Quelle: Methods, data, analyses : a journal for quantitative methods and survey methodology (mda), 13 (2019) 1, S 59-90
Inhalt: Questions on earnings are counted among sensitive topics that often produce high rates of item nonresponse or measurement error. Both types of bias are well documented in the literature and are found to concentrate in the tails of the earnings distribution. In this paper, we explore whether measurement error on earnings could be explained by socially desirable
reporting and whether the error is impacted by interviewer characteristics. Using the linked dataset NEPS-SC6-ADIAB, which contains survey data from the German National Educational Panel Study, Starting Cohort "Adults", linked to administrative earnings records from the German Federal Employment Agency, we analyze the extents of over- and underreporting and the influence of respondent and interviewer characteristics on these behaviors for different quartiles of the earnings distribution. Our results show that the average level of misreporting is relatively low (approximately 6% of median earnings). Our main logistic model reveals that female and more highly educated respondents report significantly more accurately while those with higher earnings misreport to a significantly greater extent. Regarding the impact of personality traits on reporting accuracy, we find significant positive effects for more agreeable respondents and significant negative effects for extraverted respondents. When differentiating by the direction of misreporting, we find, for instance, that women are less likely to overreport across all earnings quartiles. However, the influence of interviewer characteristics is negligible.
Non-Randomized Response Models: An Experimental Application of the Triangular Model as an Indirect Questioning Method for Sensitive Topics
Autor/in:
Erdmann, Anke
Quelle: Methods, data, analyses : a journal for quantitative methods and survey methodology (mda), 13 (2019) 1, S 139-167
Inhalt: When it comes to sensitive questions, data is often affected by bias due to non-response or effects of social desirability. Several methods have been introduced to eliminate answer bias by using randomization processes and probabilistic theory to obscure the respondent’s answer and create anonymity, thus facilitating honest answers. The probably most traditional method is the Randomized Response Technique by Warner (1965). However, this method is loaded with certain disadvantages. Therefore, in the last decade, newer methods were introduced that aim at balancing the disadvantages and weaknesses of previous methods, for instance, the non-randomized models Crosswise Model and Triangular Model (Yu et al. 2008) as well as the Parallel Model (Tian 2014). Although especially the Triangular Model is easy to implement in a study, there is only little empirical evidence on its application in different survey modes and populations. Further, it is to assume that certain questions are not equally sensitive for everybody due to specific personal characteristics. Thus, indirect questioning might not be effective in general but only for certain populations. The present study extends prior work on the Triangular Model by evaluating it for different subgroups. The conducted experiment asks for sensitive characteristics in the context of mental stress among students. The Triangular Model achieves significantly higher percentages than conventional direct questioning for illegal drug use among persons that answer socially desirable according to the characteristic of Self-Deception. For the other analyzed subgroups (Impression Management, gender, and depressiveness), the Triangular Model could not achieve higher prevalence rates compared to direct questioning on a sufficient probability level. But still, hard evidence on the effectiveness of indirect questioning models is thin and further critical discussion is needed.
Schlagwörter:Umfrageforschung; survey research; Datengewinnung; data capture; Datenqualität; data quality; Antwortverhalten; response behavior; soziale Erwünschtheit; social desirability; Anonymität; anonymity; psychische Gesundheit; mental health; Drogenkonsum; drug use; Triangular Model; Indirect Questioning; Survey Methodology; Non-Randomized Response
SSOAR Kategorie:Erhebungstechniken und Analysetechniken der Sozialwissenschaften
How to weight survey data with a dyadic multi-actor design?
Autor/in:
Pasteels, Inge
Quelle: Survey Methods: Insights from the Field, (2015) , 11 S
Inhalt: This paper deals with adjustment for nonresponse in dyadic multi-actor survey designs. It presents a multi-dimensional approach to
weighting that addresses the various analytical units represented in such data, so that sampling design weights are correctly accounted
for and so that consistency between weights is achieved. This approach is demonstrated by using the primary respondents in the Divorce
in Flanders study, which is a typical example of a dyadic multi-actor design. Five sets of weighting coefficients are made available
whereby different subsets of data, according to different analytical units, are selected: the subset of the dyads, the subset of men and
women respectively, and two subsets of marriages. Post-stratification – with the year of marriage, status of the reference marriage at the
sampling date and five-year divorce cohort as auxiliary variables – was chosen as the weighting adjustment technique.
