A minimum data set and empirical model for population health status assessment.

BACKGROUND: The need to assess the health status of American communities in a comprehensive and systematic manner has been widely acknowledged. This study attempts to empirically derive a minimum core data set of indicators, in order to produce a uniform parsimonious model for population health status monitoring. METHODS: Five years of secondary data (1992-1996) for 113 indicators of community health for each of Florida's 67 counties were organized into 11 conceptual groups. Principal component analysis with orthogonal rotation was conducted separately on each group of indicators for each year. The component scores were converted to standard scores to further study the relationships among the conceptual groups measuring community health. A causal model was hypothesized and tested using ordinary least-squares path analysis. RESULTS: Nineteen principal components composed of 78 indicators were identified. The model demonstrated a large difference in the ability to explain variance in adult mortality (56%) compared with variance associated with adverse birth outcomes (13%). Both demographic and socioenvironmental factors have a direct effect on adult mortality. Socioeconomic factors, on the other hand, influence adult mortality indirectly through adequacy of primary care and other available resources. CONCLUSIONS: Minimum core data sets of indicators drawn from extant databases can be used to uniformly describe and explain variation in adult mortality. This research suggests caution in regard to the creation of integrated indices that combine mortality, morbidity, and other concepts such as quality of life into a single measure of community health. Further validation research employing a national sample of counties is recommended.

Improving the peer interactions of students with emotional and behavioral disorders through self-evaluation procedures: a component analysis and group application.

We conducted two experiments examining the effects of a self-evaluation package on the peer interactions of students described as emotionally or behaviorally disordered. Experiment 1 assessed the additive effects of various components of a self-evaluation package on the frequency of inappropriate and appropriate peer interactions. The components assessed were rewards alone, rewards plus discussion, and self-evaluation plus rewards. Results showed limited effectiveness when rewards alone and rewards plus discussion were implemented. However, substantial improvements in peer interactions were observed when the self-evaluation component was added. Experiment 2 examined the efficacy and feasibility of the procedures when implemented in a group setting. Students in three classrooms served as participants. Direct observation data collected for 8 of the participants showed the procedures to be effective in improving peer interactions when implemented in a group context.

Relationships between suicidal behavior and personality types.

This study examined the hypotheses that introversion in general and the introverted-intuitive-feeling-perceiving (INFP) type as measured by the Meyers-Briggs Type Indicator are related to suicidal ideation and behaviors. Analyses of data from 330 university students found four interactive suicidal behavior risk models: introverted-perceiving (IP) (females); introverted-sensing-feeling (ISF) (males); introverted-intuitive-perceiving (INP) (males); extraverted-intuitive-judging (ENJ) (males). Study findings were interpreted in terms of implications for risk assessment and intervention.

A Study of Multivariate Permutation Tests Which May Replace Hotelling's T2 Test in Prescribed Circumstances.

Multivariate permutation tests are described and studied which may be profitably substituted for Hotelling's one-sample P test in situations commonly arising in behavioral science research. These tests (a) may be computed even when the number of variables exceeds the number of subjects, (b) are distribution-free, (c) may be tailored for sensitivity to specific treatment alternatives, and (d) provide one-sided as well as two-sided tests of hypotheses. Power comparisons were made between the permutation tests and Hotelling's T(2) test under a variety of treatment effect model, correlation structure and number of variables combinations. Results show that the permutation tests have significant power advantages over the T(2) in a variety of circumstances, but may have considerably less power in others.