Das unterschätzte Potenzial hoher Fallzahlen - Stärken und Limitierungen des Mikrozensus am Beispiel von Fertilitätsanalysen
Titelübersetzung:The unrecognised potential of large N - strengths and limitations of the German Microcensus exemplified by fertility research
Autor/in:
Bujard, Martin; Dorbritz, Jürgen; Herter-Eschweiler, Robert; Lux, Linda
Quelle: Zeitschrift für Familienforschung, 27 (2015) 3, S 343-372
Inhalt: "Aufgrund der hohen Fallzahlen (688.931 Personen in 2012) ist das Potenzial des Mikrozensus (MZ) für demografische und familiensoziologische Analysen groß, zudem inhaltlich durch die Erhebung der Geburten von Frauen noch gestiegen. Der Beitrag hat zum Ziel, methodische Hintergründe, Limitierungen und Potenziale des MZ am Beispiel von Fertilität systematisch aufzuzeigen. Die Analyse von Antwortverweigerungen zeigt, dass die Art der Befragung einen starken Einfluss hat, deren Verzerrungen durch Imputationen deutlich reduziert werden konnten. Die Limitierungen sind v.a. das Querschnittsdesign sowie die fehlenden Daten zur Kinderzahl von Männern und zur Binnenmobilität. Die hohen Fallzahlen des MZ ermöglichen tiefenscharfe Differenzierungen bezüglich Sozialstruktur, Paritäten und Zeitverläufe, was anhand von Konfidenzintervallen diskutiert wird. Drei Forschungsdesigns, die die Analysepotenziale nutzen, werden vorgestellt: Dendrogrammanalysen ermöglichen gruppenspezifische Differenzierungen nach fünf Ebenen und zeigen so ein detailliertes Muster der Kinderlosigkeit. Der Verlauf von Paritätsmustern im Kohortenvergleich verdeutlicht die je nach beruflicher Bildung unterschiedlichen Verläufe. Die Analysen zur altersspezifischen Häufigkeit von Erstgeburten ermöglichen, Trends von nachholenden Geburten für verschiedene Bildungsgruppen zu identifizieren." (Autorenreferat)
Inhalt: "Due to a large number of cases (688.931 persons in 2012) the German Microcensus has a huge potential for analyses in demography and family sociology. Content is further improved by the new voluntary question on biological children of women. This paper aims to systematically assess the methodological background as well as limitations and potentials of the Microcensus using the example of fertility data. The analyses of nonresponse show that the way of questioning has a strong impact. The related bias could, however, be reduced by imputations. The main limitations are the cross-sectional design and missing data on the children of men and internal migration. The potential of the high case numbers of the Microcensus is systematically discussed by analyzing confidence intervals: It allows for in-dept differentiations for analyzing social structure, birth parities and time patterns. Finally, we discuss three research designs which are able to use this potential: Dendrogram analyses allow group differentiations for up to five levels and thus reveal detailed patterns of childlessness. Parity patterns by cohorts illustrate differences by occupational education. Third, the age-specific analyses of first births among several cohorts allow for identifying trends of recuperation which differ considerable between educational groups." (author's abstract)
Schlagwörter:Mikrozensus; microcensus; Antwortverhalten; response behavior; Befragung; survey; Interview; interview; Datengewinnung; data capture; Schätzung; estimation; Kinderlosigkeit; childlessness; Fruchtbarkeit; fertility; Geburtenentwicklung; birth trend; Sozialstruktur; social structure; soziale Faktoren; social factors; demographische Faktoren; demographic factors; sozioökonomische Faktoren; socioeconomic factors; Bildungsniveau; level of education; Federal Republic of Germany; Imputationen
SSOAR Kategorie:Erhebungstechniken und Analysetechniken der Sozialwissenschaften, Familiensoziologie, Sexualsoziologie, Bevölkerung
The impact of method bias on the cross-cultural comparability in face-to-face surveys among ethnic minorities
Autor/in:
Kappelhof, Joost W. S.
