Conference Agenda

Session
RN21_01: Potentials and boundaries of analytical techniques
Time:
Wednesday, 21/Aug/2019:
11:00am - 12:30pm

Session Chair: Modesto Escobar, Universidad de Salamanca
Location: GM.326
Manchester Metropolitan University Building: Geoffrey Manton, Third Floor 4 Rosamond Street West Off Oxford Road

Presentations

A Phantom Menace: Random Data, Model Specification and Causal Inference in Qualitative Comparative Analysis

Alrik Thiem, Lusine Mkrtchyan

University of Lucerne, Switzerland

To date, the method of Qualitative Comparative Analysis (QCA) has been employed by hundreds of researchers. At the same time, the literature has long been convinced that QCA is prone to committing causal fallacies when confronted with random data. Specifically, beyond a certain case-to-factor ratio, QCA is believed not to be able to distinguish anymore between random and real data. In consequence, applied researchers relying on QCA for the analysis of their empirical data have worried that the explanatory models presented to them would be nothing but algorithmic artifacts. So as to minimize that risk, benchmark tables of boundary case-to-factor ratios have been proposed by Marx and Dusa (2011). We argue in this article that fears of inferential breakdown in QCA are unfounded as every set of data generated by any proper causal structure can always be reproduced by an isomorphic set of purely stochastic data. In this connection, we furthermore demonstrate that Marx and Dusa's benchmarks do not prevent but force QCA to commit causal fallacies. Ultimately, we argue that random data are a phantom menace which applied researchers need not worry about when designing their analyses with QCA.



Quantitative Grounded Theory – A Neglected Tool?

Peter Kevern, Edward Tolhurst

School of Health Studies, Staffordshire University, United Kingdom

Grounded theory is a methodology for structuring and interpreting research activity, irrespective of the type of data to be gathered. Despite this, ‘Quantitative Grounded Theory’ is an almost entirely neglected methodological tool. This neglect suggests two things: first, the effectiveness with which grounded theory has been ‘annexed’ by qualitative researchers, and secondly the degree to its distinctive contribution as a methodology has been lost because of its reduction to a lexicon and set of methods for constructivist qualitative research. This is reflected in the discrepancy between the abundance of methodological writing on qualitative grounded theory, and the limited number of studies that actually adhere closely to its key tenets.

Nevertheless, a strong case can be made for applying grounded theory to quantitative research. The distinctive strength of grounded theory is the way that pays close attention to the inductive phase of research i.e. processes of theory-building and the development and refinement of concepts which are frequently ‘bracketed out’ of accounts of quantitative research. Furthermore, because it aims only to generate working hypotheses for further testing and stops short of claiming to have arrived at robust findings, it has the potential to bridge the divide between university-led ‘clean and slow’ research and the needs of commissioners for ‘quick and dirty’ answers.



Interactive Graphs for Lineal and Logistic Regressions Networks

Modesto Escobar

University of Salamanca, Spain

Graphs have been employed not only to solve topographic problems and to represent social structures, but also to show the correlation between variables according to casual models. Path analysis and structural equations models are indeed well known by sociologists, but both were restricted to quantitative variables at their early stages. In this paper, we propose a new way to display the connections between qualitative variables in a similar way to the correspondence analysis, but using another set of multivariate techniques, such as lineal and logistic regression, mixed with network analysis.

The proposed representations are based on solving several equations and selecting only those coefficients with a significant positive relationship. By doing so, graphs are obtained selecting the categories with predicted proportions or means significantly greater than those of the population. Furthermore, to increase their analytic power, they have interactive characteristics, which include either the selection of the elements according to their size or attributes, and the filter of the most central and strongest links.

The first part of the paper deals with the statistical basis of these representations; the second proposes programs to make them possible, and the third gives examples of their use in international comparative surveys.



Measuring the Permeability of Class-Boundaries: An Information Theoretic Approach

Georg P. Mueller

Univ. of Fribourg, Switzerland

Mutual information (MI) is a concept from mathematical information theory, which describes the information-gain about a random-variable Y that results from knowledge about a related variable X. The concept is obviously of interest for intergenerational mobility research: in open societies with permeable class-boundaries the MI of the parental class-position X with regard to the status Y of their children is generally small, whereas in feudal societies with strict class-barriers the corresponding MI is generally higher. Thus MI is an indicator of the inequality of chances.

One of the advantages of MI is the possibility to decompose it by the class (a) of the parents and (b) of their children. The first case refers to the information of the class of origin about the child's destination: this is an indicator of the conservatism of the related class as a result of social heritage. In the second case we acquire knowledge about the information contained in the class of destination about the child's origin: this indicates the closedness of a class due to the exclusion of social newcomers.

In order to show the utility of the mentioned indicators, the author uses the European Values Study (2008) for an analysis of the intergenerational transfer of educational attainments in different countries. The high standardization of the data-source allows to compare countries with regard to conservatism and closedness of primary, secondary, and tertiary education. The analyses show for the three educational strata a considerable variation of closedness and conservatism within and between countries.