Jacques A. Hagenaars

Loglinear models with latent variables

Introduction The Loglinear Model The Latent Class Model Loglinear Modeling with Latent Variables Internalizing External Variables Causal Models with Latent Variables A Modified LISREL Approach Latent Variable Models for Longitudinal Data Problems and New Developments

Categorical longitudinal data : log-linear panel, trend, and cohort analysis

The Analysis of Social Change Log-Linear Analysis of Categorical Data Latent Class Analysis and Log-Linear Models with Latent Variables Panel Analysis Answering Typical Panel Questions Panel Analysis Investigating Causal Hypotheses Trend Analysis Cohort Analysis A Summary View

Latent Structure Models with Direct Effects between Indicators

A basic assumption of latent structure models is that of local independence: given the score on the latent variable, the scores on the manifest variables are independent of each other. This basic assumption is violated when test-retest effects, response consistency effects, correlated response errors, and so forth are present. However, it is possible to reformulate the latent class model in such a way that these direct relations between the indicators (manifest variables) can be accounted for. The reformulation proposed here, takes place within the framework of log-linear modeling.

The Multiple Propensity Score as Control for Bias in the Comparison of More Than Two Treatment Arms An Introduction From a Case Study in Mental Health

Background and Objective:The propensity score method (PS) has proven to be an effective tool to reduce bias in nonrandomized studies, especially when the number of (potential) confounders is large and dimensionality problems arise. The PS method introduced by Rosenbaum and Rubin is described in detail for studies with 2 treatment options. Since in clinical practice we are often interested in the comparison of multiple interventions, there was a need to extend the PS method to multiple treatments. It has been shown that in theory a multiple PS method is possible. So far, its practical application is rare and a practical introduction lacking. Methods:A practical guideline to illustrate the use of the multiple PS method is provided with data from a mental health study. The multiple PS is estimated with a multinomial logistic regression analysis. The multiple PS is the probability of assignment to each treatment category. Subsequently, to estimate the treatment effects while controlling for initial differences, the multiple PSs, calculated for each treatment category, are included as extra predictors in the regression analysis. Results:With the multiple PS method, balance was achieved in all relevant pretreatment variables. The corrected estimated treatment effects were somewhat different from the results without control for initial differences. Conclusions:Our results indicate that the multiple PS method is a feasible method to adjust for observed pretreatment differences in nonrandomized studies where the number of pretreatment differences is large and multiple treatments are compared.

Categorical Causal Modeling Latent Class Analysis and Directed Log-Linear Models with Latent Variables

Latent class analysis (LCA) is an extremely useful and flexible technique for the analysis of categorical data, measured at the nominal, ordinal, or interval level (the latter with fixed or estimated scores). It is, first, a general measurement model, a particular kind of latent structure model that can be used for the investigation of the reliability and validity of categorical measurements, taking both random and systematic response errors into account. When dealing with test-retest effects, response consistency effects, unobserved heterogeneity, response effects from varying survey (interviewing) conditions, and so on, it is useful to take a consistent causal modeling point of view and to integrate LCA into a general causal log-linear model with latent variables. The resulting directed log-linear modeling approach integrates insights from Goodmans modified path approach, the modified LISREL approach, and directed graphical modeling.

Estimating True Changes when Categorical Panel Data are Affected by Uncorrelated and Correlated Classification Errors An Application to Unemployment Data

Conclusions about changes in categorical characteristics based on observed panel data can be incorrect when (even a small amount of) measurement error is present. Random measurement errors, referred to as independent classification errors, usually lead to over-estimation of the total amount of gross change, whereas systematic, correlated errors usually cause underestimation of the transitions. Furthermore, the patterns of true change may be seriously distorted by independent or systematic classification errors. Latent class models and directed log-linear analysis are excellent tools to correct for both independent and correlated measurement errors. An extensive example on labor market states taken from the Survey of Income and Program Participation panel is presented.

LCAG — Loglinear Modelling with Latent Variables: a Modified LISREL Approach

In order to explicate the potentialities of loglinear modelling Goodman introduced the phrases ‘modified multiple regression approach’ and ‘modified path analysis approach’ (Goodman, 1972, 1975, 1973). As he pointed out, the loglinear techniques are not exactly identical with the classical regression techniques, but there is indeed a rather striking analogy between the two (see also Brier, 1979). As such, these phrases were very aptly chosen. They convey the general impression that the questions to be answered by the analyses of data measured at a nominal scale are essentially the same as the questions one tries to answer while analysing interval or ratio level data.

Conceptual issues of research methodology for the behavioural, life and social sciences

Summary. Research methodology (RM) must be clearly separated from substantive fields, such as medicine, psychology, education, sociology and economics, and, on the other side, from the philosophy of science and statistics. RM starts from substantive research problems and uses statistical knowledge, but it goes its own way in developing and applying new methods and concepts, all of which are relevant to RM consultants. Examples of topics are given that are typical for RM research and consultancy. Finally, it is argued that RM needs its own research programme and curriculum for RM consultants.

Analyzing Change in Categorical Variables by Generalized Log-Linear Models

This article discusses how several hypotheses about change in discrete variables can be tested on data obtained in a longitudinal study. A first class of hypotheses pertain to the invariance of certain characteristics of marginal distributions. A second class of hypotheses derive from assumptions about the causal relations between the variables. In this article, the authors show how all these hypotheses can be tested by means of a generalization of log-linear modeling developed by Lang and Agresti. By means of the same approach, it is also possible to test conjunctions of several hypotheses from both classes.

Lessons learnt from an unsuccessful clinical trial of homœopathy: Results of a small-scale, double-blind trial in proctocolitis*

Abstract In a randomized, double-blind study, the effects of individualized homœopathy in patients suffering from proctocolitis were investigated. The study was controlled against conventional therapy and placebo. Individually indicated homoœopathic medicines were prescribed in the 30c, 200c or 1,000c potencies. The researchers did not succeed in attracting sufficient numbers of participants to arrive at statistically significant conclusions. The total number of patients in the three combined groups was only 20, which was far below the target figure. The causes of this are discussed. Bowel habit and subjective parameters were assessed by a weekly health calendar. This calendar had not been previously validated. The conventional group scored best: eight out of ten patients completed the year of research and improved both with regard to their proctocolitis and their general well-being.