Quelle: Methods, data, analyses : a journal for quantitative methods and survey methodology (mda), 8 (2014) 1, S 79-118
Inhalt: "This article investigates the impact of several sources of method bias on the cross-cultural comparison of attitudes towards gender roles and family ties among non-Western minority ethnic groups. In particular, it investigates how interviewer effects, the use of an interviewer with a shared ethnic background, interview language, interviewer gender, gender matching, the presence of others during the interview and differences in socio-demographic sample composition of non-Western minority ethnic groups affect the cross-cultural comparison of attitudes towards gender roles and family ties between these groups. The data used in this study come from a large scale face-to face survey conducted among the four largest non-Western minority ethnic groups in The Netherlands for which Statistics Netherlands drew a random sample of named individuals from each of the four largest non-Western minority populations living in The Netherlands. Furthermore, methods are introduced to estimate the potential impact of method bias on cross cultural comparisons. The results show that measurement of both gender roles and family ties constructs are full scalar invariant across the different ethnic groups, but that observed differences in attitudes between ethnic groups especially towards gender roles are influenced by method bias. This in turn leads to biased comparisons between ethnic groups because of differences in the size of the various sources of method bias, the differential impact of the same method bias between ethnic groups and the combination thereof." (author's abstract)
Schlagwörter:ethnische Gruppe; ethnic group; Minderheit; minority; Befragung; survey; Interview; interview; Geschlechtsrolle; gender role; Familie; family; Niederlande; Netherlands; soziale Integration; social integration; kulturelle Integration; cultural integration; interkultureller Vergleich; intercultural comparison; Forschungsreaktivität; reactivity effect; Antwortverhalten; response behavior; Umfrageforschung; survey research; methods bias; non-Western ethnic minorities; incomparability of samples; multi group Mimic
SSOAR Kategorie:Erhebungstechniken und Analysetechniken der Sozialwissenschaften
Testing the Validity of Gender Ideology Items by Implementing Probing Questions in Web Surveys
Autor/in:
Behr, Dorothée; Braun, Michael; Kaczmirek, Lars; Bandilla, Wolfgang
Quelle: Field Methods, 25 (2013) 2, S 124-141
Inhalt: This article examines the use of probing techniques in web surveys to identify validity problems of items. Conventional cognitive interviewing is usually based on small sample sizes and thus precludes quantifying the findings in a meaningful way or testing small or special subpopulations characterized by their response behavior. This article investigates probing in web surveys as a supplementary way to look at item validity. Data come from a web survey in which respondents were asked to give reasons for selecting a response category for a closed question. The web study was conducted in Germany, with respondents drawn from online panels (n = 1,023). The usefulness of the proposed approach is shown by revealing validity problems with a gender ideology item.
Schlagwörter:Befragung; survey; Online-Befragung; online survey; Gender; gender; Ideologie; ideology; Validität; validity; Panel; panel; Antwortverhalten; response behavior; Fehler; error; Deutungsmuster; pattern of interpretation; Interpretation; interpretation; Analyse; analysis; Indikator; indicator; Egalitarismus; egalitarianism; Rollenverteilung; role distribution; Rollenverständnis; role conception; ISSP; ISSP; Federal Republic of Germany; probing; web survey; gender ideology; ZA3880: International Social Survey Programme: Family and Changing Gender Roles III - ISSP 2002
SSOAR Kategorie:Erhebungstechniken und Analysetechniken der Sozialwissenschaften
Quelle: Wissenschaftszentrum Berlin für Sozialforschung gGmbH; Berlin (Discussion Papers / Wissenschaftszentrum Berlin für Sozialforschung, Forschungsschwerpunkt Zivilgesellschaft, Konflikte und Demokratie, Abteilung Migration, Integration, Transnationalisierung, SP IV 2011-703), 2011. 261 S
