Reliability and validity of single-item self-reports: with special relevance to college students' alcohol use, religiosity, study, and social life
Related Experiment Videos
Stephen J Dollinger1, Danielle Malmquist
1Department of Psychology, Southern Illinois University, Carbondale 62901, USA. dollngr@siu.edu
Abstract:
The authors tested the assumption that single-item measures have unacceptably low reliability and validity. On 2 occasions 11 weeks apart, college students reported on the frequency and quantity of alcohol consumption, 2 religious behaviors, time of study and of socializing (focal items), and other qualities and characteristics. Most test-retest reliabilities were good to excellent; objective facts were more reliable than subjective evaluations; and target items had good validity when correlated with 2-week nightly log records of corresponding behaviors in a multimethod multitrait matrix. The exception was self-reported study, with relatively low reliability and validity, suggesting the non-trait-like quality of this behavior. Single-item measures may be better than commonly thought.
Related Concept Videos
Reliability and Validity
Self-Report Tests of Personality
Self-Presentation: Self-Monitoring and Self-Handicapping
Surveys
Strategies of Self-Presentation III: Self-Monitoring
Self-Discrepancy Theory