The autonomy‐validity dilemma in mechanical prediction procedures
The quest for a compromise
Bibliographic Data
A robust finding in psychological research is that combining information with a mechanical rule results in more valid predictions than combining information holistically in the mind. Nevertheless, information is typically combined holistically in practice, resulting in suboptimal predictions and decisions. Earlier research showed that decision makers are more likely to use mechanical prediction procedures when they retain autonomy in the decision‐making process. However, it remains largely unknown how different autonomy‐enhancing features affect predictive validity. Therefore, in two preregistered studies (total N = 342), we investigated if and how prediction procedures can be designed such that they satisfy decision makers' autonomy needs and acceptance without reducing predictive validity. Based on archival application data from a university admission procedure, participants predicted applicants' first‐year grade point average and chance of dropout. The results of Bayesian analyses showed that participants preferred prediction procedures in which they retained autonomy by choosing consistent predictor weights of a mechanical rule or by holistically adjusting the predictions of an optimal regression model. In general, these prediction procedures resulted in slightly higher predictive validity compared with fully holistic prediction. Providing participants with predictor validity information slightly increased predictive validity when participants could choose predictor weights but not when making holistic predictions or adjusting optimal model predictions. Giving decision makers a role in designing mechanical rules through choosing weights based on explicit predictive validity information could help promote the implementation and validity of mechanical prediction in practice
Affect (linguistics · Autonomy · Compromise · Construct validity · Criterion validity · Dilemma · External validity · Incremental validity · Machine learning · Predictive modelling · Predictive validity · Psychometrics · Behavioral Health and Interventions · Clinical Psychology · Computer Science · Decision-Making and Behavioral Economics · Mathematics · Neural and Behavioral Psychology Studies · Psychology · Social Psychology
A Theory of Cognitive Dissonance
Moving to a World Beyond “ p < 0.05”
Overcoming Algorithm Aversion
A practical solution to the pervasive problems ofp values
The accuracy of recidivism risk assessments for sexual offenders
The Meaningfulness of Effect Sizes in Psychological Research
General Evaluability Theory
Bayesian Assessment of Null Values Via Parameter Estimation and Model Comparison
Why We (Usually) Don't Have to Worry About Multiple Comparisons
The Bayesian New Statistics
RWA Web
Clinical versus mechanical prediction
When eliminating bias isn’t fair
The Time Has Come
Need Satisfaction, Motivation, and Well-Being in the Work Organizations of a Former Eastern Bloc Country
The New Statistics
In defense of clinical judgment … and mechanical prediction
A systematic review of algorithm aversion in augmented decision making
The earth is round (p < .05)
The effects of predictive ability information, locus of control, and decision maker involvement on decision aid reliance
The relative influence of advice from human experts and statistical methods on forecast adjustments
The recalcitrance of overconfidence and its contribution to decision aid neglect
What People Want From Their Professionals
The Meta-Analysis of Clinical Judgment Project
The "What" and "Why" of Goal Pursuits
Distinction Bias
Comparative efficiency of informal (subjective, impressionistic) and formal (mechanical, algorithmic) prediction procedures
Linear models in decision making
Measurement and prediction, clinical and statistical
Effects of cognitive feedback on performance
Determinants of linear judgment
Resistance to Medical Artificial Intelligence
A Contribution to the Study of Actuarial and Individual Methods of Prediction
| Unique citing works | 1 |
|---|---|
| Citations per year | 0,33 |
| Citation span | 2023 - 2023 (1) |
| Citation velocity | historical |
| Highly cited | No |
| Citation types | Neutral: 1 |