Inhalt: Zum Zusammenhang zwischen ethnischer Heterogenität, Sozialkapital (Vertrauen, freiwilliges Engagement, Mitgliedschaft in Organisationen und Vereinen etc.) und der Bereitstellungöffentlicher Güter lassen sich aus bislang vorliegenden Studien keine eindeutigen Befunde ableiten. Dies gilt insbesondere für den europäischen Kontext. Das Projekt Ethnische Vielfalt, soziales Vertrauen und Zivilengagement, gefördert vom Bundesministerium für Familie, Senioren, Frauen und Jugend, hatte zum Ziel, dieses Thema durch den Vergleich von drei Ländern, von unterschiedlichen Regionen und Städten in nationalem Kontext sowie durch Untersuchungen auf lokaler Ebene in Schulen besser zu beleuchten. Zur Datengewinnung wurden drei unterschiedliche methodische Ansätze miteinander verbunden: Im Rahmen der Umfragestudie Ethnic Diversity and Collective Action Survey (EDCAS) wurden insgesamt 10 200 Einwohner in 74 ausgewählten Regionen Deutschlands, Frankreichs und der Niederlande befragt. Die Daten ermöglichen den Vergleich von Städten, die sich nach dem Ausmaß ethnischer Diversität sowie in ihren integrationspolitischen Ansätzen voneinander unterscheiden. Um Einblick in die Prozesse zu gewinnen, die hinter den Effekten ethnischer Diversität stehen, wurden Fallstudien zur Elternpartizipation in Berliner und Lyoner Schulen organisiert. Darüber hinaus wurden Online- und Feldexperimente mit Anwohnern durchgeführt, um die kausalen Mechanismen dieser Effekte zu ergründen. Die hier dargelegten Ergebnisse beziehen sich auf ausschließlich auf die deutschen Daten. Bezogen auf die kognitive Dimension von Sozialkapital haben die Projektergebnisse den erwarteten negativen Zusammenhang zwischen objektiver, d.h. statistisch ausgewiesener ethnischer Vielfalt einerseits sowie Kooperationsvermögen und Vertrauen andererseits bestätigt. Ferner stellte sich heraus, dass auch die subjektive Wahrnehmung ethnischer Diversität einen eigenen Einfluss jenseits objektiver Diversitätsmaße ausübt. Mittels besonderer Betonung von Diversität in Experimenten konnte gezeigt werden, dass als hoch wahrgenommene ethnische Vielfalt die kausale Ursache für den Rückgang von Vertrauen ist. Im Hinblick auf die strukturelle Dimension von politischer Beteiligung und Mitgliedschaften in Organisationen sind die Ergebnisse jedoch weniger eindeutig: Zwar konnten wir mit einem Briefwurf-Experiment belegen, dass ethnische Diversität negative Auswirkungen auf tatsächliches pro-soziales Verhalten haben kann. In Bezug auf Engagement in Organisationen oder Beteiligung an Demonstrationen ließen sich jedoch keine signifikanten Effekte objektiver oder wahrgenommener Diversität nachweisen. Entgegen Putnams Deutung, dass Vielfalt zu einem Rückzug ins Private führe, gelangen wir zu der Schlussfolgerung, dass Diversität zwar Vertrauen und pro-soziales Verhalten beeinträchtigt, Menschen aber auch dazu aktivieren kann, der mangelnden Bereitstellung öffentlicher Güter in diversen Gemeinschaften organisiert entgegenzuwirken. (Autorenreferat)
Inhalt: Existing evidence on the relationship between ethnic heterogeneity, social capital (trust, voluntarism, associational membership etc.), and levels of public goods provision is inconclusive, especially for the European context. The project Ethnic Diversity, Social Trust, and Civic Engagement, financed by the German Federal Ministry of Family Affairs, Senior Citizens, Women and Youth, was intended to advance our knowledge of this relationship by conducting a comparative analysis across three countries, across regions and cities, as well as in the local organisational context of school communities, and using three complementary methodological approaches. First, the Ethnic Diversity and Collective Action Survey (EDCAS) across local populations in 74 selected regions in Germany, France, and the Netherlands with a total sample size of 10 200 allowed comparisons of cities with different levels of ethnic diversity and different policy approaches to deal with immigration. Second, case studies on parental participation in schools in Berlin and Lyon provided qualitative insight into the processes behind ethnic diversity effects. Finally, survey, online and field experiments with local residents were conducted to investigate the causal mechanisms behind these effects. Results presented in the paper refer only to the German data. (author's abstract)
