Günter L. Huber

Computer-aided qualitative data analysis : theory, methods and practice

Introduction - Udo Kelle An Overview of Computer-Aided Methods in Qualitative Research PART ONE: GENERAL METHODOLOGICAL ISSUES Computer Use in Qualitative Research and Issues of Validity - Udo Kelle and Heather Laurie User Experiences of Qualitative Data Analysis Software - Raymond M Lee and Nigel G Fielding Grounded Theory as an Emerging Paradigm for Computer-Assisted Qualitative Data Analysis - Marrku Lonkila Different Functions of Coding in the Analysis of Textual Data - John Seidel and Udo Kelle PART TWO: COMPUTERS AND QUALITATIVE THEORY BUILDING Introduction: Using Linkages and Networks for Theory Building - Gerald Prein and Udo Kelle in discussion with Lyn Richards and Tom Richards Reducing Fragmentation in Qualitative Research - Ian Dey Using Hierarchical Categories in Qualitative Data Analysis - Tom Richards and Lyn Richards Designing and Refining Hierarchical Coding Frames - Luis Araujo PART THREE: COMPUTERS AND QUALITATIVE HYPOTHESIS EXAMINATION Introduction - Udo Kelle in discussion with Ernest Sibert, Anne Shelly, Sharlene Hesse-Biber and G[um]unter L Huber Hypothesis Examination in Qualitative Research Using Logic Programming for Hypothesis Generation and Refinement - Ernest Sibert and Anne Shelly Hypothesis Testing in Computer-Aided Qualitative Data Analysis - Sharlene Hesse-Biber and Paul Dupuis Qualitative Hypothesis Examination and Theory Building - G[um]unter L Huber PART FOUR: COMPUTERS AND TRIANGULATION Introduction - Gerald Prein and Udo Kuckartz in discussion with Edeltraud Roller, Charles C Ragin and Udo Kelle Between Quality and Quantity Case-Oriented Quantification - Udo Kuckartz Hermeneutic-Classificatory Content Analysis - Edeltraud Roller, Rainer Mathes and Thomas Eckert Using Qualitative Comparative Analysis to Study Configurations - Charles C Ragin An Overview of Software - Gerald Prein, Udo Kelle and Katherine Bird

Media Violence and Youth Violence: A 2-Year Longitudinal Study

The frequency of exposure to media violence and eight additional risk factors were path-analyzed in a 2-year longitudinal study. The exposure to media violence (total score) affected students’ later violence (β = .28) and later violent delinquency (β = .30) more strongly than other risk factors. Direct effects were also caused by risk factors assessed at Time 1, which in turn were reinforced by the remaining risk factors on the second or third stratum of analysis. Of particular importance are the findings that (1) playing violent electronic games is the strongest risk factor of violent criminality and (2) both media-stimulated and real experiences of aggressive emotions associated with the motive of revenge are core risk factors of violence in school and violent criminality. The results of our study show that the more frequently children view horror and violence films during childhood and the more frequently they play violent electronic games at the beginning of adolescence the higher will these students’...

Uncertainty orientation and cooperative learning: individual differences within and across cultures

Abstract Four studies are presented with different samples of students (grade school, high school, university, teachers) from different cultures (Canadian, German, Iranian). Across these samples it was found that, when asked to state their learning preferences, all samples preferred cooperative learning to other types of learning. However, this preference was greater for uncertainty-oriented persons than certainty-oriented persons as predicted by the theory of uncertainty orientation (Sorrentino & Short 1986). Also as predicted, when actually placed in different learning situations, whereas uncertainty-oriented students had a greater preference and performed better in a cooperative situation (the Jigsaw) than a traditional expository situation, certainty-oriented students were more negative and performed worse in the cooperative than traditional learning situation. Student teachers showed similar preferences as a function of their uncertainty orientation. Implications for advocates of cooperative learning, as well as other forms that do not account for individual differences, are discussed.

Computer assistance for testing hypotheses about qualitative data: The software package AQUAD 3.0

The analysis of qualitative data seems to be inevitably bound to tedious and time-consuming mechanical work, especially if we proceed to more complex analytical procedures like testing hypotheses. In this article we describe the general tasks of qualitative analysis, i.e., reducing data, structuring findings, and drawing conclusions. Then we outline, referring to the software package AQUAD 3.0, how software facilitates the work of qualitative researchers. By describing a practical application of this software, basic features of computer assistance to hypotheses testing are explained.

The ambiguous use of language in the paradigms of QUAN and QUAL

Whether they teach or apply quantitative (QUAN) or qualitative (QUAL) methodology, textbooks as well as research reports often use ambiguous language in the sense that many formulations cannot be assigned unequivocally to one of the main methodological orientations. An analysis of research reports on the linguistic level shows that QUAN and QUAL paradigms cannot be separated strictly on the level of language. The use of language consolidates the strands of research activities and, therefore, the logic and hidden meanings behind scientific conclusions are revealed. QUAL studies often quantify at least their findings, and QUAN approaches implicitly introduce concepts, ideas and strategies from the field of QUAN. Examples from empirical research and the teaching of statistics demonstrate this ambiguity. Supposing that both approaches actually do not represent isolated paradigms but rather more or less clearly identified steps in the research process, we compare the essential criteria of quality in the domains of QUAN and QUAL. A common ground for both paradigms is identified within basic assumptions from measurement theory and its associated criteria of quality. To make these interrelations more explicit, stronger efforts should be undertaken to document the inputs, the outputs and the methods applied during the various stages of the research process.

