame of reference. Journal of Applied Psychology, 86, 304-315.

Stark, S., Chernyshenko, O. S., Chan, K., Lee, W. C., & Drasgow, F. (2001). Effects of testing situation on item responding: Cause for concern. Journal of Applied Psychology, 86, 943-953.

Thornton, C., & Miller, C. E. (revise & resubmit). Using the Correction for Attenuation Equation to Obtain Accuracy Scores. Journal of Applied Psychology.

Williamson, L. G., Campion, J. E., Malos, S. B., Roehling, M. V., & Campion, M. A. (1997). Employment interview on trial: Linking interview structure with litigation outcomes. Journal of Applied Psychology, 82, 900-912.

Zickar, M.J., & Drasgow, F.(1996).Detecting faking on a personality instrument using appropriateness measurement. Applied Psychological Measurement, 20, 71-87.