This chapter introduces the basic multigroup latent class (LC) model, and discusses two important extensions of the basic model, an extension for dealing with ordinal indicators and for modeling the latent variables as ordinal variables. It analyses measurement invariance using multigroup LC models, discussing the general procedure as well as methods for parameter estimation and evaluation of model fit. The multigroup extension of the standard LC model has been developed for the analysis of latent structures of observed categorical variables across two or more groups. The chapter presents three parameterizations of the multigroup LC models: probabilistic, log-linear, and logistic parameterizations. Multigroup LC models assume the presence of three types of categorical variables: observed variables; an unobserved variable that accounts for the relationships between the observed variables. LC models are usually estimated by means of maximum-likelihood under the assumption of a multinomial distribution for the indicator variables in the model.

The form-resistant hypothesis states that alternative ways of measuring the same values should be small if method-specific features are taken into account. However, previous research that compared rating and ranking questionnaires for measuring values has shown mixed results. We suggest that adopting a latent class segmentation approach helps to explain these mixed results by identifying segments with similar item preference structures and segments linked to one format only. Our approach is applied to a Dutch survey on work values. In both ranking and rating mode, we find two similar segments reflecting the intrinsic and extrinsic preference structure, while other segments differed between modes. In line with the modified form-resistant hypothesis, the results suggest the same latent preference structure has guided particular segments in a population to respond similarly to rating and ranking questions.

Survey data are often used to map cultural diversity by aggregating scores of attitude and value items across countries. However, this procedure only makes sense if the same concept is measured in all countries. In this study we argue that when (co)variances among sets of items are similar across countries, these countries share a common way of assigning meaning to the items. Clusters of cultures can then be observed by doing a cluster analysis on the (co)variance matrices of sets of related items. This study focuses on family values and gender role attitudes. We find four clusters of cultures that assign a distinct meaning to these items, especially in the case of gender roles. Some of these differences reflect response style behavior in the form of acquiescence. Adjusting for this style effect impacts on country comparisons hence demonstrating the usefulness of investigating the patterns of meaning given to sets of items prior to aggregating scores into cultural characteristics.

The key research question asked in this research is to what extent the respondents’ answers to ranking a set of items is mirrored in the response pattern when using rating questions. For example: Do respondents who prefer intrinsic over extrinsic work values in a ranking questionnaire also rate intrinsic values higher than extrinsic values when ratings are used? We adopt a modified version of the form-resistant hypothesis, arguing that each questionnaire mode yields unique features that prevent it from establishing a perfect match between both modes. By adopting a unified latent class model that allows identifying latent class profiles that share a particular preference structure in both question modes, we show that a large portion of respondents tend to identify similar preferences structures in work values regardless of the questionnaire mode used. At the same time the within-subjects design we use is able to answer questions regarding how non-differentiators in a rating assignment react to a ranking assignment in which non-differentiation is excluded by design. Our findings are important since – contrary to popular belief – ranking and ratings do produce results that are more similar than often thought. The practical relevance of our study for secondary data analysts is that our approach provides them with a tool to identify relative preference structures in a given dataset that was asked by rating questions and hence not directly designed to reveal such preferences.

Many variables crucial to the social sciences are not directly observed but instead are latent and measured indirectly. When an external variable of interest affects this measurement, estimates of its relationship with the latent variable will then be biased. Such violations of “measurement invariance” may, for example, confound true differences across countries in postmaterialism with measurement differences. To deal with this problem, researchers commonly aim at “partial measurement invariance” that is, to account for those differences that may be present and important. To evaluate this importance directly through sensitivity analysis, the “EPC-interest” was recently introduced for continuous data. However, latent variable models in the social sciences often use categorical data. The current paper therefore extends the EPC-interest to latent variable models for categorical data and demonstrates its use in example analyses of U.S. Senate votes as well as respondent rankings of postmaterialism values in the World Values Study.

Measuring values in sociological research sometimes involves the use of ranking data. A disadvantage of a ranking assignment is that the order in which the items are presented might influence the choice preferences of respondents regardless of the content being measured. The standard procedure to rule out such effects is to randomize the order of items across respondents. However, implementing this design may be impractical and the biasing impact of a response order effect cannot be evaluated. We use a latent choice factor (LCF) model that allows statistically controlling for response order effects. Furthermore, the model adequately deals with the known issue of ipsativity of ranking data. Applying this model to a Dutch survey on work values, we show that a primacy effect accounts for response order bias in item preferences. Our findings demonstrate the usefulness of the LCF model in modeling ranking data while taking into account particular response biases.

