Validity and Reliability of Self Report Measures of Physical Activity: An Information-Processing Perspective

Abstract The published data on the reliability of self report measures of physical activity most commonly used in the cardiovascular disease epidemiology literature revealed high test-retest reliability coefficients for two of the measures, and modest to nonexistent intertest correlations. Validity coefficients were low to modest. The published data on the accuracy of the self report measures, however, revealed memory decay, memory of rare events alone, and lack of motivation in memory recall. An information processing model, composed of encoding, storage, and retrieval processes is proposed to understand the memory of physical activity, and to identify necessary skills for accurate self report identified at each step in the process. Questions requiring further research to specify this model and, in turn, improve accuracy of recall, are raised.