Assessment of type A behavior in children: a comparison of two instruments

The agreement of classifications yielded from two instruments used to assess children's Type A-Type B behavior, the Matthews Youth Test for Health (MYTH) and Hunter-Wolf (HWolf), was evaluated with a sample of rural children from the southern United States. Fifth grade children (N = 276) served as subjects. MYTH and HWolf scores were found to be only weakly correlated and the agreement of Type A-Type B classifications occurred at a rate only slightly above chance. To assess the psychometric properties of the instruments, both were subjected to factor analysis and reliability/internal consistency estimates were obtained and compared with previous results. A factor structure remarkably similar to that of previous reports was found for the MYTH but not the HWolf. The results closely parallel those of another recent report and provide further support for the recommendation that these instruments should not be considered interchangeable measures of Type A behavior and that when multiple measures cannot be employed for research, the investigator should use the MYTH. Caution is indicated in interpreting scores of either measure, however, since neither has yet been shown to relate to later development of coronary heart disease.