The Program for International Student Assessment (PISA) is a large-scale cross-national study that measures academic competencies of 15-year-old students in mathematics, reading, and science from more than 50 countries/economies around the world. PISA results are usually aggregated and presented in so-called “league tables,” in which countries are compared and ranked in each of the three scales. However, to compare results obtained from different groups/countries, one must first be sure that the tests measure the same competencies in all cultures. In this paper, this is tested by examining the level of measurement equivalence in the 2009 PISA data set using an item response theory approach (IRT) and analyzing differential item functioning (DIF). Measurement in-equivalence was found in the form of uniform DIF. In-equivalence occurred in a majority of test questions in all three scales researched and is, on average, of moderate size. It varies considerably both across items and across countries. When this uniform DIF is accounted for in the in-equivalent model, the resulting country scores change considerably in the cases of the “Mathematics,” “Science,” and especially, “Reading” scale. These changes tend to occur simultaneously and in the same direction in groups of regional countries. The most affected seems to be Southeast Asian countries/territories whose scores, although among the highest in the initial, homogeneous model, additionally increase when accounting for in-equivalence in the scales.

Inglehart applies a four item ranking scale to measure post-materialism which is used for cross-cultural and cross-temporal comparative purposes. The aim of this research is to test measurement invariance of the scale to establish to what extent the scale produces comparable results in time and between countries. We use Eurobarometer data to test longitudinal comparability for ten countries (France, Belgium, the Netherlands, Italy, West-Germany, Luxembourg, Denmark, Ireland, Great Britain and North Ireland) over a period of 20 years (1976–1997). With the exception of Denmark the within-country longitudinal comparisons indicate that measurement invariance is a tenable assumption. However, the findings of the cross-cultural analyses indicate that the meaning assigned to the four items differs slightly between countries, indicating a lack of comparability of the average level of post-materialism between countries. Findings also suggest that a lack of unidimensionality of the scale might cause this incomparability.

Extreme response style (ERS) and acquiescence response style (ARS) are among the most encountered problems in attitudinal research. The authors investigate whether the response bias caused by these response styles varies with following three aspects of question format: full versus end labeling, numbering answer categories, and bipolar versus agreement response scales. A questionnaire was distributed to a random sample of 5,351 respondents from the Longitudinal Internet Studies for the Social Sciences household panel, of which a subsample was assigned to one of five conditions. The authors apply a latent class factor model that allows for diagnosing and correcting for ERS and ARS simultaneously. The results show clearly that both response styles are present in the data set, but ARS is less pronounced than ERS. With regard to format effects, the authors find that end labeling evokes more ERS than full labeling and that bipolar scales evoke more ERS than agreement style scales. With full labeling, ERS opposes opting for middle response categories, whereas end labeling distinguishes ERS from all other response categories. ARS did not significantly differ depending on test conditions.

AbstractThe purpose of event history analysis is to explain why certain individuals are at a higher risk of experiencing the event(s) of interest than others. This can be accomplished by using special types of methods which are usually referred to as hazard models. Two special features of hazard models are that they can deal with censored observations and with covariates that change their values during the observation period.

In this article, data from the 2005 European Working Conditions Survey are used to examine the relationship between contemporary employment arrangements and the work-related well-being of European employees. By means of a Latent Class Cluster Analysis, several features of the employment conditions and relations characterizing jobs are combined in a typology of five employment arrangements: SER-like, instrumental, precarious unsustainable, precarious intensive and portfolio jobs. These job types show clear relationships with separate indicators of job satisfaction, perceived safety climate and the ability to stay in employment, as well as with an overall indicator for work-related well-being. The findings from this multifaceted approach towards employment quality raise questions about the long-term sustainability of highly flexible and de-standardized employment arrangements.