Schlagwörter:Sozialkapital; social capital; Ethnologie; ethnology; Diversität; diversity; Vertrauen; confidence; bürgerschaftliches Engagement; citizens' involvement; regionale Integration; regional integration; Befragung; survey; Experiment; experiment; Fallstudie; case study; Federal Republic of Germany; Frankreich; France; Niederlande; Netherlands; internationaler Vergleich; international comparison; Integrationspolitik; integration policy
SSOAR Kategorie:Ethnologie, Kulturanthropologie, Ethnosoziologie, Erhebungstechniken und Analysetechniken der Sozialwissenschaften
Personalisation in advance letters does not always increase response rates: demographic correlates in a large scale experiment
Titelübersetzung:Personalisierung in Voraus-Briefen erhöhen nicht immer die Antwortraten: demographische Korrelationen in einem Großversuch
Autor/in:
Luiten, Annemieke
Quelle: Survey Research Methods, 5 (2011) 1, S 11-20
Inhalt: "This study was set up to examine whether personalizing advance letters by adding names and appropriate salutation, affects the survey cooperation of subgroups in the general population differently, in analogy to findings that subgroups react differently to advance letters. Differential reactions could be an explanation for the mixed findings in the literature on personalization of advance or cover letters. In a large scale study (N=39,518), information in communal registries was used to study (non)response patterns in subgroups, as a result of personalization. Advance letters of the Dutch CAPI Labor Force Survey were randomly assigned to addresses. In the non-personalized version (N=30,899), letters were addressed to 'the inhabitants of '. In the personalized version (N=8,619), the name or names of the household core were derived from municipal registries and used for addressing the letter. A re-interview addressed the issue whether the advance letter was read by more households when the household was addressed by name. By linking the sample to registries, it was possible to study response behavior of subgroups. The study focused on groups the literature indicates as differentially reacting to advance letters, i.e., different age, ethnic, gender, household composition and income groups, and groups with or without a listed telephone number. Hardly any difference in the overall level of cooperation was found if either a personalized or non-personalized letter was used. However, differential reactions were found for listed versus unlisted telephone owners, where only listed households reacted positively to personalization. In the other subgroups studied, no firm evidence of differential reactions were found. The paper discusses what these results signify for sample composition and the risk of bias." (author's abstract)
Sozial erwünschtes Antwortverhalten bezüglich der Teilung häuslicher Arbeit: die Interaktion von Interviewergeschlecht und Befragtenmerkmalen in Telefoninterviews
Titelübersetzung:Social desirability and response bias in case of the division of household labour: the interaction between gender of the interviewer and respondent characteristics in telephone interviews
Autor/in:
Klein, Markus; Kühhirt, Michael
Quelle: Methoden, Daten, Analysen (mda), 4 (2010) 2, S 79-104
Inhalt: "Der Beitrag untersucht den Einfluss des Interviewergeschlechts auf das Antwortverhalten von Befragten bezüglich der Teilung häuslicher Arbeit in telefonischen Interviews. Aufgrund einer Diskrepanz zwischen öffentlich vertretenen egalitären Geschlechterrollen und der noch immer geringen männlichen Beteiligung an der Hausarbeit wird auf Basis von Rational-Choice Argumenten erwartet, dass männliche Befragte bei weiblichen Interviewern sozial erwünscht antworten und ihre Beteiligung an der Hausarbeit überschätzen. Dagegen ist anzunehmen, dass Frauen ihren relativen Anteil an der partnerschaftlichen Hausarbeit eher unterschätzen. Interviewereffekte des Geschlechts sollten bei weiblichen Befragten jedoch geringer ausfallen als bei Männern. Da sich vor allem junge bzw. gebildete Befragte egalitärer Rollenvorstellungen bewusst sind, sollten vor allem diese Gruppen anfällig für derartige Interviewereffekte sein. Für männliche Befragte entsprechen die Ergebnisse weitestgehend den Erwartungen, wobei das Interviewergeschlecht in Abhängigkeit vom Alter des Befragten die stärkeren Effekte zeigt. Für weibliche Befragte zeigen sich über alle Altersgruppen und Bildungsabschlüsse dagegen keine signifikanten Effekte. Die Ergebnisse verdeutlichen, dass sozial erwünschtes Antwortverhalten auch bei Auskünften über konkretes Alltagshandeln auftreten kann. Hinzu kommt eine beträchtliche Heterogenität der Effekte über unterschiedliche Bevölkerungsgruppen." (Autorenreferat)
