You are on page 1of 3

13.

Decision-making

In personnel selection, decisions are made about individuals. Since decision-makers cannot know
in advance with absolute certainty the outcomes of any assignment, outcomes must predicted on
the basis of available information. Two-step procedure:
1). Measurement (collecting data using tests that are relevant to job performance);
2). Prediction (combining these data to enable the decision-maker to minimize predictive error in
forecasting job performance).
Measurement and prediction are simply technical components of a system designed to make
decisions about the assignment of individuals to jobs.
Linear models often are used to develop predictions.

Unit-weighting: transform all predictors in z-scores à sum them to one factor à compare to
others and make decision.
Advances of unit-weighting schemes:
1). Not estimated from the data and don’t “consume” degrees of freedom;
2). “Estimated” without error (no standard errors);
3). They cannot reverse the “true” relative weights of the variables.

Suppressor variables: can affect a given predictor-criterion relationship, even though such


variables bear little or no direct relationship to the criterion itself. (figure 14-2, p. 319).

Data combination strategies


Mechanical (statistical): if individuals are assessed on some instruments, if they are assigned
scores based on that assessment, and if the scores subsequently are correlated with criterion
measure (tests, biographical data forms, structured interviews).
Judgmental (clinical): if a set of scores or impressions must be combined subjectively in order
to forecast criterion status (assessment interviews and observations of behavior).

Strategies of Data Collection and Data Combination

  Mode of data combined

Mode of data collection Judgmental (clinical) Mechanic (statistical)

Judgmental (clinical) Pure clinical Behavior threat rating

Pure statistical
Mechanical (statistical) Profile
interpretation (best?)

Clinical composite (most Mechanical


Both
common) composite (best?)
Alternative prediction models

 Multiple-regression approach: performance on certain predictors can compensate


weaknesses on other predictors. Linear models (y = a + b1x1 + .. bixi). Not possible?
 Multiple-cutoff approach: when one predictor cannot compensate for another. Cutoff
score dependent on need of organization. Multiple regression cutoff à represents the
minimum score necessary to qualify for selection.
Methods for determining cutoff score for single predictor:
- Angoff method: based on how incompetent people score. Experts judges rate each item
in terms of the probability that a minimally competent person would answer the item
correctly. The probabilities are then averaged for each item across judges to yield item
cutoff scores, which are summed to yield a test cutoff score.
- Expectancy charts: are used to illustrate visually the impact of cutoff scores on future
hiring decisions; depict the likelihood of successful criterion performance for any given
level of predictor scores.
score > 85 (best 20%) à 90% chance of being successful
score > 50 (best 40%) à 70% chance of being successful
score > 25 (best 60%) à 50% chance of being successful
 Multiple-hurdle approach: combine multiple regression and multiple cut-off.
Sequential, decision strategy.
Two-stage, sequential selection procedure (figure 14-5, p. 327):
Assessment of ability occur in stage 1, because this information is inexpensive to obtain.
Applicants who reach the cutoff score on the ability measure progress to stage 2; the
others are rejected. Final selection decisions are based on stage 1 and stage 2 information.
Stage 2 information would normally be more expensive than ability measures to obtain,
but the information is obtained from a smaller group, thereby reducing the cost relative to
obtaining stage 2 information from all applicants.

Above are the classical validity approaches to selection decisions.


Decision-theory approach = extension/ addition of classical approach. Not only validity, but
also utility decision for individual and organization à overall worth of selection instrument.
Utility depends not only on the validity of a selection measure but also on two other parameters:

 Selection Ratio (SR): the ratio of the number of available job openings to the total
number of available applicants (figure 14-6, p. 328). As the SR approaches 1.0 (all
applicants must be selected), it becomes high or unfavorable from the organizations’
perspective. Conversely, as the SR approaches zero, it becomes low or favorable, and
therefore, the organization can afford to be selective.
 Base Rate (BR): # successful applicants/ # total applicants. The more skills needed for
the job, the lower the base rate.
 Both affect success ratio: successful hires/ total hires.

By focusing only on selection, the classical validity approach ignore the implications of selection
decisions for the rest of the HR system. Any selection procedure must be evaluated in terms of
its total benefits to the organization. The main advantage of the decision-theory approach is that
it addresses the SR and the BR parameters (if erroneous acceptance are a major concern, then the
predictor cut off score may be raised.
 

 Taylor-Russell model

Strive for a high success-ratio. Percentage of successful persons in selected group as high as
possible (validity, SR and BR). CA : FA.

 Naylor-Shine model
Not: ‘good or not good’, but: criterion score of selected group as high as possible.
A linear relationship between validity and utility. The higher the validity, the greater the
increase in average criterion score for the selected group over that observed for the total
group. Less information is required in order to use this utility model.
 Brogden-Cronbach-Gleser model
Added value of selected group as high as possible (in money). Calculates the increase in
financial pay-off resulting from the use of a selection procedure instead of selecting
randomly. The only assumption required to use this model is that the relationship
between test scores and job performance is linear; the higher the test score, the higher the
job performance and vice versa.

You might also like