Prediction with single event versus aggregate data

• Published in: Organizational behavior and human decision processes Vol. 41; no. 2; pp. 196 - 210
• Main Author: Sniezek, Janet A
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Inc 01.04.1988; Elsevier; Elsevier Science Publishing Company, Inc
• Series: Organizational Behavior and Human Decision Processes
• Subjects: Accuracy; Behavior. Attitude; Behavioral decision theory; Biological and medical sciences; Comparative analysis; Effects; Experiments; Fundamental and applied biological sciences. Psychology; Information systems; Judgments; Predictions; Psychology. Psychoanalysis. Psychiatry; Psychology. Psychophysiology; Social psychology; Studies; Variance analysis; Human; Probability; Statistical estimation; Decision making; Prediction; Judgment
• ISSN: 0749-5978; 1095-9920
• DOI: 10.1016/0749-5978(88)90026-X

The impact of statistical information on predictive judgment was studied in a cue probability learning (CPL) task. Two kinds of aggregate information about criterion events were used: the conditional mean ( Z M ) and the conditional interquartile range ( Z R ). The single event information was the exact criterion value for one randomly selected past case ( Z O ). Results showed that Z M and Z R increased prediction consistency and accuracy and reduced bias while Z O led to more appropriate cue weighting but lower consistency and accuracy. When Z M , Z R , and Z O were all provided, the unique benefits of single-event and aggregate data were combined. This was true even for subjects without any prior task experience. When able to select only one of Z M , Z R , and Z O , judges most often chose aggregate information, particularly Z R . However, the statistical information was underutilized. Recommendations for aiding predictive judgment are made.