Schlagwörter:Gewichtung; weighting; Antwortverhalten; response behavior; Stichprobe; sample; soziale Schichtung; social stratification; Analyseverfahren; analysis procedure; Zufallsauswahl; random sample; Umfrageforschung; survey research
SSOAR Kategorie:Erhebungstechniken und Analysetechniken der Sozialwissenschaften
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
Titelübersetzung:Social presence in online surveys
Autor/in:
Schmidt-Catran, Alexander W.; Hörstermann, Katharina
Quelle: Methoden, Daten, Analysen (mda), 7 (2013) 3, S 397-432
Inhalt: "Onlinebefragungen zählen mittlerweile zum Standardrepertoire vieler Forschungsinstitute, einige methodische Fragen bleiben bis dato jedoch noch unbeantwortet. Diese Studie beschäftigt sich mit der Frage, ob durch die Simulation sozialer Präsenz das Antwortverhalten der Befragten beeinflusst werden kann. Mit einem experimentellen Design werden Effekte der sozialen Präsenz auf Nonresponse sowie sozial erwünschtes Antwortverhalten untersucht. Dabei wird davon ausgegangen, dass die Implementierung sozialer Präsenz den Vorteil einer erhöhten Teilnahmebereitschaft und den Nachteil von sozial erwünschtem Antwortverhalten mit sich bringen kann. Der inhaltliche Schwerpunkt des Experiments liegt auf Einstellungen zu Sexismus, Geschlechterrollen und der Berufstätigkeit von Frauen. Zur Simulation der sozialen Präsenz werden Fotografien verwendet. Zusätzlich zum Effekt der sozialen Präsenz wird untersucht, ob das Geschlecht und die Attraktivität der auf den Fotos gezeigten Personen einen Einfluss auf das Antwortverhalten haben. Bei diesem Effekt handelt es sich um eine Art 'Gender-of-Interviewer'-Effekt. Bezüglich der Teilnahme- und Antwortbereitschaft konnten keine positiven Effekte nachgewiesen werden. Die Ergebnisse hinsichtlich möglicher Interviewereffekte in Form von sozial erwünschtem Antwortverhalten sind nicht eindeutig, es tauchen aber systematische Unterschiede zwischen den Experimental- und der Kontrollgruppe auf." (Autorenreferat)
Inhalt: "Today, online surveys belong to the standard instruments of most survey research institutes, but some methodical questions are still unanswered. This study deals with the question of whether the simulation of social presence has an effect on the response style. Using an experimental design, the effects of social presence on non-response and socially desirable answers are tested. We expect social presence to lower non-response but to induce socially desirable answers. Topics of the survey are attitudes towards sexism, gender roles and the participation of women in the labor market. Social presence is simulated by pictures of the 'investigators' of the study. In addition to the effect of social presence, the study investigates whether the gender and the attractiveness of the person on the pictures have an effect on the answers. This effect might be called a 'gender-of-interviewer'-effect. Concerning the non-response rate no positive effect of social presence could be found. The results with regard to social desirability and a possible gender-of-interviewer-effect are ambiguous but show significant differences between the control group and our experimental groups." (author's abstract)
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)
Das standardisierte Interview als soziale Interaktion: Interviewereffekte in der Umfrageforschung
Titelübersetzung:The standardized interview as social interaction: interviewer effects in survey research
Autor/in:
Haunberger, Sigrid
Quelle: ZA-Information / Zentralarchiv für Empirische Sozialforschung, (2006) 58, S 23-46
Inhalt: 'Dieser Artikel beschäftigt sich mit methodischen Problemen, die beim standardisierten face-to-face Interview durch die Interaktion zwischen Interviewern und Befragten entstehen. Auf der Datengrundlage einer Mehrthemenumfrage (Sozialwissenschaftenbus 2/96) werden zu drei Fragetypen (Antwort auf offene Fragen, Antwortverweigerung bei der Einkommensabfrage und Antworten zu Meinungsfragen) mögliche Interviewereffekte analysiert. Für alle drei Fragetypen lassen sich bedenkenswerte Zusammenhänge von Interviewermerkmalen und der Situation des Interviews feststellen: Intervieweralter und Interviewerbildung beeinflussen die Antworten auf offene Abfragen. Bei der Verweigerung der Einkommensabfrage bestehen ebenfalls deutliche Zusammenhänge mit dem Interviewergeschlecht und der Interviewerbildung sowie mit der Anwesenheit Dritter. Analysiert wurde auch ein Interaktionseffekt von Interviewer- und Befragtenmerkmalen. In 'homogenen' Bildungskonstellationen wird weniger verweigert. Zustimmungs- und Ablehnungstendenzen bei Meinungsfragen zeigen einen deutlichen Geschlechtseffekt.' (Autorenreferat)
Inhalt: 'This article focusses on methodological problems in standardized oral interviews which arise from the interaction between interviewer and respondent. With the data from Sozialwissenschaften-Bus 2/96, three possible interviewer effects are analyzed (responses to open questions, nonresponse to the income question and opinion questions). For all three topics, considerable influences of interviewer characteristics and the interview situation could be found: The responses to open questions vary with the interviewer's age and education. In nonresponse to the income question, there is a clear relation between interviewer's sex and education as well as influence of present others. There was also an interaction effect of interviewer and respondent-characteristics. Homogeneous education-groups show less nonresponse. Tendencies of acceptance or refusal to questions of opinion are related to the sex of the interviewer.' (author's abstract)|