Inhalt: "In this paper the authors examine the impact of the interviewer's gender on respondents' self-reported share of housework in telephone interviews. Due to a discrepancy between modern, egalitarian gender roles dominating public discussion and men's marginal participation in housework they expect male respondents to answer in a socially desirable way and exaggerate their share of housework vis à vis female interviewers. At the same time, they assume that female respondents underreport their contribution to the couple's housework to female interviewers. The effects of the interviewer's gender should be less strong in case of female respondents than in case of male respondents, though. Additionally, theory suggests that young and educated respondents are particularly susceptible to effects of the interviewer's gender as they are most aware of egalitarian gender roles due to their socialization and environment. For male respondents the results are largely consistent with the expectations. However, the effect of the interviewer's gender varies stronger by age than by education. Regarding female respondents the results do not indicate any significant effects across age groups and educational degrees. Overall, the results show that social desirability may also bias self-reports of everyday behaviour. Moreover, interviewer effects vary considerably between different social groups." (author's abstract)
Reasons for refusals, their collection in surveys and interviewer impact
Autor/in:
Menold, Natalja; Züll, Cornelia
Quelle: GESIS - Leibniz-Institut für Sozialwissenschaften; Mannheim (GESIS-Working Papers, 2010/11), 2010. 35 S
Inhalt: "Refusals are a significant source of non-response in surveys. During field periods of some surveys reasons for refusals are collected in call record data (as part of para-data). This article presents a study employing a content analysis of open-ended comments on reasons for refusals collected by interviewers in a survey of the German population (ALLBUS). We analysed the reasons for refusals contained in these comments, as well as to what extent these comments include information about factors relevant to participation in surveys. Additionally, we analysed the impact of interviewer characteristics – gender, age, education and experience – on data collection using various multilevel multinomial models. The results show that interviewer comments provide typical reasons for refusals, as well as specific information about target persons, their environment and the survey process. Interviewers' age and education influenced the collection of reasons for refusals. At the same time interviewer variances (obtained through multinomial multilevel models) were very high, showing that interviewers prefer to report certain reasons for refusals. The highest interviewer level variances were obtained for providing no comments at all. To improve data quality and reduce high interviewer impact, we suggest using
improved standardised instruments to collect reasons for refusals. Codings based on a categorisation scheme which we developed for our content analysis show high reliability (kappa = .81). Thus, this scheme can be used as a basis for developing such standardised instruments." (author's abstract)
Autor/in:
Bollinger, Christopher R.; Hirsch, Barry T.
Quelle: Rat für Sozial- und Wirtschaftsdaten (RatSWD); Berlin (RatSWD Working Paper Series, 165), 2010. 22 S
Inhalt: "Earnings nonresponse in the Current Population Survey is roughly 30% in the monthly surveys and 20% in the annual March survey. Even if nonresponse is random, severe bias attaches to wage equation coefficient estimates on attributes not matched in the earnings imputation hot deck. If nonresponse is ignorable, unbiased estimates can be achieved by omitting imputed earners, yet little is known about whether or not CPS nonresponse is ignorable. Using sample frame measures to identify selection, we find clear-cut evidence among men but limited evidence among women for negative selection into response. Wage equation slope coefficients are affected little by
selection but because of intercept shifts, wages for men and to a lesser extent women are understated, as are gender wage gaps. Selection is less severe among household heads/co-heads than among other household members." [author's abstract]