The International Sociological Association’s RC33 Research Committee “Logic and Methodology”

Research Committee on Logic and Methodology in Sociology, RC33, Sessions

Program Coordinators: Katja Lozar Manfreda, University of Ljubljana, Slovenia, katja.lozar@fdv.uni-lj.si; Vera Toepoel, Utrecht University, Netherlands, v.toepoel@uu.nl. The number of allocated sessions including Business Meeting: 18. Proposed sessions in alphabetical order:

Analysing and Counteracting Undesired Response Patterns in Survey Research (session in English). Session Organizers: Ben Jann, University of Bern, Switzerland, jann@soz.unibe.ch; Volker Stocké, University of Kassel, Germany, volker.stocke@uni-kassel.de; Sebastian Sattler, University of Freiburg, Germany, sebastian.sattler@psychologie.uni-freiburg.de; Peter Graeff, Bundeswehr University, Germany, peter.graeff@unibw.de. Evidence suggests that, depending on topic, respondents tend to distort their answers to comply with social norms (social desirability bias), refuse to answer certain questions (item non-response), or refrain from participating in the entire interview (unit non-response). Especially questions about deviant behaviour (e.g., drug use, plagiarism) can be affected, as well as questions about attitudes, preferences and values (e.g., attitudes towards minorities, self-reported religiosity, age-related stereotypes). These problems can vary with topic and survey technique, and also across groups of respondents (e.g., due to differing internalized social norms with respect to the behaviour under investigation), and pose a serious threat to data quality.

Providing a theoretical framework to explain the factors influencing such response patterns and investigating them are challenges of current survey research. In general, social norms are the basis on which respondents determine whether an answer might be socially desirable or not. However, situational factors are important as well. For example, respondents might gauge their answers in response to interviewer characteristics and behaviour, as well as the survey sponsor. Furthermore, undesired response patterns can be intrinsically or extrinsically motivated, and can be conceptualized as the result of “self-deception” or “other-deception”.

Various approaches such as randomized and non-randomized response techniques, vignette-based studies, or self-administered questionnaires have been suggested to improve data quality by increasing the respondents’ feeling of privacy protection and anonymity. Furthermore, imputation techniques are employed to deal with missing data and increase the amount of usable information.

This session invites contributions about: theoretical approaches for explaining and predicting undesired response patterns; techniques to reduce such response biases; empirical studies investigating undesired response patterns; statistical techniques that help to reveal undesired response patterns; and approaches to address problems of data quality after data collection.

Challenges to the Secondary Analysis of Large-Scale Cross-National Comparative Surveys (session in English, not open for submission of abstracts). Session Organizer: Kazufumi Manabe, Aoyama Gakuin University Tokyo, Japan, kazufumi.manabe@nifty.com. One of the most significant developments in contemporary social sciences is the implementation of large-scale cross-national comparative surveys. So far, such surveys have been conducted mainly in Western countries. Two such excellent survey projects include the European and World Values Survey and the International Social Survey Programme. In recent years, cross-national comparative surveys have also started in Asian countries. The examples of such survey projects are the East Asia Values Survey and the AsiaBarometer.

Data from cross-national surveys are being accumulated, and efforts are being made to conduct secondary analysis of survey data at the international level. The secondary analyses to be presented in our session will focus on: confirmation of theoretical hypotheses in the social sciences; exploratory data analysis; and examination of the question equivalence. It is our aim that by proposing new ideas and methodology, we will be able to shed light on the advantages and challenges of secondary analysis.

As examples of large-scale surveys, we are thinking of: the European Values Study, the World Values Survey, the International Social Survey Programme, the East Asian Values Survey and the AsiaBarometer. We are planning to invite a key person from each of these surveys.

