Week 3 - Selection and Assessment Flashcards
(21 cards)
Importance of context
Every industry is different
Can influence JA - competition in workplace, resistance
Open Systems View (Wilson)
Job analysis = context > receptivity > process > outcomes
Just-in-time approach
Identify JA need > determine job specific requirements > evaluate available JA methods > select method > conduct > analyse
Basically only doing JA when required
Method ‘fit’
Must consider which JA method is worthwhile (i.e. critical incident technique is best but costly)
Recruitment
Finding candidates for vacant positions (make them aware, encourage them to apply)
Selection
Matching person to the job
Need information - job requirements, candidate attributes, candidate needs, what job provides
Criterion-related validity
Correlating test score with performance measure
Temporal element problematic
Can try for concurrent validity (test + sample task) but this is also problematic
Criterion-related ‘cut-scores’
Specified point below which candidates are rejected (point depends on context)
Maximise hits/minimise misses
Job performance coefficients
Very beneficial (>.35) - cognitive ability, structured interviews, job knowledge tests, assessment centres
Likely to be useful (.21-.35) - conscientiousness, references
Less good - unstructured interviews, graphology
Incremental variance
How much additional performance is explained by other tests
Which fewest tests maximise predictive information?
Criticisms of psychological testing for selection
Homogenised workforce and maintenance of conformity (‘company men’)
Schneider’s ASA model - attraction, selection, attrition (similar people attracted, selected and stay in job)
Ethical/legal considerations
ILO Convention 111, Article 1 - discrimination that nullifies equality of opportunity (cannot be based on inherent job requirements though)
Areas of discrimination - age, sex, marital status, religion, disability, race, union membership
Individual differences
Differences in intelligence (maximal performance) and personality (typical performance) particularly important
These are stable and pervasive
Assumptions of framework - 1. variety of stable attributes, 2. people differ in these attributes, 3. differences remain after training, 4. different jobs require different attributes, 5. attributes can be measured
Interviews
Good - assesses useful information, can promote organisation
Bad - assumes right candidate will be obvious, assumed necessary
Structured, behavioural is best.
Bot interviewing also now exists (promising)
Resumes
Good - cheap, no explanation needed, easy to evaluate
Bad - no standard format, relevant info hidden, style influence, distracting info
General mental ability (‘g’)
Reason, plan, problem solve, abstract comprehension, learn
Most jobs require information manipulation, so ‘g’ is important
Greater job complexity = greater predictive value of GMA
Individual tests (SB, WAIS) and group tests
Good - stable over time, easy to test, predictive validity
Bad - specific abilities not measured, may reinforce cultural and other differences
Aptitude and ability tests
Aptitude - potential to do something in future (grad programs or low skill jobs)
Ability - things you can already do (jobs with technical/observable capacities), e.g. work sample test
Observing behaviour
Trend towards assessment of actual behaviour (games/simulations)
Assessment centres
Intensive process w/multiple methods (interviews, psychometrics, work samples, observations)
Expensive and time-consuming
Especially popular for managers
Effective for job performance (but low construct validity)
Personality tests
How is it measured? - observation, interview, psychometric tests
Why it matters? - predicts job performance, has incremental value
Must be interpreted in context of appropriate norms however
Assessing special capabilities
Expert Intensive Skills Evaluation (EXPERTise)
RAPID Framwork (Wiggins) – cue recognition, association, prioritisation, identification, discrimination (can be tailored to different domains, assesses cue utilisation)
Assessing special capabilities useful when not all job parameters are known (emerging roles)
