Comparing Self-Reported and Partnership-Inferred Sexual Orientation in Household Surveys

Abstract

Research comparing heterosexuals with bisexuals and homosexuals in economics and the social sciences typically relies on two strategies to identify sexual orientation in existing survey data of general populations. Probing respondents to self-report their sexual orientation is generally considered the preferred option. Since self-reports are unavailable in most large multidisciplinary surveys, often researchers infer sexual orientation from the gender-constellation of a respondent’s partnership instead. Based on German Socio-Economic Panel (SOEP) data, this article reviews both strategies empirically in the context of a household panel survey. The analysis shows that self-reported and partnership-inferred sexual orientation are not mutual substitutes, instead leading to substantively different conclusions about differences between heterosexuals and LGBs (Lesbian, Gays, and Bisexuals). The article discusses problems of non-coverage in partnership-inferred sexual orientation and also investigates measurement error in self-reported sexual orientation, finding notable mode and interviewer effects.

Political engagement and turnout among same-sex couples in Western Europe

“…Our measurement therefore relies on the gender of a survey respondent's current partner as a proxy of their sexual orientation. Similar measurement strategies have been used elsewhere using different datasets (Fischer et al, 2016) (for a discussion, see Kühne et al,2019) and observe comparable numbers of individuals in same-sex relationships, increasing the confidence in the validity of our approach. Moreover, given that interview effects can alter interviewees' willingness to self-report as LGB (Kühne et al, 2019), partnerinferred sexuality allows us to overcome these obstacles.…”

“…Similar measurement strategies have been used elsewhere using different datasets (Fischer et al, 2016) (for a discussion, see Kühne et al,2019) and observe comparable numbers of individuals in same-sex relationships, increasing the confidence in the validity of our approach. Moreover, given that interview effects can alter interviewees' willingness to self-report as LGB (Kühne et al, 2019), partnerinferred sexuality allows us to overcome these obstacles. The limitation of this operationalisation is that we can only include individuals in a co-habiting relationship in the analysis -that is, non-cohabiting individuals are removed from the sample to facilitate the isolation of sexuality-based divergence rather than divergence based on cohabitation.…”

“…We acknowledge the potential limitations of the analysis. Firstly, given that the measurement strategy relies on individuals being in a relationship, the extent to which the effects can be generalised across those who are and are not currently in a relationship is unclear (Kühne et al, 2019). Secondly, we are not able to distinguish between those who are bisexual and those who are gay or lesbian.…”

Multidimensional issue preferences of the European lavender vote

“…I infer ESS survey respondents to be LGB if they i) report to cohabit in the same household as their husband/wife/partner, and ii) the gender of the cohabiting individual matches that of the respondent. This measurement strategy, described and validated with similar and alternative datasets (Fischer, Kalmijn, and Steinmetz 2016), is not without its limitations (Kühne, Kroh, and Richter 2019) 7 .…”

Who wins the British lavender vote? (Mostly) Labour

“…Whilst Turnbull-Dugarte's (2020) pooled cross-national study, which includes UK voters, relies on household-inferred 8 sexuality measures based on the identification of those in same-sex relationships, here I am relying on individuals' self-identification as LGB. This has the benefit of allowing us to test the role of sexuality across the wider LGB population, rather than only those in a relationship (Kühne, Kroh, and Richter 2019). This is important as research points towards identification as LGB exhibiting greater effects on political behaviour and voting preferences than engaging in samesex sexual behaviour alone (Schnabel 2018;Swank and Fahs 2019).…”