The cinderella predictor

Anne Keegan

Stefan Mol

Gabor Kismihok

The Cinderella Predictor: Taking job knowledge to the personnel selection ball

CriterionRelatedValidity

ContentRelatedValidity

ConstructRelatedValidity

ContentRelatedValidity

ConstructRelatedValidity

CriterionRelatedValidity

Job Knowledge

CriterionRelatedValidity

“…people who are more intelligent learn more job knowledge and learn it faster, the major determinant of job performance is not GMA but job knowledge” (Schmidt and Hunter, 2001).

GMA Job knowledge Job Performance

“Validity is nearly twice as high for job specific tests than for of the-shelf tests. In fact, the levels for locally developed tests appear to rival those of general ability tests” (Dye, Reck & McDaniel, 1993,p. 156).

ρ=.62, k =59, N=3965

Job Knowledge CriterionRelatedValidity

GMA Job knowledge Job Performance

“Content-related evidence of validity has traditionally involved … rational examination of the manner in which the performance domain is sampled by the predictor” (Binning & Barrett, 1989, p. 482).

Job Knowledge CriterionRelatedValidity

ContentRelatedValidity

Job knowledge: According to the APA publication manual which statistics do you need to report in a regression analysis table?

GMA: How smart do you need to be to publish an article?

Hunter (1986, p. 353) “knowledge is a much better measure of performance than [supervisor] ratings”.

Job Knowledge CriterionRelatedValidity

ContentRelatedValidity

Job Knowledge

“construct validation and theory development imply the same basic process….” “If it can be shown that a test measures a specific construct … that has been determined to be critical for job performance…, then inferences about job performance from test scores are by logical implication justified” (Binning and Barrett, 1989, p.487, p. 480

CriterionRelatedValidity

ContentRelatedValidity

ConstructRelatedValidity

Job Knowledge CriterionRelatedValidity

ContentRelatedValidity

ConstructRelatedValidity

Job Knowledge Specificity

1 2 43 65 87 9

Substantiation

Job Performance Specificity

124 36 58 79

Towards a research agendaUnequivocally Job Specific (albeit more portable for more related occupations)

Undeniably Job General

Job Specific – Job General

Job Knowledge

Job Performanc

eGMA

Education

Development of Job Knowledge Tests: Implications

• Focus on idiosyncratic job knowledge requirements– Requires use of inductive, interpretive research methods

• Implications for status and acceptance of qualitative research within the personnel psychology community– “Qualitative research often advances the field by

providing unique, memorable, socially important and theoretically meaningful contributions to scholarly discourse and organizational life” (Gephart and Rynes 2004, p.461)

Job Knowledge as Matchmaker

• Connecting:– HRM, eLearning & Knowledge Management

(foster collaboration)– Academia and industry (Science and practice)– Employees and workplaces (Enhanced Matching)– Policy and Education (Training) – Government and Labor Market (Needs analysis)

http://www.ontohr.eu

Technology is on our side

• Flexible systems for matching employees, organizations and educational institutions

• Process Management• Semantic applications

– Data structuring – Comparability

• Text/data mining– Automatic data collection and recognition

• Content management– eLearning methodologies for content representation (LO, LOR, OER)

• Adaptive testing

Why make the effort?

With representative job knowledge tests:• We no longer need to reject candidates on

stable dispositional characteristics• We can inform candidates as to where they can

obtain missing knowledge • We can inform educational institutions about

missing job knowledge of their graduates• We can gauge the knowledge of immigrants,

informal learners, and the elderly