Cognitive Aspects of Survey Research (session in English). Session Organizers: Wander van der Vaart, University of Humanistic Studies, Netherlands, w.vandervaart@uvh.nl; and Robert F. Belli, University of Nebraska, USA, bbelli2@unl.edu. Psychological issues have always been important in methodological studies into survey data collection. In particular, the “cognitive aspects of survey methods” (CASM) movement has made a huge contribution to the field of data collection. More recently, both conversational interviewing methods and survey calendar methods again steered attention towards psychological aspects of data collection methods.

This time attention was paid not merely to specific methods like question wording but also to more general procedures such as the style of interviewing and the integration of (aided recall) tools in the interview. The rise of conversational interviewing and calendar interviewing in survey practice urges for a more profound examination of the psychological mechanisms that underlie these and related data collection procedures.

This session focuses on cognitive and conversational aspects that are of importance to these data collection methods. This session brings together studies from the several fields like linguistics, sociology and psychology, including research into: verbal interaction, framing, non-verbal behavior; memory, life histories, narratives, timeframes, aided recall, probing; and the interaction of these issues with characteristics of the study population, the type of research design, the mode of data collection, and the visual or verbal characteristics of data collection tools. The aim of the session is to discuss how conversational and cognitive insights can be used to enhance the data quality as produced by flexible interviewing methods.

Composite Likelihood Methods and Approaches to Handling Non Response in the Analysis of Longitudinal Data (session in English). Session Organizers: Roger Penn, Queen’s University Belfast, Ireland, r.penn@qub.ac.uk; and Damon Berridge, Lancaster University, United Kingdom, d.berridge@lancaster.ac.uk. This proposed session will provide a forum for papers on recent developments in composite likelihood methods. These methods are relatively new but they offer innovative approaches to the analysis of longitudinal data, particularly when there are multiple explanatory variables and multiple response variables (see Penn and Berridge, 2013).

The complex inter-relationships between a set of multiple response variables and a set of explanatory variables can be broken down into a series of relationships between pairs of variables. A composite likelihood is built up on the basis of a set of pair-wise likelihoods, thereby offering an efficient way of analysing complex multivariate longitudinal data.

The session will be open to social scientists who are engaged in the development and refinement of such methods and in their application to sociological questions, as well as those involved in the development of appropriate specialist software solutions such as Sabre R.

The proposed session builds upon the successful session that took place at the RC33 VIII Social Science Methodology Conference in Sydney in 2012 on the issues arising in the collection and analysis of incomplete longitudinal data.

Crisis, Transnational Migration, and the Gender Order in Europe (session in English, not open for submission of abstracts), integrative session with RC31 (Sociology of Migration), RC32 (Women in Society), RC38 (Biography and Society), the German Sociological Association (DGS), the European Sociological Association (ESA) and the RC33 Research Network on Women and Gender Studies.

Dealing with Non-response - Strategies to Increase Participation and Methods for Post-Survey Adjustments. Session Organizer: Vera Toepoel, Utrecht University, Netherlands, v.toepoel@uu.nl. Response rates are rapidly declining. Non-response bias is a major threat to the validity of survey answers. Researchers are developing tools to increase survey participation and use pre-survey adaptive designs or post-survey adjustments to improve representivity of survey results. Which types of tools increase participation? We know from the literature that incentives are a major trigger for responding, but are there other (less expensive) tools that increase participation as well? How can you use adaptive designs to increase participation and reduce non-response bias? And last, how can weighting be implemented in post-survey correction. Which type of weighting methods work? Which variables can best be used for weighting? How well does weighting in non-probability samples work? This section will discuss all methods that can be used to improve sample composition to produce reliable results.

Design Aspects of Response Scales in Surveys (session in English). Session Organizer: Kathrin Bogner, Leibniz Institute for the Social Sciences, Germany, kathrin.bogner@gesis.org; Natalja Menold, Leibniz Institute for the Social Sciences, Germany, natalja.menold@gesis.org; and Christof Wolf, Leibniz Institute for the Social Sciences, Germany, christof.wolf@gesis.org. This session focuses on the design aspects of response scales measuring attitudes, opinions or behavior in surveys and their potential effects on response behavior and data quality. Researchers are invited to submit papers dealing with the effects of design aspects of response scales such as number of categories, middle category, unipolar or bipolar scales, numerical and/or verbal labels, ascending or descending order of categories or the scale’s visual design/layout.