Cooperative learning in German schools

Abstract Despite decades of appreciation of CL in scientific discussions of education and in curriculae, it is a rare event in the average German classroom. Rather than looking for external circumstances impeding the realization of a promising idea, this chapter focuses on internal conditions of dissemination strategies. The discussion offers the optimistic outlook that discrepant orientations toward CL may be resolved.

Computerunterstützung von Reduktions-und Konklusionsprozessen bei der Interpretation von Texten

In den Sozialwissenschaften ist man haufig darauf angewiesen, bestimmte Sach-verhalte, Zusammenhange, Probleme usw. aus der subjektiven Sicht der beteiligten oder betroffenen Personen zu erschliesen. Standardisierte Erhebungsinstrumente scheiden fur diesen Zugang definitionsgemas aus. Die wissenschaftliche Analyse von “Alltagstheorien”, von “impliziten Theorien”, von “oral history”, um nur einige der gebrauchlichen Bezeichnungen fur den Forschungsgegenstand zu nennen, geht von qualitativen Daten aus. In der Regel liegen sie als verbale Daten vor, d.h. als Texte, die in Interviews, Tagebuchern, Protokollen “Lauten Denkens”, “Feldnotizen” teilnehmender Beobachter usw. produziert wurden. Die Datenmengen, die sich so ergeben, sind enorm. Bei einer Interviewstudie mit 100 Gesprachspartnern waren 2.000 Seiten Transkriptionen im DIN A4 Format eher im unteren Bereich des Durchschnitts einzuordnen. Alle diese Seiten enthalten Aussagen, welche die individuellen Perspektiven der Gesprachspartner in farbigen, vielgestaltigen, oft schon innerhalb eines Protokolls mehrdeutigen sprachlichen Wendungen abbilden.

Computergestützte Analysen in der qualitativen Sozialforschung — mehr als Wörter suchen und zählen

Den Ansatzen der qualitativen Methodologie wird haufig mangelnde Scharfe und Prazision ihrer Verfahren vorgeworfen. Daruber hinaus wird kritisiert, das in der Regel in den Arbeiten zur Methodologie uber das Vorgehen nach der Datenerhebung nur wenige konkrete Angaben zu finden sind. Miles [1] charakterisierte qualitative Daten daher auch gegenwartig noch zutreffend als “attraktives Argernis”. Bei der Anwendung quantitativer Methoden kann man in der Regel auf standardisierte Mesinstrumente zuruckgreifen und aus einer wohlsortierten Sammlung analytischer Prozeduren die jeweils passende begrundet auswahlen. Qualitative Forscher dagegen konnen sich nur auf ein breites Repertoire von Methoden zur Erhebung von Daten stutzen, was in ihrem Ansatz bedeutet, Zugange zur Einzigartigkeit spezifischer sozialer Settings oder zur Subjektivitat persoinlicher Weltsichten zu erschliesen. Dann aber sind qualitative Forscher ziemlich allein mit einem schon physisch fast uberwaltigenden Berg facettenreicher, farbiger, meist verbaler Daten und mit ihrem Einfallsreichtum oder ihrer Intuition, daraus klug zu werden. Aus qualitativen Daten “kIug werden” erfordert zunachst einmal, die meist wortreichen und redundanten Beschreibungen, Erklarungen, Rechtfertigungen, Feldnotizen, Beobachtungsprotokolle usw. auf eine uberschaubare Menge sinnvoller und vergleichbarer Bedeutungseinheiten zu reduzieren, dann diese Einheiten strukturiert, d.h. uberschaubar darzustellen und schlieslich urnfassendere Schlusse zu ziehen und zu uberprufen. Miles & Huberman [2] umreisen ihr Verstandnis von “Qualitativer Analyse” mit diesen drei Gruppen von Aktivitaten zur Datenauswertung, betonen aber gleichzeitig, das diese Techniken als Schritte des Forschungsprozesses in Wechselwirkung stehen und daher in der Regel nicht unabhangig voneinander oder einfach sequentiell abgearbeitet werden. Abhangig von den spezifischen Anforderungen des Forschungsprozesses wechseln sich Reduktionen, Datenaufbereitung und Konklusionen immer wieder ab und beeinflussen sich wechselseitig.

Verwendung von Computern in der qualitativen Sozialforschung

“Hilfe, ich ersticke in Texten!” So kennzeichnete Sudmersen (1983) ihre Situation bei der Auswertung Narrativer Interviews. Diese Situation durfte allen, die mit qualitativen Methoden arbeiten, gut vertraut sein aufgrund der Menge an Informationen, die dabei in relativ unstrukturierter Form als Daten anfallen. Solche Erfahrungen sind es aber auch, die den Ausgangspunkt fur den immer haufiger zu verzeichnenden Ruckgriff auf Computer bei qualitativer Forschung darstellen.