Fairness&ethics Rsw3

Front 1

validiteitscoefficient

Back 1

de correlatie tussen scores op procedure x en een extern criterium y (in hoeverre de testscores gerelateerd zijn aan een psychologisch betekenisvollle karaktereigenschap/gedrag)

Front 2

adverse impact

Back 2

members of one group are selected at substantially greater rates than members of another group (.80 ratio)

Front 3

differential validity

Back 3

differences in calidity coefficients across group. computed validity coefficients are significant different for different groups

Front 4

differential prediction

Back 4

the best prediction equations and/or standard errors of estimate are significant different for different groups.
--> different intercepts & different slopes

Front 5

direct vs indirect discrimination

Back 5

direct: differential treatment based on a specific characteristic
indirect: any provision, criterion or practice which is neutral on its face but is not)

Front 6

positive validity

Back 6

people high on test perform high on job.
predictor: criterion relationship is the same for both subgroups and elliptical in shape

Front 7

zero validity

Back 7

predictor criterion relationship is the same for both subgroups but circular in shape

Front 8

valid predictor with adverse impact

Back 8

predictor-criterion relationship shows differences per subgroup and elliptical in shape

Front 9

selective ratio

Back 9

number given raise / number employees

Front 10

adverse impact

Back 10

minority sel. ratio/ majority sel. ratio

Front 11

predictor

Back 11

testscore (reject/accept)

Front 12

criterion

Back 12

performance rating (unsatisfactory/satisfactory)

Front 13

strategies reduce adverse impact

Back 13

1. improve recruiting for minorities
2. use g-test in combination with non g-predictors
3. use measures of specific, not general g-abilities
4. use diferent weighting for various criterion facets deload criterion of g
5. use alternate modes of presenting test stimuli (video based)
6.implement test-score banding vs strict top-down selection

Front 14

adverse impact justifyable when

Back 14

-wrong criteria
-justified by nature of job
-positive action; measures to prevent or compensate for existing inequalities

Front 15

Affirmative action

Back 15

policy that exceed equal employment opportunity by requiring to correct past discriminitory practices by increasing the numbers of minorities & women in specific positions

Front 16

reverse discrimination

Back 16

giving preference to members of protected classes to the extent that unprotected individuals believe they are suffering discrimination

Front 17

performance feedback preperation

Back 17

-communicate often with subordinates about perf.
-get training in performance app.
-judge yourself first before judging others
-encourage subordinates to prepare for the interview
-use priming info to help retrieve info from memory

Front 18

performance feedback during appraisal

Back 18

-warm up & encourage subordinate participation
-judge performance
-be specific, first + then -
-be an actie listener-empathize
-avoid destructive criticism & threats to employees ego
-set mutually agreeable & formal goals for future improvement

Front 19

performance feedback after appraisal

Back 19

-communicate often with subordinates about perf.
-periodically assess progress toward goals
-make orga. rewards contingent on performance

Front 20

organization responsibility

Back 20

context-specific organizational actions & policies that take into account stakeholders expectations and the triple bottom line of economic, social & environmental performance

Front 21

stakeholders

Back 21

groups & people who (can be) affect(ed) by orga. includes: owners, employees, surrounding community etc.)

Front 22

moral standards

Back 22

standards of conduct of give profession/group at given time

Front 23

ethics & morality

Back 23

behaviours about which society holds certain values

Front 24

ethical choice

Back 24

considered choice among alternative courses of action.
-interrests of all parties have been clarified & risks and gains evaluated

Front 25

confidentiality

Back 25

treatment of info provided with expectation that it will not be disclosed to others

Front 26

code of ethics

Back 26

a written set of standards of condct (ethical values) that governs relations with employees and the public. provides a basis for the orga & managers to evaluate their plans and actions

Front 27

Fair information practices

Back 27

-peridoically & systematially review HR recordkeeping practices
-articulate, communicate, implement fair info-practice policies

Front 28

ethical obligations to the people evaluated

Back 28

-accuracy, equality of opportunity- fairness
-securing privacy, guaranteeing confidentiality
-obtaining applicants & employees informed consent
-respect employees right to know
-imposing time limitations ons data
-minimizing erroneus acceptance/rejections
-treating employees with respect & consideration

Front 29

ethical obligations to the employees

Back 29

convey accurate expectations of evaliations procedures
-ensure high quality for HR decisions
-periodically reviw accuracy
-respect employers proprietary rights
-balance the vested interests of employer with law, code etc.
-high executive commitment of tester!
-corporate ethics programs

Front 30

differentiele predictie/predictie bias

Back 30

als een selectie-measure oneerlijke discriminerend is, en een bepaalde groep dus slechter presteert dan een andere groep op de measure maar hetzelfde presteert op de baan die de measure voorspelt (zelfde criterium, ander prediction)

Front 31

minimax strategie

Back 31

het minimaliseren v/h maximale verlies in utiliteit, in vergelijking met top-down selectie (verplaatsen van band na 1e selectie naar 1na hoogste score)

Front 32

predictor vs criterium referenced banding

Back 32

-criterion-refband. is niet altijd mogelijk
-bij crit. worden bredere banden gebruikt zodat de false negatives worden verlaagt, maar economische utiliteit verlaagd.
-crit. data vaak imperfect/beperkt
-worden 2 ipv 1 betrouwbaarheidscoefficient gebruikt bij crit.band (zowel predictor als criterium)