Search results

1

Estimating Support for Extremism and Its Correlates: The Case of Pakistan

100%

Al Baghal T.

ASK. Research and Methods

|

2014

|

vol. 23

35-56

EN

The extent of support of extremist ideology is a major area of concern for both policy makers and academic researchers. Identifying the extent and correlates of a difficult to measure concept such as extremist ideology is often limited by the use of a single imperfect indicator. This paper outlines one approach, latent class analysis (LCA), to overcome this issue and uses the example of estimating support for such ideology in Pakistan. Using survey data from Pakistani men, the level of support is estimated using LCA employing several indicators related to extremism. The results suggest that although most Pakistanis are not supportive of extremist ideology, a substantively important portion of men are supportive. LCA also allows for class assignment, which is useful for understanding covariate relationships with the latent variable. Based on the results of the LCA, respondents are assigned to different classifications of extremist support, and a continuation-ratio logistic regression model is employed allowing for more covariates to be examined. The results suggest that there are a number of characteristics important in influencing support within this subset of the population. In particular, younger and less educated men are more likely to support extremism ideology. The results suggest a potentially useful methodology in understanding extremism, as well as a greater understanding of the problem of extremist support.

2

Patterns of social embeddedness in later adulthood: gender and other covariates

100%

Mikulionienė S., Gaižauskaitė I., Morkevičius V.

Gender a výzkum / Gender and Research

|

2021

|

vol. 22

|

issue 1

36-58

EN

The concept of social exclusion is widely used in academia and social policy discourse in European countries. However, one of the constituents of social exclusion, namely, exclusion from social relationships, remains unclear and underdeveloped both conceptually and empirically. Moreover, knowledge on the patterns of exclusion from social relationships among men and women in later adulthood is even less advanced. Therefore, we developed a multidimensional scale for measuring an individual’s level of embeddedness in social relationships and examined the gender patterns of social embeddedness. We applied latent class analysis to survey data collected on a sample of 2015 residents of Lithuania, Latvia, and Estonia aged 50+. We derived a seven-class model based on a continuum from strong to weak social embeddedness. We identified two classes with a high level of exclusion from social relations and, conversely, two classes with a high level of social embeddedness. Subsequent multinomial regression analysis revealed that gender was a statistically significant predictor in the cases of the most excluded class and one of the most socially embedded classes.

3

Latent class analysis in the evaluation of items in survey research

100%

Brzezińska J.

Econometrics. Ekonometria. Advances in Applied Data Analytics

|

2018

|

issue vol. 22 no. 3

55-65

EN

Latent class analysis has been widely used in the measurement models. Models based on latent variables have a wide range of applications in the presence of repeated ob-servations, longitudinal data, and multilevel data. In this paper we present and apply log-linear analysis as a method for the analysis of multi-way tables. We also present a latent variable model based on a variable that is not directly observed. The basic model postulates an underlying categorical latent variable; within any category of the latent variable the manifest or observed categorical variables are assumed independent of one another (axiom of conditional independence). In this paper we present the results of a survey research based on categorical data and the author`s questionnaire. We present the results of the latent class analysis in the classification of respondents into clusters characterized by similar attitudes and features in economic research. We also conduct a prior log-linear analysis for a multi-way contingency table. All the calculations are conducted in R.

4

Teoria reakcji na pozycję w podejściu modelowym w takso- nomii

100%

Genge E.

Econometrics. Ekonometria. Advances in Applied Data Analytics

|

2016

|

issue 1 (51)

9-19

EN

Item response theory is considered to be one of the two trends in methodological assessment of the reliability scale. In turn, latent class models can be viewed as a special case of model-based clustering, for heterogenous multivariate discrete data. We used the approach combining item response theory and latent class models to find groups of Polish households’ with similar saving ability levels. We analyzed data collected as part of the Polish Social Diagnosis using MultiLCIRT package of R.

5

Item response theory models in the measurement theory with the use of LTM package in R

100%

Brzezińska J.

Econometrics. Ekonometria. Advances in Applied Data Analytics

|

2018

|

issue vol. 22 no. 1

11-25

EN

Item Response Theory (IRT) is an extension of the Classical Test Theory (CCT) and focuses on how specific test items function in assessing a construct. They are widely known in psychology, medicine, and marketing, as well as in social sciences. An item response model specifies a relationship between the observable examinee test performance and the unobservable traits or abilities assumed to underlie performance on the test. Within the broad framework of item response theory, many models can be operationalized because of the large number of choices available for the mathematical form of the item characteristic curves. In this paper we introduce several types of IRT models such as: the Rasch, and the Birnbaum model. We present the main assumptions for IRT analysis, estimation method, properties, and model selection methods. In this paper we present the application of IRT analysis for binary data with the use of the ltm package in R.

6

Poczucie śląskości wśród Ślązaków – analiza empiryczna z wykorzystaniem modeli klas ukrytych

100%

Genge E.

