Ask: Research and Methods. Volume 16, Issue 1 (2007)

Permanent URI for this collection

https://kb.osu.edu/handle/1811/69454

Choosing among Discrete Choice Models for Voting Behavior
Dubrow, Joshua Kjerulf pp. 9-23
Article description | Article Full Text PDF

Are Non-Respondents Similar to Respondents? Findings from the ESS-2004 in Poland
Sztabinski, Franciszek; Sztabinski, Pawel B.; Przybysz, Dariusz pp. 25-54
Article description | Article Full Text PDF

Hierarchical Modeling: Properties and Application
Wilk, Katarzyna pp. 55-66
Article description | Article Full Text PDF

Wpływ niezrealizowania części wywiadów na trafność wnioskowania statystycznego w badaniach społecznych. Technika wywiadu kwestionariuszowego oraz telefonicznego w świetle błędów nielosowych [The Impact of not Carrying out Some Part of Interviews on the Validity of Statistical Inference in Social Research. Telephone Interviews and CATI in View of Systematic Errors]
Jabkowski, Piotr pp. 67-86
Article description | Article Full Text PDF

Analiza danych jakościowych przy użyciu programów komputerowych [Computer Assisted Qualitative Data Analysis]
Bielinski, Jacek; Iwinska, Katarzyna; Rosinska-Kordasiewicz, Anna pp. 89-114
Article description | Article Full Text PDF

Skale ryzyka nadmiernego używania internetu [Excessive Internet Use Risk Scales]
Kaliszewska, Katarzyna pp. 115-131
Article description | Article Full Text PDF

Browse

Now showing 1 - 6 of 6

Skale ryzyka nadmiernego używania internetu
(IFiS Publishers, 2007) Kaliszewska, Katarzyna
In the article the theoretical bases and the psychometric characteristics of the Excessive Internet Use Risk Scales were presented. The scales were constructed for the needs of empirical research conducted in order to psychologically characterize the Excessive Internet Use phenomenon. The psychometric parameters of the scales are satisfactory and it may be assumed that this tool can be successfully used in individual and group research as well as in inter-group comparisons of the Excessive Internet Use variable.
Analiza danych jakościowych przy użyciu programów komputerowych
(IFiS Publishers, 2007) Bieliński, Jacek; Iwińska, Katarzyna; Rosińska-Kordasiewicz, Anna
Qualitative and quantitative research are embedded in different epistemological traditions and typically differ in tools applied in data analysis. The computer software was mostly used within quantitative design to carry out statistical analysis. Among qualitative researchers there is a widespread opinion that computer software applied in qualitative analysis might kill (by subsuming to the standardized, uniformed procedures) the freshness and relevance of rich and non-reductionable empirical data. Nevertheless, since early nineties, the computer software has been growing more and more popular in qualitative research process. The software is more and more widely used in cataloguing, sorting out and data retrieval done throughout the qualitative data analysis process. We argue that the CAQDAS (computer assisted qualitative data analysis software) provide a useful tool for improving the rigor of analytical work as well as it might be a help in arranging data in synthesis in qualitative research. The aim of this article is to show the possibilities and usefulness of computer assisted qualitative data analysis (mostly: Atlas.ti and Nudist) based on examples drawn from two qualitative researches.
Wpływ niezrealizowania części wywiadów na trafność wnioskowania statystycznego w badaniach społecznych. Technika wywiadu kwestionariuszowego oraz telefonicznego w świetle błędów nielosowych
(IFiS Publishers, 2007) Jabkowski, Piotr
The article concerns the question of inaccessible units set in the perspective of the theory of statistical conclusion. The focus is on consequences of omitting of non-sampling errors resulting from the lack of knowledge of distribution of the opinions of people who were selected but not included in the sample. It has been shown that non-sampling errors resulting from not carrying out of part of the test (if they are not restricted efficiently) disturb the results to such an extent that drawing any reasonable conclusions becomes practically impossible. The article consists of three parts connected with each other. In the first the problem of not carrying out of a part of the test have been formulated and statistical theoretical bases for further arguments have been given. In the second part procedures of evaluation of non-sampling assessment errors resulting from not carrying out of assumed test at the beginning of the test have been derived. In the third part a critical analysis (in the light of the problems discussed) of two research techniques–telephone surveys and questionary surveys– has been provided.
Hierarchical Modeling: Properties and Application
(IFiS Publishers, 2007) Wilk, Katarzyna
Empirical analyses of hierarchical data are important in various disciplines, but are most common to the social sciences. Until the 1980’s, when the method of multilevel modeling was introduced, researchers dealt with the problem of nested data in a variety of ways, none of which was completely effective or accurate. The method of hierarchical modeling, and softwares such as HLM or MLwiN, provide the most appropriate available tools for dealing with the nested data. This article intends to introduce this strategy, as well as provide an empirical example to illustrate the relative advantages of using it to perform analysis.
Are Non-Respondents Similar to Respondents? Findings from the ESS-2004 in Poland
(IFiS Publishers, 2007) Sztabiński, Franciszek; Sztabiński, Paweł B.; Przybysz, Dariusz
In Poland, like in other countries, participation in surveys is declining over time. The growing non-response rates increase a risk of systematic bias of the results depending on the differences between non-respondents and respondents. In this article we attempt to assess how different are non-respondents from the survey participants in Poland. I approaching this question we refer to the two basic hypotheses concerning non-responses. The first hypothesis concerns the relationship between participation in surveys and the socio economic status (SES); the second hypothesis concerns relationship between the participation and social involvement/social isolation, and liberalism/conservatism. These hypotheses are tested using data from the pilot study of the European Social Survey 2004 and the main ESS 2004. Our analyses reveal that both the socio-demographic and the socio-psychological characteristics of individuals affect the non-response in Poland in a specific way.
Choosing among Discrete Choice Models for Voting Behavior
(IFiS Publishers, 2007) Dubrow, Joshua Kjerulf
Analyses presented in this paper aim at testing demographic cues hypothesis, which explains voting behavior as a function of the distance between the voter and the object of the vote, expressed as demographic similarity. Four types of multivariate regression models–binomial logistic (BNL), multinomial logistic (MNL), contrast logistic (CONTRAST), and conditional logistic (CLOGIT)–are applied to explain vote choice among Polish parties in the 1997 parliamentary election. For all models I use survey data combined with information on political parties derived from characteristics of the electoral candidates. The results demonstrate that for testing demographic cues hypothesis CLOGIT and BNL are the most advisable options in terms of elucidation of the regression coefficients; MNL and CONTRAST involve cumbersome interpretation and their fit to the theory is questionable.

Browse

Recent Submissions