Mental Health Among Elderly Native Americans (Psychology Revivals)

James Narduzzi, James L. Narduzzi

• Publisher: Routledge
• Year: 2015
• Print ISBN: 9781138861084
• eBook ISBN: 9781317506027
• Topic: Psychology
• Subtopic: Mental Health in Psychology
• Index: Psychology

In the 1990s providing mental health services to the elderly and particularly to elderly Native Americans had been an issue of some concern for the last several decades. Despite this, many public decisions made at the time were based on inadequate data. Due to this lack of data, there had been little research devoted to determining the factors associated with mental health among elderly Native Americans. Instead, the growing body of mental health research had "been based on limited samples, primarily of middle-majority Anglos." Originally published in 1994, the purpose of this research was to utilize existing data to close the gap in our understanding of mental health among elderly Native Americans.

V
Results: The Stress Process Model Applied to Elderly Native Americans

In this chapter, the stress process model will be applied to the various samples and the results will be reported. Two prior steps remain, however, before the model can be dissected in earnest. The first is to determine if either sex or place of residence have effects on mental health independent of the various stressors and mediators, thereby confounding the regression results. Thus, the chapter commences with a brief discussion of the results of the analysis of variance and covariance conducted on sex and place of residence. The second step is to observe the statistical trends and characteristics relative to the relationships between the exogenous constructs. Specifically, correlation and partial correlation analysis will be performed to determine the precise relationships between these variables and the results will be summarized and reported. Then, attention will shift to the regression results. The full model will be applied to 5 samples: the entire population, males, females, reservation-dwellers, and urban dwellers, and the results reported. Then, the parsimonious model for each sample will be examined. Next, tests to determine the interaction effect of the mediators will be performed and discussed. The chapter will end with a summary of the major results of the research.

Analysis of Variance and Covariance

As noted previously, the stress process model will be examined taking both sex and place of residence into account. Before that analysis can be done, however, it is important to determine if sex and/or place of residence have an effect on mental health that is independent of the exogenous variables. In other words, are there significant differences to begin with between the sexes and/or place of residence on mental health? If not, then it can be safely argued that the observed regression results are not confounded by either of these factors.

Analysis of variance and analysis of covariance were employed to test for these differences. Appendix G provides a brief discussion of both each technique and the respective results when applied to this data. For now, it is important to note that both types of analysis indicate that there are no significant differences between sex and place of residence with respect to mental health. In effect, the regression results can be reported knowing that neither sex nor residence will confound the results of that analysis.

Correlation and Partial Correlation Analysis

The last step before turning to the regression results relates to the observation of statistical trends and characteristics of the independent variables to determine if there are interdependencies among these constructs. Correlation and partial correlation analysis was thus performed with that end in mind. Appendix H presents a discussion of cach technique along with the complete results for each technique when applied to each sample. For now, it is instructive simply to review and summarize the results of that analysis.

Three sets of interdependencies are expected, the first relating to the operationalization of socioeconomic status and the second two relating to the hypothesized interaction between mediators and stressors. Respectively, these interdependencies are as follows:

l. SES —	between INC and ED;
2. COP —	and INC
	and ED
	and PH;
3. SS —	and INC
	and ED
	and PH.

The table below presents the relevant results from the correlation and partial correlation analysis that was conducted. Correlation analysis is a statistical technique that can be employed to summarize the strength of the relationship between a pair of variables. In column one, the product-moment correlations between the exogenous variables for each sample with r greater than .20 (either positive or negative) are reported. In column two, the partial correlation results are reported. In partial correlation analysis, the variation in the dependent variable which is attributable to other independent variables is controlled, thereby isolating the effects of the interdependent variables. The procedure involves allowing the control variable to explain all of the variation in the dependent variable before the effect of the test variable is considered. Specifically, column two in the table shows the reduction in the correlations between the independent variables and mental health greater than .05 when the controls (i.e. the other independent variables) are added to the equation. The independent variable is reported first, followed by the control variable, and the differ...

Table of contents

• Cover
• Title
• Copyright
• Original Title
• Original Copyright
• Dedication
• Contents
• PREFACE
• ACKNOWLEDGEMENTS
• LIST OF TABLES
• I INTRODUCTION
• II BACKGROUND AND LITERATURE REVIEW
• III THE RESEARCH MODEL
• IV METHODS
• V RESULTS: THE STRESS PROCESS MODEL APPLIED TO ELDERLY NATIVE AMERICANS
• VI ANALYSIS AND IMPLICATIONS
• VII SUMMARY AND CONCLUSIONS
• APPENDICES
• SELECTED BIBLIOGRAPHY
• INDEX