•The issue of the Cleary Model having longer-run increases in adverse-impact vs. short-run effects seems to be an issue yet it is a popular model in selection (at least at this time). Are companies required to continue to validate the model/selection procedures overtime even after initial determination of this? I would think long-run bias effects would be an issue in the legal aspect and bring the model to be considered unreliable.
This article, and the comparison of different fairness models in the Arvey and Faley chapter, suggests that while the law supports the Cleary model, there are advantages and disadvantages to defining fairness with different models. Why is the Cleary model the one most supported by the EEOC and legal system? Is there any other legal support or backlash for using one of the other fairness models in personnel selection?
Do you think a 20 year period is a reasonable amount of time? Is this too long as many things change in that time? Or is it too short in that these methods we use tend to stick around for a long time?
I am left with similar questions to Vicki. What kind of impact did this research have? Have other definitions of fairness (other than bias) as modeled here held up in court as supporting the goals of Title VII (for their decreased adverse impact on Black employment)?
These authors came to the conclusion that the Cleary model was better because you start to see utility differences over time. But even after twenty years (looking at Table 5), the Clearly model seemed to do only slightly better than the Thorndike method. Shouldn't these authors conclude that as long as you are taking into account one of the more recent definitions you are ok? I guess I don't really see the difference of using one over the other...
I just am not a fan of these simulation studies that use computer algorithms to predict long term social/behavioral outcomes. If anything is going to examine the generalization of Cleary on long-term adverse impact, it needs to be a longitudinal study. SO MUCH can happen throughout a person’s career and tenure at an organization. Do you think the authors are over stepping their boundaries to draw such strong conclusions using a simulation model?
So, it appears that while the Cleary rule is currently satisficing, it is only a matter of time before a court case comes along that invalidates it. Based on the findings of Steffy and Ledvinka (1989) the rule maximizes productivity at the expense of Black employment in the short run and even sacrifices overall productivity in the long run. Does it seem to anyone else that this is a very complex issue?
•The issue of the Cleary Model having longer-run increases in adverse-impact vs. short-run effects seems to be an issue yet it is a popular model in selection (at least at this time). Are companies required to continue to validate the model/selection procedures overtime even after initial determination of this? I would think long-run bias effects would be an issue in the legal aspect and bring the model to be considered unreliable.
ReplyDeleteThis article, and the comparison of different fairness models in the Arvey and Faley chapter, suggests that while the law supports the Cleary model, there are advantages and disadvantages to defining fairness with different models. Why is the Cleary model the one most supported by the EEOC and legal system? Is there any other legal support or backlash for using one of the other fairness models in personnel selection?
ReplyDeleteDo you think a 20 year period is a reasonable amount of time? Is this too long as many things change in that time? Or is it too short in that these methods we use tend to stick around for a long time?
ReplyDeleteI am left with similar questions to Vicki. What kind of impact did this research have? Have other definitions of fairness (other than bias) as modeled here held up in court as supporting the goals of Title VII (for their decreased adverse impact on Black employment)?
ReplyDeleteThese authors came to the conclusion that the Cleary model was better because you start to see utility differences over time. But even after twenty years (looking at Table 5), the Clearly model seemed to do only slightly better than the Thorndike method. Shouldn't these authors conclude that as long as you are taking into account one of the more recent definitions you are ok? I guess I don't really see the difference of using one over the other...
ReplyDeleteI just am not a fan of these simulation studies that use computer algorithms to predict long term social/behavioral outcomes. If anything is going to examine the generalization of Cleary on long-term adverse impact, it needs to be a longitudinal study. SO MUCH can happen throughout a person’s career and tenure at an organization. Do you think the authors are over stepping their boundaries to draw such strong conclusions using a simulation model?
ReplyDeleteSo, it appears that while the Cleary rule is currently satisficing, it is only a matter of time before a court case comes along that invalidates it. Based on the findings of Steffy and Ledvinka (1989) the rule maximizes productivity at the expense of Black employment in the short run and even sacrifices overall productivity in the long run. Does it seem to anyone else that this is a very complex issue?
ReplyDelete