Papers should focus on the effects of response scale design on response behavior and reliability and validity of the responses. Contributions analyzing the moderating influence of cognitive or motivational factors on the effects of response scale design are also welcome. Furthermore, specifics in design of response scales in different survey modes, their comparability in mixed-mode surveys, as well as their intercultural comparability are further topics of interest.

Facing the Challenges of Data Collection via Mobile Internet (session in English). Session Organizers: Marika de Bruijne, CentERdata, Netherlands, m.debruijne@uvt.nl; and Arnaud Wijnant, CentERdata, Netherlands, wijnant@uvt.nl. Engaging mobile devices for data collection is relatively new in social sciences. Mobile surveys challenge the typical questionnaire design and introduce new ways to interact with the questionnaire. Compared to surveys, mobile devices provide an opportunity to collect different types of non-reactive data such as geographical data or accelerometer measurements through the numerous types of sensors present in the devices.

In this session, we wish to discuss the recent findings of methodological research on the effects of mobile surveys, new ways of recruiting and approaching respondents, alternative data collection methods enabled by mobile devices, the challenges of response rate and bias in (mobile) online surveys, and considerations on mobile survey user interfaces and questionnaire design. This session will also discuss future developments in survey interaction.

How to Think Methodology without Notion of Object or Subject? (session in English). Session Organizers: Charlotte Baarts, University of Copenhagen, Denmark, cba@soc.ku.dk; and Lina Hauge Katan, University of Copenhagen, Denmark, lhk@soc.ku.dk. Recently, new theories of science referred to as “new materialisms” have introduced frameworks of thoughts about scientific practices that refuse completely any notion of object or subject. It is not at all new that the subject-object dualism as a fundamental presumption for the traditional practice of social sciences is under attack. It has been so since the rise of modernity and not at least since the first appearances of methodologies such as reflexivity in the 1960s. These methodologies sought to moderate the distance and separation implied by the dichotomised dualism by taking into account the effect that the scientist had on her object and the influence of the object on her – in other words, the relation of inter(ex)changeability between the two. Within the last couple of decades, however, new materialisms have brought this confrontation of the dichotomy as far as to a complete abandonment of both object and subject as points of reference. Concepts such as “engagement”, “entanglement” and “intra-action” announce an inherent relationality of separately indeterminate materialities including that of the scientist. No object pre-exists nor is evoked by a subject; instead properties and boundaries are appearances that emerge as effects of continuous becomings rather than delineations of separate entities. In the disappearance of subject-object as given positions of difference and distance, the points of reference they represent for epistemology and ontology also fade, thus evoking new methodological challenges for the practice of the social sciences.

New materialism is thus a move beyond methodological concepts that have functioned as central presumptions for the social sciences. Hence this move leaves behind questions to be explored within this theoretical framework about the practice of science and our own part in it. Explorations in this field of questions could for instance reflect the possibilities of imagining a methodology that omits any notion of object or discuss how sociology should approach its field of study, not to mention identify it, if this process does not involve the concept of object. Furthermore, relevant questions to be pondered could be: who the social scientist becomes in the lack of relation to an object; how to conceive science as a practice of a subject if there is no object; or how to even begin conceptualizing methodological approaches, if we cannot establish anything to be approached?

This session welcomes contributions of theoretical, empirical, practice-based orientation that somehow explore, assess, elaborate or connect these or related considerations with regard to the formative stages, recent actualizations or future potentials of knowledge production.

