• Journal of the Korean Data and Information Science Society
• /
• v.26 no.1
• /
• pp.141-150
• /
• 2015

Graphical paradigms for assessing the adequacy of models in logistic regression are discussed. The residual plot has been widely used as a graphical tool for evaluating the adequacy of the model. However, this approach works well only for linear models with constant variance, and the alternative approach, the marginal model plot, has its defects as well. We suggest a Chi-residual plot that overcomes the potential shortcomings of the marginal model plot.

• The Korean Journal of Applied Statistics
• /
• v.28 no.6
• /
• pp.1191-1208
• /
• 2015

Most statistical methods are dependent on the summary statistic. However, with graphical approaches, it is easier to identify the characteristics of the data and detect information that cannot be obtained by the summary statistic. We present various graphical methods to assess the adequacy of models in logistic regression that include checking log-density ratio, structural dimension, marginal model plot, chi-residual plot, and CERES plot. Through simulation data, we investigate and compare the results of graphical approaches under diverse conditions.

• Journal of Korean Academy of Nursing
• /
• v.29 no.6
• /
• pp.1192-1207
• /
• 1999

The purpose of this study was to clarify the dynamic relationships among risk factors of arteriosclerosis and to develop and examine a model which could explain this relationship clearly. Data were collected from medical records of 400 male clients who visited a university hospital located in Inchon for physical examinations, from May 1996 to December 1996. Data were analyzed using the LISREL (Linear Structural Relationship) 8 program. To test the fitness of the hypothesized model, chi-square, RMSR (root mean square residual), GFI (goodness of fit index), CN (critical number) and Q-plot were used. Most of the fitness measurements, except the chi-square showed that the hypothesized model complimented the real data. According to the results, there were trends that obesity and hyperlipidemia were prevalent in heavier smokers, higher alcohol intakers, and groups who excercised less. Also, hypertension was more prevalent in older age, higher alcohol intaker, and higher serum lipid level groups. In contrast to the hypothesis, alcohol intake did not significantly affect serum lipid levels. This might be due to the serum lipid measurements (total cholesterol and trigryceride) used in this study to estimate hyperlipidemia. The direct effect of smoking on hypertension was not significant. However, the total effect of smoking on the hypertension was significant since indirect effects of smoking on hypertension, such as obesity and hyperlipidemia, were significant. The total effect of obesity on hypertension was significant since the indirect effect of obesity on hypertension via hyperlipidemia was significant, although the direct effect of smoking on hypertension was not significant. The degree of explaining hyperlipidemia with smoking, exercise, and obesity was high (60%), however, the degree of explaining obesity with age, smoking, alcohol intake, and exercise was very low (7%). On the basis of these results, high risk factors of arteriosclerosis such as hypertension, hyperlipidemia, or obesity are either directly or indirectly correlated each other. Therefore, it is difficult to predict outcomes for increasing or decreasing the risk factors by simply modulating a factor. Smoking, alcohol, and exercise both directly and indirectly affected major risk factors of arteriosclerosis. Therefore, correcting these variables is required to decrease risk factors. Finally, the relationship among other risk factors which have been known to be related with arteriosclerosis (diet, stress or hereditary) should be clarified in further studies.

• Research in Community and Public Health Nursing
• /
• v.12 no.2
• /
• pp.450-458
• /
• 2001

Purpose: The purpose of this study was to establish a structural equation model on social roles and health for married working mothers. to determine the effects of social roles on Korean women's health and to explore the mediating role of self-esteem in the relationship between social roles qualities and health. Method Data were collected from 323 employed women with partner and children. lived in Seoul and near Seoul. Response rate was 62.3%. The instruments for measurements were Role Quality Scale developed by Park et al. (1999). Rosenberg's Self-Esteem Scale. and 31 items from SF-36 developed by Ware & Sherboune(l992). Results: The effect of marital role quality on self-esteem and the effect of parental role quality on health were not significant. Modification model fitted with the collected data very well. as evidenced by the small chi-square(0.58), the very high goodness-of-fit(GFI = 1.00), and adjusted goodness-of-fit (AGFI = 0.99), and very small root mean square residual(RMSR=0.0072), and the slope of Q-plot is over 1. 41% of the variance in self-esteem and 21% of the variance in health were accounted for by these variables. Conclusion: Further research concerned with the mediating effects of self-esteem in the role and health relationship should be covered the issue of various role combinations. And it is necessary to examine the influence of subfactor of quality of social roles on health.