Die Integration qualitativer und quantitativer Forschung – theoretische Grundlagen von „Mixed Methods“

Mixed Methods have been used in social research for quite a long time, as can be seen by looking on influential empirical studies from the first half of the 20^th century, where the combined use of qualitative and quantitative methods was justified in most cases only pragmatically (if at all). Unfortunately, the discussions led from the 1970s onwards about theoretical justifications of mixed methods designs often suffered from a focus limited on formal and epistemological issues. This holds true for the debates about the notion of “triangulation” as well as for the controversies about the paradigms presumably underlying quantitative, qualitative or mixed methods research. By contrasting “positivism”, “constructivism” and “pragmatism” outdated frontlines were often preserved and complex philosophical problems oversimplified. Furthermore, the focus on epistemology led to a neglect of substantial issues relating to the research field (which are always critical for the use of methods).The necessity for using (and combining) quantitative and qualitative methods follows from the nature of social structures relevant for the concrete field of investigation which may differ widely according to their stability and flexibility. Thereby structures with limited scope or “contingent structures” create specific problems for qualitative and quantitative methods which can be detected and dealt with by drawing on the strengths of the respective other methodological tradition in mixed methods designs.     

Datenanalyse in der Mixed-Methods-Forschung

This article discusses data analysis in mixed methods approaches. The challenge of integration during the stage of analysis has recently become a main point of the discourse on mixed methods that for more than a decade had been centered on topics of research design and compatibility of paradigms. The article focuses on key aspects of integration strategies as well as on points of the integration, which depend largely on the research design of the project. Ten different integration strategies in three areas are presented in detail, namely result-based, data-based and sequence-oriented integration strategies. Joint displays, which jointly present data and/or results of the qualitative and the quantitative research strands a particularly important.     

Seat Biases in Proportional Representation Systems with Thresholds

In proportional representation systems, apportionment methods are used to convert the number of votes of a party into the number of seats allocated to this party. An interesting characteristic of any such method are the seat biases, that is, the expected differences between the actual seat allocation and the ideal share of seats, separately for each party, when parties are ordered from largest to smallest. For electoral systems with a threshold, that is, with a minimum percentage of votes that parties must reach in order to be eligible to participate in the apportionment process, we show that seat biases decrease from their maximum to zero, as the threshold increases from zero to its maximum, and that all seat biases decrease linearly.     

Ethical inequality measures and the redistribution of income when needs differ

The paper considers nested social welfare functions and ethical inequality measures when households may differ in needs. Several principles of transfers between different household types are introduced and systematically examined. Their implications for the form of welfare functions and inequality measures are derived. The corresponding classes are completely described.      

Bias Correction in Generalized Nonlinear Models with Dispersion Covariates

We derive general formulae for the second-order biases of maximum likelihood estimates of the parameters in generalized nonlinear models with dispersion covariates. This result generalizes previous work by Botter and Cordeiro (1998 Botter , D. A. , Cordeiro , G. M. ( 1998 ). Improved estimates for generalized linear models with dispersion covariates . J. Statist. Comput. Simul. 62 : 91 – 104 .[Taylor & Francis Online], [Web of Science ®] , [Google Scholar]) and Cordeiro and McCullagh (1991 Cordeiro , G. M. , McCullagh , P. ( 1991 ). Bias correction in generalized linear models . J. Roy Statist. Soc. B 53 : 629 – 643 . [Google Scholar]). The practical use of such bias corrections is illustrated in a simulation study.     

Regardless of where they give birth,women living in non-metropolitan areas are less likely to have an epidural than their metropolitan counterparts

QuestionCan differences in Australian birth intervention rates be explained by women's residence at the time of childbearing?.MethodsData were collected prospectively via surveys in 1996, 2000, 2003, 2006 and 2009 from women, born between 1973 and 1978, of the Australian Longitudinal Study on Women's Health. Analysis included data from 5886 women who had given birth to their first child between 1994 and 2009. Outcome measures were self-report of birth interventions: pharmacological pain relief (epidural and spinal block analgesia, inhalational analgesia and intramuscular injections), surgical births (an elective or emergency caesarean section) and instrumental births (forceps and ventouse).FindingsPrimiparous women residing in non-metropolitan areas of Australia experienced fewer birth interventions than women residing in metropolitan areas: 43% versus 56% received epidural analgesia; 8% versus 11% had elective caesarean sections; and 16% versus 18% had emergency caesarean sections. Differences in maternal age and private health insurance status at first birth accounted for differences in surgical birth rates but did not fully explain differences in epidural analgesia.ConclusionNon-metropolitan women had fewer birth interventions, particularly epidural analgesia, than metropolitan women. Differences in maternal age and private health insurance do not fully explain the differences in epidural analgesia rates, suggesting care provided to labouring women may differ by area of residence. The difference in epidural analgesia rates may be due to lack of choice in maternity services, however it could also be due to differing expectations leading to differences in birth interventions for primiparous women living in metropolitan and non-metropolitan areas.     

Social welfare,inequality, and poverty when needs differ

The paper examines income distributions of a finite population consisting of households which may differ with respect to needs. Since observed incomes are not directly comparable, income distributions have to be adjusted. Incomes are transformed to equivalent incomes interpreted as living standards and measured for a reference type, and the latter are supplemented by weights depending on needs. A general class of social welfare orderings (being based on adjusted rank-ordered income distributions) is characterized by a set of properties. Severe limitations for the form of the adjustment process are implied. The consequences for the measurement of inequality and poverty are demonstrated, and corresponding orderings are derived.