Jae-On Kim

The Treatment of Missing Data in Multivariate Analysis

Procedures for treating missing data in the statistical analysis of survey data are reviewed. The main topics covered are: (1) how to assess the nature of missing data especially with regard to randomness, (2) a comparison of listwise and pairwise deletion, and (3) methods for using maximum information to estimate (a) parameters or (b) missing values.

Multivariate Analysis of Ordinal Variables

This article examines the assumptions underlying two multivariate strategies commonly used in analyzing ordinal data. Both strategies employ as a descriptive tool the ordinary multiple regression algorithms; the crucial difference between the two is that the first, ordinal strategy, uses the matrix of Kendalls s as the building block of multivariate analysis, while the second, parametric strategy, uses the matrix of Pearsons s. These two strategies are evaluated and constrasted in terms of their usefulness in answering basic research questions that arise in multivariate analysis. One overriding conclusion is that, contrary to the claims of its proponents, the ordinal strategy is no better than the parametric strategy at meeting some of the basic requirements of multivariate analysis. It is argued that parametric strategy, when accompanied by careful evaluation of the validity of the implict quantification of ordinal variables, is more amenable to one of the goals of scientific research: successive approximation and refinement.

Standardization in Causal Analysis

In comparative study, it is argued that (1) the standardization of variables and scales should be separated from the habitual use of standardized coefficients; (2) the use of standardized coefficients implies standardizing every variable using group specific standards, and, therefore, it is not appropriate even if some variables have group specific metrics or some variables do not possess commonly accepted metrics; and (3) the explicit standardization of some or all variables can be fruitfully combined with the use of unstandardized coefficients.

Standardized and Unstandardized Coefficients in Causal Analysis

The inadvisability of using standardized regression coefficients for the purpose of comparing causal relationships across populations or establishing causal laws is addressed by explicating, both abstractly and by example, the relationship between the underlying causal structure and the variance-covariance structure of a system of variables.

PRU Measures of Association for Contingency Table Analysis

This paper introduces new predictive measures of association that may be used for contingency table analysis much as the simple, partial, and multiple correlation coefficients are used for regression analysis. The new measures are based on the combination of (1) the logic of the PRE (proportional reduction in error) measures of association, (2) the use of uncertainty (or maximum-likelihood-ϰ2) as indirect measures of errors in prediction, and (3) analysis strategy of log-linear models for contingency tables.

AN APPROACH TO SENSITIVITY ANALYSIS IN SOCIOLOGICAL RESEARCH

Sociologists, in particular, and social scientists, in general, are plagued by both weak theory and weak data. When the existing theory is insufficient to make all the relevant parameters of a statistical model identified, we often introduce dubious assumptions to make the model identifiable. In a similar way, when the variables under examination do not meet the measurement levels required by the statistical model, we sometimes pretend as if they do. As a means of evaluating the extent to which the parameter estimates are sensitive to such potential errors in model specification and data, a new approach to sensitivity analysis is proposed. The sensitivity index, which is the key to the new approach, has been widely used in different disciplines under a variety of different names.

Salvaging national pride: The 2010 taekwondo controversy and Taiwan’s quest for global recognition

In this article, we analyze Taiwan’s grassroots reactions to the disqualification of taekwondo icon Yang Shu-chun in the 2010 Guangzhou Asian Games, in order to examine how a technical dispute induced political and popular campaigns that variously blamed the governing party, the People’s Republic of China, and South Korea for inflicting shame on both the athlete and on Taiwan. Our research combines a historiography of Taiwan’s post-WWII experience, analysis of the nature of the Internet, and an assay of Taiwan’s three major newspapers. We find that the Yang incident became a symbolic vehicle for expressing the feeling of the nation’s citizens that Taiwan is trapped in international politics, and for restoring their national pride by transforming Yang into a virtuous heroine. We propose an attention to local, contingent narratives on sports, through which even regional games or failures serve to reinforce national identity.

Paths to Advancement in Japanese Bureaucracy

Based on biographical data, this study analyzes the factors affecting advancement in Japans higher civil service. Through path analysis, it evaluates in a systematic way not only the relative importance of three widely acknowledged factors of advancement—university background, field of study, and higher civil service examinations—but also the manner in which they interact with one another. It also assesses the impact of a civil servants “track record” on advancement. Finally, after examining the seniority principle in the context of Japanese cultural norms and the findings pertaining to merit-oriented variables, the study suggests that the seniority principle does not produce stagnation but rather facilitates mobility at the highest levels of Japanese bureaucracy.

Prognostic factors predicting control of chronic asthma symptoms in children receiving prophylactic bronchodilator therapy.

The factors predicting the control of asthma symptoms were studied in 312 consecutive children with chronic asthma who were receiving chronic prophylactic bronchodilator therapy. Asthma symptoms were controlled in 232 (74.4%) patients but not in 80 (25.6%) patients after 18 months of therapy. The cumulative percentage of patients in whom asthma symptoms were controlled increased with the duration of therapy. During the first 6 months of therapy, children who had a history of exacerbation of symptoms by fear stimulus or by dust had respectively an 83.3% (p = 0.0001) and a 35.4% (p = 0.003) lower rate of control of asthma symptoms than children without such a history. Similarly, children with a history of worsening asthma symptoms following ingestion of cold foods or drinks had a 71.4% (p = 0.002) lower rate of control of asthma symptoms after 6-12 months of therapy than children without such a history. Certain factors are important for predicting control of asthma symptoms at different times following the initiation of therapy and may relate to the lability of the airways to these stimuli.

Participation and Political Equality: A Seven-Nation Comparison

In this survey of political participation in seven nations-Nigeria, Austria, Japan, India, the Netherlands, Yugoslavia, and the United States-the authors examine the relationship between social, economic, and educational factors and political participation.