Showing papers by "Joop J. Hox published in 1998"

An introduction to structural equation modeling

Abstract: This article presents a short and non-technical introduction to Structural Equation Modeling or SEM. SEM is a powerful technique that can combine complex path models with latent variables (factors). Using SEM, researchers can specify confirmatory factor analysis models, regression models, and complex path models. We present the basic elements of a structural equation model, introduce the estimation technique, which is most often maximum Likelihood (ML), and discuss some problems concerning the assessment and improvement of the model fit, and model extensions to multigroup problems including factor means. Finally, we discuss some of the software, and list useful handbooks and Internet sites. What is Structural Equation Modeling? Structural Equation Modeling, or SEM, is a very general statistical modeling technique, which is widely used in the behavioral sciences. It can be viewed as a combination of factor analysis and regression or path analysis. The interest in SEM is often on theoretical constructs, which are represented by the latent factors. The relationships between the theoretical constructs are represented by regression or path coefficients between the factors. The structural equation model implies a structure for the covariances between the observed variables, which provides the alternative name covariance structure modeling. However, the model can be extended to include means of observed variables or factors in the model, which makes covariance structure modeling a less accurate name. Many researchers will simply think of these models as ‘Lisrel-models,’ which is also less accurate. LISREL is an abbreviation of LInear Structural RELations, and the name used by Joreskog for one of the first and most popular SEM programs. Nowadays structural equation models need not be linear, and the possibilities of SEM extend well beyond the original Lisrel program. Browne (1993), for instance, discusses the possibility to fit nonlinear curves. Structural equation modeling provides a very general and convenient framework for statistical analysis that includes several traditional multivariate procedures, for example factor analysis, regression analysis, discriminant analysis, and canonical correlation, as special cases. Structural equation models are often visualized by a graphical path diagram. The statistical model is usually represented in a set of matrix equations. In the early seventies, when this technique was first introduced in social and behavioral research, the software usually required setups that specify the model in terms of these matrices. Thus, researchers had to distill the matrix representation from the path diagram, and provide the software with a series of matrices for the different sets of 1 Note: The authors thank Alexander Vazsonyi and three anonymous reviewers for their comments on a previous version. We thank Annemarie Meijer for her permission to use the quality of sleep data. Introduction Structural Equation Modeling 2 parameters, such as factor loadings and regression coefficients. A recent development is software that allows the researchers to specify the model directly as a path diagram. This works well with simple problems, but may get tedious with more complicated models. For that reason, current SEM software still supports the commandor matrix-style model specifications too. This review provides a brief and non-technical review of the basic issues involved in SEM, including issues of estimation, model fit, and statistical assumptions. We include a list of available software, introductory books, and useful Internet resources. Examples of SEM-Models In this section, we set the stage by discussing examples of a confirmatory factor analysis, regression analysis, and a general structural equation model with latent variables. Structural equation modeling has its roots in path analysis, which was invented by the geneticist Sewall Wright (Wright, 1921). It is still customary to start a SEM analysis by drawing a path diagram. A path diagram consists of boxes and circles, which are connected by arrows. In Wright’s notation, observed (or measured) variables are represented by a rectangle or square box, and latent (or unmeasured) factors by a circle or ellipse. Single headed arrows or ‘paths’ are used to define causal relationships in the model, with the variable at the tail of the arrow causing the variable at the point. Double headed arrows indicate covariances or correlations, without a causal interpretation. Statistically, the single headed arrows or paths represent regression coefficients, and double-headed arrows covariances. Extensions of this notation have been developed to represent variances and means (cf. McArdle, 1996). The first example in Figure 1 is a representation of a confirmatory factor analysis model, with six observed variables and

Multilevel modeling: When and why

Abstract: Multilevel models have become popular for the analysis of a variety of problems. This chapter gives a summary of the reasons for using multilevel models, and provides examples why these reasons are indeed valid. Next, recent (simulation) research is reviewed on the robustness and power of the usual estimation procedures with varying sample sizes.