Econometrics. Ekonometria. Advances in Applied Data Analytics

|

2013

|

issue 4(42)

48-59

EN

The paper focuses on latent class models and their application for quantitative data. Latent class modeling is one of multivariate analysis techniques of the contingency table and can be viewed as a special case of model-based clustering, for multivariate discrete data. It is assumed that each observation comes from one of the numbers of subpopulations, with its own probability distribution. We used latent class analysis for grouping and detecting homogeneity of Silesian people using poLCA package of R. We analyzed data collected by the Department of Social Pedagogy, University of Silesia in Katowice.

7

Veřejná podpora přímé demokracie : analýza heterogenity politických postojů v České republice

88%

Dvořák T.

Sociologický časopis. Czech Sociological Review

|

2019

|

vol. 55

|

issue 4

419-443

EN

This paper focuses on the sources of popular support for direct democracy in the Czech Republic. The analysis first replicates the standard approaches used in previous scholarly research, testing theories of cognitive mobilisation and alienation using the standard regression modelling approach. The results are, however, somewhat inconclusive, as have been the results of previous empirical research in other European countries. It is hypothesised that one reason for this could be data heterogeneity. Both tested concepts could be valid, but for different social groups, resulting in overall inconclusive results. Therefore, latent class analysis (LCA) is then used to show that in the Czech Republic direct democracy is supported both by people alienated from politics and by those satisfied with democracy and democratic governance. The results of empirical analysis show that support for direct democracy in the Czech Republic cannot be explained by any one theory and that different concepts apply to different groups. The article represents a methodological innovative step in the empirical study of sources of popular support for direct democracy.

8

Patterns of Relapse Risks and Related Factors among Patients with Schizophrenia in Razi Hospital, Iran: A Latent Class Analysis

75%

Noroozi M., Alibeigi N., Armoon B., Rezaei O., Sayadnasiri M., Nejati S., Fadaei F., Ghahestany D. A., Dieji B., Ahounbar E.

Polish Psychological Bulletin

|

2018

|

vol. 49

|

issue 3

9

Wykorzystanie modeli ze zmiennymi porządkowymi w badaniach satysfakcji klienta

63%

Błoński K., Putek-Szeląg E.

Zeszyty Naukowe Uniwersytetu Ekonomicznego w Krakowie

|

2013

|

issue 909

39-47

PL

Pomiar satysfakcji klienta możliwy jest z wykorzystaniem skal ilościowych, jak również jakościowych. Jednak stosowanie każdej z nich związane jest z problemami i ograniczeniami. Niektórzy autorzy, np. N. Hill i J. Alexander w publikacji "Pomiar satysfakcji i lojalności klientów" (2003) jednoznacznie wskazują najlepszy według nich rodzaj skal do pomiaru satysfakcji – skalę liczbową. Mimo takich wskazówek najczęściej stosowana jest skala porządkowa. Jej użycie powoduje konieczność wykorzystania metod statystycznych pozwalających na analizowanie danych jakościowych. Artykuł przedstawia zastosowanie analizy klas ukrytych w badaniach satysfakcji. Do budowy modelu wykorzystano dane z pomiaru dokonanego na skalach porządkowych.

EN

Customer satisfaction can be measured using both quantitative and qualitative scales. However, each of them has problems and limitations. Some authors, such as N. Hill and J. Alexander in their “Handbook of Customer Satisfaction and Loyalty Measurement” (2003), clearly believe numerical scales are the best for measuring satisfaction. Nonetheless, ordinal scale, which requires statistical methods for the analysis of qualitative data, remains the most commonly used. This paper presents the use of analysis of latent class in satisfaction surveys. To build a model, data from measurements made with ordinal scales were used.

10

Kryteria wyboru liczby skupień w binarnym modelu klas ukrytych – analiza symulacyjna

63%

Kapłon R.

Przegląd Statystyczny

|

2010

|

vol. 57

|

issue 1

66-84

EN

When using latent class analysis the number of clusters need to be known in advance. In order to decide on this, one can use information criteria. In such a case selection procedure is as follows: estimating a few models with different number of classes, computing information criteria and choosing a model for which a criterion takes the smallest value. Because there are many information criteria one need to determine which of them ought to be decisive. Unfortunately, by virtue of the differences among these criteria, their reliability alter depending on model class. Simulations confirm it as well. Taking into account the fact that simulations mainly concern finite mixtures of normal density functions, therefore in this paper we broaden research to latent class analysis.

PL

Wykorzystanie analizy klas ukrytych (LCA) wymaga przyjęcia a priori liczby klas. W celu rozstrzygnięcia, ile ma ich być, można wykorzystać kryteria informacyjne. Procedura selekcji sprowadza się do: szacowania kilku modeli o różnej liczbie klas, obliczenia wartości kryterium informacyjnego oraz wyboru modelu, dla którego odnotowano najmniejszą wartość tego kryterium. Ponieważ istnieje wiele kryteriów informacyjnych, więc należy zadecydować, które powinno rozstrzygać. Niestety, nie można jednoznacznie wskazać na konkretne kryterium, gdyż w zależności od klasy modelu, zmienia się ich wiarygodność. Taki wniosek wynika z badań symulacyjnych. Biorąc pod uwagę fakt, że najczęściej badania takie dotyczyły mieszanek rozkładów normalnych, dlatego celem niniejszego opracowania jest rozszerzenie tych badań o analizę klas ukrytych.

