DCI Staff Attend SIOP’S 27th Annual Conference Held in San Diego

Notably, EEOC Chair Jacqueline Berrien provided the keynote address for the conference’s theme-track. A “theme track” is essentially a conference within a conference that took place on the first day of the conference. The track included back-to-back sessions that were all focused on one topic, which was the “Science and Practice Perspectives on Contemporary Workplace Discrimination”, a particularly relevant topic for the federal contractor community. Summaries for many of the theme track sessions, as well as other relevant sessions, are provided below. We’ve grouped sessions into three sections: (1) Compliance, (2) Diversity and Inclusion, and (3) Discrimination. DCI Consulting Group staff members were involved in a number of those presentations.

• No one job analysis methodology is best for every situation, and even good job analysis methodologies will be scrutinized by judges or colleagues. Sometimes, a method must be “invented” in order to effectively to meet the objectives of the organization and the requirements of the courts.
• Consultants/psychologists should perform the job of interest to get a unique perspective that allows for better questions and research.
• Involve a physician whenever there is an issue of ADA compliance. It’s always important to get information from “the right people” and in the case of ADA issues; the right people are doctors and not incumbents (who are traditionally the subject matter experts).
• Judges and jurors typically like information from direct observation and structured surveys. Subject matter expert ratings and confirmation of results tend to “push a judge over the edge”
• The courts respond positively to the use of “rater confidence” metrics provided by SMEs.
• It’s a difficult balance to get the right level of detail; it’s a fight between trivializing the job with generalities and penalizing with minutia. There is also a balance between a good, thorough job analysis and the cost to clients.
• Documentation of the job analysis process and findings is vital to legal defensibility.
• An organization’s job structure should not be automatically trusted to provide accurate representations of the relationships between jobs. Many organizations have poor classifications.
• If the organization is unionized, union reps should be involved as early as possible to avoid conflict with, or sabotage by, union members.
• Cumulative knowledge (i.e., O*NET) is a great starting point for job analysis but should not be relied on exclusively.

Another interesting session was a community of interest on EEO issues. Community of interest sessions are informal and allow participants to come and go as they please, and ask any questions that they would like. Dr. Dunleavy facilitated this community of interest to a full room of participants. A number of interesting questions raised focused on the Americans with Disabilities Act, content oriented validation strategies, recent EEOC and OFCCP enforcement patterns, and regulatory change related to EEO/AA for veterans and persons with disabilities. Notably, the session was attended by EEOC staff and they contributed substantially to the conversation. The session ended with a preview of the Supreme Court’s decision in Fisher versus the University of Texas, which is the latest challenge to affirmative action and the use of race in college admission decisions.

• Information is power: Stay on top of current research and best practices
• Use structured selection tools
• Measure adverse impact in strategic and reasonable ways that mirror reality
• Consider reasonable alternatives and document
• Think strategically about what validation strategies make sense in a given context

Another interesting presentation at the conference was a symposium entitled “Racial Differences in Personnel Selection: Complex Findings and Ongoing Research”. The session was chaired by Dr. Leaetta Hough and included a balanced set of paper authors from academia and practice. The first presentation from Oppler et al. focused on whether the extent to which test taker performance is over- or under-predicted or could be a factor in identifying reasonable alternatives as is required by the uniform guidelines. The second presentation by Berry et al. focused on recent research on differential validity, and suggested that variance differences were not an explanation for why differential validity may exist. Next, Scherbaum et al. summarized some recent research on the Sienna Reasoning Test (SRT) which is a relatively new measure of cognitive ability that apparently has lower subgroup differences and similar prediction as compared with traditional measures of cognitive ability. McDaniel et al. presented an alternative explanation for SRT results, which is called Spearman's hypothesis. This approach suggests that cognitive ability tests with lower subgroup differences are in fact measuring less cognitive ability and more of other constructs. The session then became heated as a debate began over the extent to which the developers of the SRT have made all background and research data available to the SIOP community. The session ended with an interesting set of discussants. Dr. Jim Outtz focused on some of the positive aspects of SRT research, while Dr. Deniz Ones conducted a series of analyses on the data made available to the panel and suggested that more research is needed to understand the SRT.

• Presenters felt that the ADA has not made a large enough impact. They noted that after Title VII was put into effect, there was a quick, and major, shift in the employment rates of women and minorities. This change hasn’t been seen for individuals with disabilities and it has been 20 years since ADA.
• When asked about the proposed changes to Section 503 by the OFCCP, the presenters felt that they would be difficult for organizations to implement. It was noted that organizations typically have a difficult job in obtaining more than a 2% population of individuals with a disability, so the 7% goal would be very difficult to achieve. Self-identification pre-offer was also met with criticism.
• Often a perception of employers is that individuals with disabilities will result in high costs to the organization because of expensive accommodations and healthcare increases.
• Individuals with disabilities are detached from the workforce.
• The differences between subtle and overt mistreatment, as individuals with disabilities tend to experience the former more often.
• Psychological and neurological disabilities are more often discriminated against than physical because organizations are worried performance will be less stable and more difficult to predict.

• EEOC settled 3,671 cases over the past decade with the most settlements arising from the Philadelphia district.
• Sex- and race-based cases comprised the large majority, at approximately 59% of total cases and 23% of total cases, respectively.
• Alarming from an industrial-organizational psychology perspective was the fact that very few cases required the implementation of structured processes as part of the remediation plan.
• An interesting trend in the OFCCP space concerns the increase in compensation-related settlements. As noted by the panel, in 2011, approximately 1/3 of all OFCCP settlements were compensation related, although the settlements tended to be small in pure dollar amounts.

Another interesting session focused on the Supreme Court ruling and Dukes versus Wal-Mart. Dr. Eric Dunleavy provided a short summary of the Supreme Court ruling, as well as the earlier district and appeals court rulings. Readers of this blog know that the Supreme Court ruled in favor of Wal-Mart, decertifying the class because the plaintiffs did not have “the glue” that would connect the employment decisions and potential victims together. Next, Dr. William Bielby, a sociologist who was the plaintiff expert in the Wal-Mart case presented the latest research on social framework analysis. Recall that Justice Scalia was not a fan of the social framework analysis, and suggested in the written decision that it did not inform on whether there was any causal mechanism to potential disparities against women. Perhaps the point that is most interesting from Justice Scalia’s treatment of social framework analysis is the fact that it was unclear whether discrimination was the explanation for 5% of the decisions or 95% of the decisions. Dr. Bielby suggested that the social framework analysis was misunderstood by the Supreme Court, and that it may be a more useful paradigm for defense attorneys.

Dr. Jerry Barrett then presented a detailed criticism of social framework analysis, with particular emphasis on laboratory research. Dr. Barrett reviewed research concluding that there was an inherent bias that was unconscious and emphasized that none of this social psychology research conducted in the lab has been replicated in organizational settings. Dr. Barrett suggested that there is no evidence that generalizes to the workplace. The session concluded with a presentation by Dr. Mike Campion, who discussed what the Wal-Mart ruling may have changed in terms of best practices in I/O psychology, and what best practices are still best practices. One of these points focused on the fact that unstructured and subjective decision-making at Wal-Mart was decentralized and extremely subjective, and this was one reason why the class could not be certified. However, structured scientific approaches to selection may still offer more defensibility. The session concluded with a reminder that the Wal-Mart saga is not over, and that smaller classes of females are taking their cases to individual courts. Stay tuned.

As you can see by the above summaries, once again the SIOP conference included a number of informative and cutting edge EEO presentations. We expect the same next year at the SIOP conference in Houston, as I/O Psychologists continue to perform important work in the EEO context.