ESRA 2019 Draft Programme at a Glance

World Values Survey: New Research Horizons and Methodological Challenges 2

Session Organisers Dr Kseniya Kizilova (Head of the Secretariat, World Values Survey Association)
Professor Christian Haerpfer (President, World Values Survey Association)
TimeFriday 19th July, 13:00 - 14:00
Room D19

The World Values Survey (WVS) is one of the world's largest and longest time-series global social research programs. WVS studies changing values and their impact on social and political life. In 1981-2018, the WVS has carried out representative national surveys in over 110 countries containing 92% of the world’s population. Work on analysing this data has been invaluable for a global network of scholars and international development agencies, including the World Bank, the UNDP, the WHO, regional development banks etc.
WVS wave 7 constitutes the next round of the program. Around 100 000 respondents in over 70 world countries will be interviewed in the course of 2017-2019 on a great scope of issues, including social values, attitudes & stereotypes; societal well-being; social capital, trust and organizational membership; economic values; corruption; migration; post-materialist index; science & technology; religious values; security; ethical values & norms; political interest and political participation; political culture and political regimes. New survey findings on these and other topics will be presented within this panel.
The ambition of the WVS-7 survey round includes testing a number of new survey items belonging both to the WVS research agenda (new forms of political participation, attitudes towards migration, corruption, security etc.) as well as designed to measure the Sustainable Development Goals indicators such as inclusive and responsive decision-making. Results of this pilot in different national and cultural contexts and conclusions on the reliability and validity of the new items will be presented in the panel.
In wave 7 WVS continues extending its geographical coverage and includes countries lacking reliable statistical and census information (i.e. Lebanon or the UAE). National teams in these countries elaborated their own methodological approaches to build national representative samples. WVS scholars will share their sample design techniques within the frame of this panel.
WVS-7 survey round is also innovative in application of mix-mode survey methods, combining online panels, face-to-face interviews and telephone surveys. Panel will feature presentations comparing findings collected in parallel via different survey methods in the same countries allowing to estimate the reliability of mixed-mode surveys as well as the challenges and prospects of methods combination.
We invite submissions dealing both with the analysis of WVS 1-7 waves data and methodological aspects of the WVS surveys implementation. We welcome submissions from both members of the WVS network and independent scholars.

Keywords: World Values Survey, comparative, cross-sectional, cross-cultural, mix-mode methods, survey technique, survey methodology

Individual, Situational, and Country-Level Determinants of Acquiescent Responding – an analysis based on the World Value Survey

Professor Beatrice Rammstedt (GESIS - Leibniz-Institute for the Social Sciences ) - Presenting Author
Dr Clemens Lechner (GESIS - Leibniz-Institute for the Social Sciences )
Ms Melanie Partsch (GESIS - Leibniz-Institute for the Social Sciences )
Professor Daniel Danner (University of Applied Labour Studies (UALS))

Acquiescence (“yea-saying”) can seriously harm the validity of self-report questionnaire data. Toward a better understanding of why some individuals and groups acquiesce more strongly than others do, we developed and tested a unified conceptual framework of acquiescent responding. Our framework holds that acquiescent responding is a joint function of respondent characteristics (e.g., age, education, values), situational or survey characteristics (e.g., interview privacy, respondents’ interest), and cultural characteristics (e.g., social norms, economic development). Moreover, the framework specifies two main mechanisms through which these characteristics relate to acquiescence: deferential communication styles and cognitive processing capacities.
Multilevel analyses using data from 60 heterogeneous countries (N = 90,347) from the World Values Survey (WVS) lent support to our framework’s main tenets. Acquiescence was higher among respondents with a higher age (>55 years), lower education, stronger deference values (i.e., conformity and tradition), and, unexpectedly, male gender. Interview privacy corresponded to lower acquiescence, but this association was small and vanished after including individual respondent characteristics. Unexpectedly, interviewee’s interest in the interview related to higher, not lower, acquiescence. Finally, acquiescence was considerably higher among interviewees from countries with stronger deference norms. We discuss theoretical and practical implications of these findings and discuss how our framework can guide future inquiries into acquiescent responding.

Mixed Mode Administration of Comparative Social Surveys: Evidence from the World Values Survey

Dr Benjamin Phillips (The Social Research Centre) - Presenting Author
Dr Jill Sheppard (The Australian National University)
Ms Anna Lethborg (The Social Research Centre)

In this paper, we contrast the results of a parallel administration of a subset of World Values Survey (WVS) items on Life in Australia™, a probability-based online panel, and the actual Australian WVS Wave 7. Life in Australia™ recruits its members using dual-frame RDD and includes offline respondents via CATI administration. The Australian WVS draws its sample from an address-based sample with the survey administered via mail with a push-to-web recruitment strategy.

Although the administration mode of ongoing and comparative survey research has traditionally been constant, the evolving survey environment threatens this constancy. The WVS has traditionally been administered by face-to-face interview based on cluster sampling of address-based, area-probability or other geographic frames. However, persistent coverage errors, high costs and the introduction of new member nations in which face-to-face interviewing is not feasible have led to increasingly mixed modes of administration.

Out of 63 measures from WVS Wave 7 fielded across both samples and modes, we find significant differences in responses to only seven questions after Hochberg’s (1988) adjustment for multiple testing. Of those differences, the biggest is the importance of religion, with Life in Australia™ panellists more likely to say religion was an important part of their lives. The second largest is the importance of politics in respondents’ lives, with online panellists placing a higher importance on politics. The other differences—though statistically significant—are negligible. Our findings contribute to the emerging literature arguing that mode effects are small enough that they do not preclude mixed mode survey research and that probability-based panels, available in increasing numbers of countries, may be suitable for cross-national research.