A comparison of randomized response, CASAO, and direct questioning; eliciting sensitive information in the context of social security fraud

Abstract: The aim of the study is to assess the validity of the responses to sensitive questions on social security fraud obtained using four different methods. We compare two different varieties of randomized response with CASAQ and direct questioning in an experimental setting. Validity could be assessed because all respondents interviewed had already been identified as committing social security fraud. The experiment was set up in such a way that the interviewers did not know that respondents had been caught for fraud, and the respondents did not know that the researchers had this information. Since the actual status of the respondents is known, it is possible to compare the results of the four approaches by comparing the percentage of false negatives. Two additional questions are these: which respondents are willing to admit to having practised fraud, and are the respondent characteristics that predict positive responses to the sensitive questions the same for all methods? For the direct question method and CASAQ these questions are answered by straightforward logistic regression. For the randomized response answers, logistic regression models are adjusted. Although the RR conditions perform much better than more traditional approaches, the percentage of respondents admitting to fraud is far less than 100 %. Some reasons for this are discussed.

Nonrespons in surveys: een overzicht

Abstract: SAMENVATTING. Nonrespons, oftewel het niet kunnen verkrijgen van gegevens van potentiële respondenten, vormt een ernstige bedreiging voor de geldigheid van enquêteresultaten. Nationaal en internationaal groeit momenteel de bezorgdheid om een toenemende nonrespons. In dit artikel wordt eerst een overzicht gegeven van internationale cijfers over nonresponse en wordt aandacht geschonken aan theorieën die een mogelijke verklaring geven van het verschijnsel. Daarna worden een aantal manieren besproken om de nonrespons het hoofd te bieden. De nadruk ligt hierbij op theoretisch onderbouwde veldwerkstrategieën om nonrespons te bestrijden, maar ook zal kort aandacht besteed worden aan statistische wegingstechnieken.

Interviewer opinions, attitudes and strategies regarding survey participation and their effect on response

Abstract: Nonresponse is a threat to the validity of conclusions based on survey data. In general, two strategies are used to counteract this threat. The first strategy is to reduce the proportion of nonresponse as far as possible, the second is to statistically adjust for the remaining nonresponse. Interview surveys are still the norm for official statistics, social studies und market research in the Netherlandr, und interviewers are an important factor in the battle against nonresponse. We focus on interviewers' opinions on nonresponse und their attitudes regarding the role of the interviewer in persuading potential respondents. In a special project at Statistics Netherlands the continuous survey on living conditions (POLS) was redesigned. During that study interviewer data were collected. It is shown that interviewer attitude und response rate are correlated. Interviewers with a positive attitude towards persuasion strategies attain a higher response rate. No dlferences between interviewers are found regarding self-reported 'door step' behaviour.

Fighting nonresponse in telephone interviews: Successful interviewer tactics

Abstract: Bei Telephoninterviews haben Interviewer weitaus weniger Zeit als in face-to-face-Interviews, einen Befragten zur Kooperation zubringen. Da sie ihren Gesprachspartner nur horen konnen, fehlt es ihnen auch an Informationen, um sich selbst optimal zu verhalten. Gleichwohl verfugen erfahrene Telephoninterviewer uber einen Vorrat an taktischen Varianten, die sie einsetzen konnen. In diesem Beitrag werden Taktiken beschrieben, mit denen sich Antwortverweigerung bekampfen lasst und die von erfahrenen Telephoninterviewern bei Statistics Netherlands angewandt werden. (ICEUbers)

On Going Research/Recherche En Cours Detecting Aberrant Response Patterns in Multi-Item Scales: Some Nonparametric Indices and a Computer Program

Abstract: Detecter des formes de reponses aberrantes sur des echelles multi-reponses: Des indices non-parametriques et un logiciel. La litterature de psychologie experimentale comprend plusieurs indices de mesurer censes identifier des formes de reponses aberrantes. Cet article examine trois indices non-parametriques bien connus, donne des exemples d'application et presente un logiciel qui calcule ces indices.

Detecting aberrant response patterns in multi-item scales: some nonparametric indices and a computer program

Abstract: La litterature de psychologie experimentale comprend plusieurs indices de mesure censes identifier des formes de reponses aberrantes. Cet article examine trois indices non-parametriques bien connus, donne des exemples d'application et presente un logiciel qui calcule ces indices.