Assessment Methods for Selection & Placement

Assessment Methods for Hiring

• The best predictor for future performance is past performance

  • ideally wed hire based on ^

  • but we rarely have that info

• instead, we measure KSAOs as ‘predictors of criteria’

Psychological Tests

• test: a standardized series of problems or questions that assess KSAOs

  • 1. ability

  • 2. psychomotor ability

  • 3. personality

  • 4. emotional intelligence

  • 5. integrity

  • 6. interests

Testing Formats

• Speed vs Power

  • speed tests have strict time limits (e.g. typing tests, math attack)

  • Power tests, time is not a factor (e.g. first exam)

• Individual vs Group

  • Individual tests require 1-1 test administration (e.g. driving tests)

  • groups test can be taken by multiple people at once (e.g. our fist exam, SAT)

• Paper ‘n’ Pencil Tests

  • test takers respond to quesitons on paper (or computer screen)

    • e.g. fill-in-blank, multiple choice

• Performance Tests

  • involves manipulating an object/equipment to test applied skills

    • e.g. flight simulators, computer programming

Key Issues to consider when using tests for hiring

• Characteristics of good tests revisited:

  • how well does it predict job performance: criterion validity '

  • is the test free from random error: reliability

  • will the benefits of testing exceed the cost”: practicality

  • will the test disadvantage anyone: fairness

  • does the test effectively capture individual differences: sensitivity

Combining multiple tests

• incremental validity: the additional variance explained in an outcome by adding a predictor

  • ex) interview scores explain roughly 18% of total variance in job performance

  • that leaves around 82% unaccounted for

The Predictive validity of different assessment methods for job performance

• results shown here here based on many meta-analyses

• validity estimate = average correlation with job performance

• although this is a very informative table, it does not show differences across jobs, organizations, contexts, etc.

General Cognitive Ability Tests

• a common, relatively easy-to-adminster selection measure

• cognitive ability is important for most jobs, but is a stronger predictor of performance in more complex jobs

  • began with Army Alpha and Beta (remember the WWI history)

  • ex): WAIS, Wonderlic, Ravens Progressive Matricies

Issues with General Cognitive Ability Tests

• group differences in scores across racial and socio-economic backgrounds

  • racial differences smaller in recent years, but still exist

• test scores related to years of formal schooling and test prep

  • test-taking wiseness

On measuring ‘innate’ ability

• you cant measure innate intelligence any more than you can measure the innate height of a stalk of corn

• cognitive test scores are products of genes + enviornment

Specific cognitive ability tests

• tests spedific abilities that are required on the job (vs. general capacity to learn)

  • more similar to job knowledge

  • r= .30 to .40 with performance

    • ex)

      • mechanical (bennet mechanical test)

      • spatial (Space relations test)

      • clerical (minnesota clerical test)

Psychomotor Ability Tests

• assess speed and accuracy of mtor and sensory coordination

  • movement, vision, hearing, etc

  • e.g. Purdue Pegboard

• r= .30 to .40 with performance

  • but only in jobs that require these abilities = job analysis