Lect 5 Flashcards
(29 cards)
what is a small, medium and large effect size
o r = .10 → Small effect size.
o r = .30 → Medium effect size.
o r = .50 → Large effect size.
what are cognitive test
Cognitive ability tests measure general mental ability (g), like reasoning, problem-solving, and learning capacity. They are good predictors of job and training performance, especially in complex jobs, but may lead to adverse impact across demographic groups
it is a selection procedure of high validity, it predicts more stringly the more complecxt the job is
it aso predicts organisational citizenship behavior
however it is not well liked
what are the problems with g
1 can lead to adverse impact
2. does not predict counterprocudtive behavior
3. can be improved over time
4. better at predicting max rther than typical performance
what are personality inventories and its ussues
Personality inventories measure traits like the Big Five (e.g., conscientiousness, extraversion) to predict job performance.
Conscientiousness = most consistent predictor of task
performance across jobs, most generalizable (also positive
relationship with OCB). there can be tpp much conscientiousness but the adverese impacts of contientiousnesss are worse in simple athan complex jobs
the main issue is distortion
what are work samples
Work samples are assessments where candidates perform tasks similar to those on the job (e.g., role-plays, in-basket tests, SJTs).
Pros:
High validity (~.33)
High face validity & acceptance
Less adverse impact than cognitive tests
Cons:
Can be costly and time-consuming
Need to match the sample to actual job tasks carefully
what to think about when diesigning a work sample
When designing a work sample, consider:
1. Group vs individual exercises – Does the job require teamwork or solo work?
2. Bandwidth – How broad is the range of skills you’re assessing?
3. Fidelity – How closely does the sample mirror real job tasks?
4. Necessary experience – Does the task assume prior job knowledge?
5. Task type – Is it cognitive, interpersonal, or physical?
6. Mode of delivery – Will it be paper-based, online, or in-person?
what are some types of work samples
a) Leaderless group discussions
b) In-basket test
c) Role-plays
d) Situational Judgment Tests (SJT)
e) Structured interviews based on CIT (from last week)
what are leaderless group discussions
Aim: To observe applicants in a group setting and
measure various aspects such as leadership,
communication, cooperation, persuasion, initiating
structure, consideration, conflict management etc.
Leaderless Group Discussions (LGDs) are group exercises where candidates discuss a problem without a designated leader.
Purpose: Assess interpersonal, leadership, and teamwork skills
Pros: Reveals natural behavior under pressure
Cons: Costly—requires trained assessors to observe and code behaviors
Common use: Often part of assessment centersRWS 5.
what is an in-basket test
An in-basket test (or in-tray exercise) simulates a job’s daily tasks—emails, memos, problems to solve.
Goal: Assess decision-making, prioritization, and problem-solving
What assessors look for: How well candidates handle info under time pressure
Tips: Beware of tasks that seem urgent but aren’t actually important
what are role plays
Role plays simulate critical interpersonal situations (e.g., handling a conflict or giving feedback).
Purpose: Assess communication, emotional control, and social skills
Focus: How candidates react under pressure and deal with difficult interactions
t or f
its better to use one predictor than to combine many
fasle - combinations of the top five predictors and the mean validity was high = .51
what are international assigments
International assignments involve sending employees to work in another country for their company.
Goal: Fill global roles, develop talent, or transfer knowledge
Challenge: Success depends more on adapting to the cultural context than technical skills
Key predictors: Personality, cultural adaptability, language skills, and international experience
when creating an international assigment what should you focus on and why
a) context
b) content
Focus on a) context.
Why?
Success in international assignments depends on how well a person adapts to the new environment (culture, customs, norms), not just their job-specific skills. The content (i.e., technical tasks) is often already known or assumed to be sufficient
how to make the decisions for selecetion
You can make selection decisions using data combination methods:
Multiple regression – Combines weighted scores from multiple predictors to estimate job success.
Multiple cutoff – Applicants must pass minimum scores on all predictors.
Multiple hurdle – Applicants are assessed in steps; only those passing move on.
Also, use utility models to evaluate the cost-effectiveness of your selection method:
Taylor-Russell, Naylor-Shine, Brogden-Cronbach-Gleser models estimate hiring success and value
what are the types of prediction models
Prediction models combine information from selection tools to forecast job performance.
Main types:
Multiple regression – Weights predictors based on how well they predict the criterion.
Multiple cutoff – Set minimum scores for each predictor; fail one = rejected.
Multiple hurdle – Predictors used in stages; only those who pass each stage move forward
what ia the multiple regression approach
The multiple regression approach combines several predictors (e.g., test scores, interview ratings) by weighting them based on their statistical relationship with job performance.
Goal: Maximize prediction accuracy
Use: All applicants are assessed on all predictors
Outcome: A total score is calculated to rank applicants
multiple cutoff approach
The multiple cutoff approach sets a minimum score (cutoff) for each predictor (e.g., test, interview).
Rule: Applicants must meet or exceed all cutoffs to be considered
Use: Screens out candidates weak in any critical area
Limitation: Strong performance in one area can’t compensate for weakness in another
what is the multiple hurdle approach
The multiple hurdle approach is a step-by-step selection method.
Sequentially eliminates candidates in multiple stages. It is cost-effective for intensive training or high-risk jobs
Process: Applicants must pass one predictor (hurdle) before moving to the next
Example: First a cognitive test → then an interview → then a work sample
Benefit: Saves time and resources by narrowing down candidates early
Limitation: Early rejection may overlook strong later-stage potential
what are the utility models and on what 4 factors do they depend on
utility models are always comparing the statuc quo to the new selection procedure
Utility (added benefit/value) of a predictor depends on 4
key factors:
1. The accuracy or validity of the predictor (criterion validity)
2. The selection ratio
3. The base rate of success
4. The costs and benefits of selection decisions (e.g.,
recruitment, selection, training etc.)
what is the selection ratio
Selection ratio (SR) - % of applicants from the population of applicants that are accepted
what is the base rate
Base rate (BR) à % of current employees who are considered effective using current/existing/old selection procedures
A company received 100 job applications for a new position. After the full selection process, they decided to hire 20 people. The company also reports that 40 out of the 100 applicants would be considered successful employees if they were all hired.
calculate Selection Ratio (SR) and Base Rate (BR)
Selection Ratio (SR) = 20 / 100 = 0.20 (or 20%)
Base Rate (BR) = 40 / 100 = 0.40 (or 40%)
what does it mean if selection and base rate rate are too high/ too low, what is the bets combo
if base rate is high that means that many of the employees are performing well, so it is difficult that a new selection procedure can improve on it
. A low base rate (only 20% succeed) suggests the current selection method is ineffective and new methods could greatly enhance hiring quality
- A high selection ratio (SR close to 1) means most applicants are hired, reducing selectivity and predictive accuracy. A low selection ratio (SR close to 0) allows for greater selectivity and better hiring decisions.
whta are the 3 utility models
The 3 utility models are:
Taylor-Russell – Shows improvement in hiring success.
Naylor-Shine – Estimates average performance difference between selected vs. unselected.
Brogden-Cronbach-Gleser – Calculates selection’s financial value (in € or $)