In survey research, acquiescence response style/set (ARS) and extreme response style/set (ERS) may distort the measurement of attitudes. How response bias is evoked is still subject of research. A key question is whether it may be evoked by external factors (e.g. test conditions or fatigue) or whether it could be the result of internal factors (e.g. personality or social characteristics). In the first part of this study we explore whether scale length—the manipulated test condition—influences the occurrence of ERS and/or ARS, by varying scale length from 5 till 11 categories. In pursuit of this we apply a latent class factor model that allows for diagnosing and correcting for ERS and ARS simultaneously. Results show that ERS occurs regardless of scale length. Furthermore, we find only weak evidence of ARS. In a second step we check whether ERS might reflect an internal personal style by (a) linking it to external measures of ERS, and by (b) correlating it with a personality profile and socio-demographic characteristics. Results show that ERS is reasonably stable over questionnaires and that it is associated with the selected personality profile and age.

In this article employment quality in the EU27 is investigated by means of a typological approach, based on several features of the employment conditions and relations characterising jobs. The analyses are drawing on data from the 2005 European Working Conditions Survey. Results of Latent Class Cluster Analyses show that it is empirically and theoretically possible to reduce a multitude of factors determining the quality of employment into five different types of jobs regarding their employment quality: SER-like jobs, instrumental jobs, precarious unsustainable jobs, precarious intensive jobs and portfolio jobs. These five types of jobs are strongly related with important covariates such as the socio-demographic profile of workers, organisation level features and indicators of the intrinsic nature of work tasks. Moreover, they are clearly distributed differently between countries within the EU27. The findings from this innovative approach towards the quality of employment are discussed in terms of the implications for the measurement of contemporary employment arrangements in Europe.

Can existing longitudinal surveys profit from the (financial) advantages of web surveying by switching survey mode from face-to-face interviews to web surveys? Before such a radical change in data collection procedure can be undertaken, it needs to be established that mode effects cannot confound the responses to the survey items. To this end, the responses of the Dutch European Values Study of 2008 were compared to the responses of a time parallel web survey. The responses on 163 of the 256 items differed significantly across modes. To explain these response differences between modes, an exploratory crisp set qualitative comparative analysis approach was used. Five sufficient conditionscombinations of survey mode characteristicsbut no necessary conditions for response differences between survey modes were found. Two survey characteristics were neither necessary nor sufficient to produce the outcome. Results suggest that switching modes may affect comparability between waves in a longitudinal survey.

Transformational, transactional, and laissez-faire leadership are considered to be three distinct leadership styles. In this research we argue that response style behaviour in the form of acquiescence and extreme response style can distort the measurement of these dimensions of leadership. Using a sample of 864 employees selected from 135 work teams, this research demonstrates that (1) response styles affect measurement; (2) divergent validity of the three dimensions increases when response styles are taken into account; (3) gender is spuriously related to leadership upon response styles; and (4) team ratings substantially change when controlling for response styles bias. As a secondary topic of this research, we elaborate on a relatively new approach in diagnosing response styles, i.e., a confirmatory latent-class factor analysis. We explain the advantages of this approach and illustrate the steps a researcher has to take in conducting this type of analysis.

Generational Differences in Political Value Orientations: An International Comparison. In this artiele we take up the discussion, introduced by Ronaid Inglehart, that cohorts tend to create a political identity in response to their historical positioning. In doing so, we examine the differential spacing of cohorts for each of the four concepts constituting the Materialist Postmaterialist dimension. It has proved to be relevant to distinguish an economie and a non-economie domain: the latter shows the most pronounced spacing, whereas the former only has minor differences between cohorts, largely due to the level of education.

Minorities' attitudes can be compared to attitudes of fellow citizen within the host country as well as to attitudes of the motherland. Given the heterogeneity of Luxembourg's minority groups, this country is a relevant example case in which the comparison needs to involve answering a two-folded question. First, the authors analyze the level of measurement equivalence, that is, the extent to which different groups can be compared. Second, the authors examine whether ethnic-cultural groups within Luxembourg resemble citizens from their native country more than their country of residence. Using European Values Study (EVS) data from 2008 the authors demonstrate different types of outcomes. Results indicate that cultural background is more important than national context in the case of culturally more distant minorities to Luxembourg's resident population and that national setting is the prevailing factor when minorities are from neighboring countries. The effect of a common national setting is also important with regards to the issue of measurement equivalence, where it contributes to greater comparability of intranational, cross-ethnic comparisons.