11

Prior distributions for Bayes factors and latent class model selection

51%

KAPŁON R.

Operations Research and Decisions

|

2009

|

vol. 19

|

issue 3

87-94

PL

Na etapie wyboru liczby segmentów w analizie klas ukrytych kryteria informacyjne są często stosowane. Szczególne miejsce zajmuje tutaj kryterium bayesowskie BIC, które można wyprowadzić – dokonując pewnych uproszczeń – z koncepcji czynnika bayesowskiego. W czynniku tym pojawia się rozkład a priori parametrów, którego nie ma w BIC. Z tego względu w pracy podjęto próbę znalezienia takiego rozkładu a priori, aby skuteczność tak powstałego kryterium była większa niż skuteczność BIC.

12

The Experience of Peer Aggression as a Heterogeneous Phenomenon: A Latent Class Analysis

51%

Frankowiak J., Kotyśko M.

Studia Paedagogica Ignatiana

|

2022

|

vol. 25

|

issue 4

45-66

PL

Agresja jest wpisana w relacje rówieśnicze. Jednak nasilona i realizowana w sposób systematyczny, niesie ze sobą negatywne skutki dla prawidłowego rozwoju psychospołecznego. Głównym celem badań było wyłonienie wśród uczniów klas ze względu na podobieństwo w zakresie doświadczania poszczególnych form i przejawów agresji rówieśniczej. Dane zgromadzono za pomocą autorskiego kwestionariusza, z którego do analizy użyto 10 pytań tworzących wskaźniki doświadczanej agresji (pięć form, po dwa przejawy). W badaniu wzięło udział 1050 uczniów gimnazjum (po 525 chłopców i dziewcząt) w wieku 13 i 14 lat. Uczniowie najczęściej doświadczali agresji werbalnej i relacyjnej, a najrzadziej przejawów agresji seksualnej. Chłopcy częściej informowali o doświadczaniu agresji fizycznej i werbalnej (bycie wyzywanym), a dziewczęta o doświadczeniu plotkowania na swój temat i cyberagresji (obraźliwe komentarze). Wykorzystując Analizę Klas Latentnych, poddano analizie model z sześcioma klasami: Wszystkie wskaźniki niskie, Wysoka werbalna i relacyjna, Wysoka fizyczna i werbalna, Wysokie wszystkie poza seksualną, Wysokie wszystkie wskaźniki oraz Wysoka seksualna, cyber i relacyjna. Uzyskane rezultaty pokazują, że zjawisko agresji rówieśniczej nie jest homogeniczne, a projektowanie działań profilaktycznych powinno uwzględniać specyfikę doświadczeń jej ofiar.

EN

Peer aggression, when escalated and regular, has negative consequences for proper psychosocial development. The main aim of this study was to identify classes among middle school students according to similarity in terms of experiencing particular forms and manifestations of peer aggression. Data were collected using a proprietary questionnaire, from which 10 questions forming indicators of experienced aggression (five forms, two manifestations each) were used for analysis. A total of 1.050 middle school students (525 boys and girls each) aged 13 and 14 participated in the study. The students most often experienced verbal and relational aggression, and least often sexual aggression. Boys were more likely to report experiencing physical and verbal aggression (being insulted), while girls reported being the target of gossip and cyber aggression (offensive comments). Using Latent Class Analysis, we analyzed a model with six classes as follows: Low all, High verbal and relational, High physical and verbal, High all without sexual, High all and High sexual, cyber and relational. The results show that the phenomenon of peer aggression is not homogeneous and that the design of preventive measures should take into account the specificity of the experiences of its victims.

Refine search results

Estimating Support for Extremism and Its Correlates: The Case of Pakistan

Patterns of social embeddedness in later adulthood: gender and other covariates

Latent class analysis in the evaluation of items in survey research

Teoria reakcji na pozycję w podejściu modelowym w takso- nomii

Item response theory models in the measurement theory with the use of LTM package in R

Poczucie śląskości wśród Ślązaków – analiza empiryczna z wykorzystaniem modeli klas ukrytych

Veřejná podpora přímé demokracie : analýza heterogenity politických postojů v České republice

Patterns of Relapse Risks and Related Factors among Patients with Schizophrenia in Razi Hospital, Iran: A Latent Class Analysis

Wykorzystanie modeli ze zmiennymi porządkowymi w badaniach satysfakcji klienta

Kryteria wyboru liczby skupień w binarnym modelu klas ukrytych – analiza symulacyjna

Prior distributions for Bayes factors and latent class model selection

The Experience of Peer Aggression as a Heterogeneous Phenomenon: A Latent Class Analysis