1
$\begingroup$

Traditionally, employee selection tests focus on cognitive abilities such as reading, reasoning, and math. These have large Black-White (B-W) effect sizes (standardized mean score differences). The range of other abilities that have been tried may be too narrow. This note is part of an effort to identify abilities that might plausibly be related to job performance and might have small B-W effect sizes but that linkage is not the topic of this question.

As pointed out in a comment, this topic is quite broad. I do not expect a direct answer. Rather I am hoping to get citations to recent published literature with information to answer this question: What abilities (or ability sub-areas) show little or no B-W mean score differences? For example, are there effect size estimates (and available measures) for: creativity, facial recognition (as in choosing a face out of a crowd), memory for faces, voice quality (as in "command presence" for a police officer), cooperativeness, fairness, calmness under pressure, warmth, etc. (The jobs I have in mind right now are Police Officer and Firefighter.)

Note: This question has a different focus than this related question: Reducing adverse impact, discrimination and unfairness when using psychological assessments for selection purposes

Addendum: In response to comments, let me clarify the reason for this broad question. Industrial psychologists tend to use a narrow range of cognitive ability areas when developing employee selection procedures, and the abilities considered typically are those that were known to the field in the mid 1900's, or before. Perhaps that is part of the reason that the best employee selection tests explain, at most, 25% of the variance in job performance. Beyond that, most employee selection procedures are supported by content validation evidence, not criterion related. Subject matter experts (SMEs) may be incapable of articulating entry-level cognitive abilities (i.e., not job knowledge) in other than the broadest ways. Typically, SMEs rate the ability areas provided to them in a questionnaire. If an ability is not included in the questionnaire, it is not rated and not considered for inclusion in the selection procedure. I am wondering if there are other cognitive ability areas that I might review to see if they are related to job performance, particularly abilities that show little or no B-W differences in mean scores. I am reaching out beyond my area of specialty. I hope to find review articles that summarize the information I seek.

| improve this question | |
$\endgroup$
  • $\begingroup$ I'm a bit confused by the question. The smallest differences would be cases where there is no effect of race, and there are certainly a huge number of examples where this is the case. I think you are asking something more specific, but the question is unclear to me. $\endgroup$ – Josh de Leeuw Apr 17 '15 at 19:47
  • $\begingroup$ @Josh Yes. I am seeking attributes, especially cognitive attributes, with small or zero effect size. Can you give some illustrations of the "huge number of examples" or point me to one or more sources? Thanks. $\endgroup$ – Joel W Apr 17 '15 at 22:14
  • $\begingroup$ The issue Josh is raising is essentially that the question is too broad to answer as written. You list a wide range of unrelated cognitive categories, so the question cannot be satisfactorily answered within the scope of SE. It would be better to pick a few cognitive attributes if you are interested in the general population, or to restate your question in terms of B-W differences among police officers or firemen specifically. $\endgroup$ – Christian Hummeluhr Apr 18 '15 at 7:52
  • $\begingroup$ @christian-hummeluhr Yes. I cast a wide net, deliberately. Let me explain why by editing the question. $\endgroup$ – Joel W Apr 19 '15 at 2:45
  • $\begingroup$ I sympathize. However, different jobs require different abilities, and you are (still) asking, as the question is currently written, for an answer that would require evaluating all possible job-ability combinations. Such a review paper almost certainly does not exist. I am therefore maintaining my close vote until such a time that the question scope is narrowed down to sufficiently define and specify either job (type) or abilities of interest. $\endgroup$ – Christian Hummeluhr Apr 19 '15 at 7:30