Personnel Selection: Adding Value Through People - A Changing Picture
Format: PDF / Kindle (mobi) / ePub
This is a fully updated edition of Personnel Selection, a seminal text on the psychometric approach to personnel selection by a noted expert in the field.
- Focuses on cutting-edge topics including the influence of social networking sites, adverse impact, age differences and stereotypes, distribution of work performance, and the problems of selecting new employees using research based on incumbent employees
- Questions established beliefs in the field, especially issues that have been characterized as “not a problem,” such as differential validity, over-reliance on self-report, and “faking good”
- Contains expanded discussion of research and practice in the US and internationally, while maintaining the definitive coverage of UK and European selection approaches
- Provides comprehensive yet accessible information for professionals and students, as well as helpful pedagogical tools (technical and statistical boxes, simplified figures and tables, research agenda boxes, key point summaries, and key references)
from –0.45 to 0.75. Such wide variation means that employers can never be sure whether a selection test will work (moderately) well, or not very well, or not at all. Work psychologists initially hoped to find moderator variables (v.i.) to explain the wide variation: type of job, type of organization, type of applicant, etc. Moderator variables were found for some selection methods, but not sufficient to explain all the variation in validity. Work VALIDITY OF SELECTION METHODS 47 Table 2.5
interviews do create some adverse impact on females, whereas structured interviews do not. The research reviewed is nearly all North American, so one cannot safely assume similar results will be found in other countries, given that attitudes to gender vary widely. Are interviewers biased by race? Huffcutt et al.’s (2001) meta‐analysis shows that unstructured interviews do create some adverse impact on non‐white Americans, especially interviews that assess intellectual ability and experience.
1980s. The GATB data base includes mostly ‘ordinary’ jobs rather than higher‐level ones. Hunter and Hunter (1984) showed that an uncorrected average validity of 0.25 for GMA rose to 0.47 when corrected for unreliability and range restriction, confirming their re‐analysis of Ghiselli’s data. VGA also shows that validity of ability tests does not vary as much as Ghiselli thought his distributions showed. Schmidt and Hunter (2004) argue that validity of ability tests does not ‘really’ vary at all;
Goldberg’s bank, which contains analogues to many popular PQs, has another unusual feature: it is in the public domain, and can be used by everyone without restriction or payment, through his IPIP website (address at end of chapter). Users will have to generate their own normative data. RESEARCH AGENDA • Survey of which types of PQ items are seen as intrusive in selection. Key points In Chapter 7 you have learned the following: • Personality is a vaguer and more diffuse concept than ability,
conscientiousness is having principles and standards, and feeling that one must adhere to them, which could be difficult to distinguish by behavioural tests from conforming through fear or expedience. Ethics and acceptability The OSS tests probably pushed As further and harder than many employers would want to do today, for fear of complaints or damaging the organization’s reputation (problems that tend not to arise in wartime). However, if a job does involve a lot of stress and conflict, the