Latent Constructs or Simply Descriptive Measures? Index-based Measurement and Applications (session in English). Session Organizers: Peter Graeff, Bundeswehr University Munich, Germany, peter.graeff@unibw.de; and Robert Neumann, Dresden University of Technology, Germany, robert.neumann@tu-dresden.de. A large part of research within the social sciences relies on item-based indices (such as the Index of Materialism/Postmaterialism or the Happy Planet index). The measurement qualities of indices for socio-economic and socio-political phenomena are topics which have received too little attention from scholars across social science disciplines, which might be due to the different understanding about how the phenomenon of interest should be measured.

While sociologists and psychologist usually apply the idea of latent constructs, it seems that econometricians (such as Word Bank researchers) understand the measurement process in a rather descriptive way. As a result, they do not elaborate techniques of measurement and focus instead on statistical techniques for analyzing data. This might be problematic as the most prominent and large data sets such as those of the World Bank or similar institutions are gathered and constructed by economic researchers. These data sets (including their scales and indices) are usually picked up in (comparative) social science research and applied in multivariate models without reflecting their measurement properties or meanings. While there is an abundant use of such indices when comparing, for example, institutions, quality of governance, welfare regimes or educational systems both theoretically and empirically, the origins or the construction of these indices is only seldom regarded.

This session proposal aims at theoretical and empirical contributions which: contribute to stimulating the missing debate about criteria of assessing indices for comparative research and multivariate modeling; address and identify criteria that will improve the comparability of different indices; address new forms of parameter estimation (e.g., Bayesian approach) or new procedures for assessing the measurement quality (for instance, structural equation models evaluating the validity and reliability of data); apply indices within a multivariate approach to measure a phenomena of interest on the group or aggregate level indirectly. Papers which contribute to the “latent” properties of variables or items will be highly appreciated, altogether with contributions that show the applicability of measurement criteria in research fields dealing with indices (democracy, governance, conflicts or values). Since the application of indices is prominent in sociology, psychology and political science, the topic can be of interest for a broad audience.

Measuring and Making Use of Social Embeddedness in Survey Research (session in English). Session Organizers: Dominique Joye, Université de Lausanne, Switzerland, dominique.joye@unil.ch; and Christof Wolf, Leibniz Institute for the Social Sciences, Germany, christof.wolf@gesis.org. That human behavior is influenced by other humans is a key assumption of social sciences. In the 1940s, Paul Lazarsfeld probably was the first to design a quantitative study in which this assumption was tested. This early study was followed by Edward Laumann's investigation of the influence of three best friends in the Detroit Area study, by Claude Fischer's multiple name generator instrument implemented in his seminal North California Community Study and by Ronald Burt's single-item name generator designed for the GSS, to name but a few influential developments in the collection of ego centric network data. In spite of their significance, these instruments have been criticized in different ways. One criticism emphasizes that these instruments only capture the core of social relations, neglecting “weak” ties that can have considerable impact in some circumstances. Other critics have pointed to methodological short comings of these instruments; e.g. the comparatively large interviewer effects produced by these instruments.

Alternative approaches, like Nan Lin's position generator or Martin Van der Gaag's resource generator, try to overcome these problems. The collection of network data today is not confined to a few specialized studies but has been introduced in many major national and international surveys which often contain at least some aspects of social embeddedness. Aside from collecting network data, survey research has also utilized the social fabric to develop novel sampling procedures; i.e. network sampling, a method particularly pertinent to capture hard to reach populations.

The session aims at discussing current developments of measuring social networks and social resources. Contributions comparing different operationalizations or contributions investigating the cross-national stability of variability of network indicators are particularly welcome. Also, research based on network sampling will be highly welcomed.

Methods of Social Network Analysis (session in English). Session Organizers: Peter Carrington, University of Waterloo, Canada, pjc@uwaterloo.ca; and Anuska Ferligoj, University of Ljubljana, Slovenia, anuska.ferligoj@fdv.uni-lj.si. Social network analysis is the application to social research of the concept of the network – a set of entities, or nodes, connected by relationships, or ties. Conceptualisation of social structures as social networks has been fruitful in many areas of the social sciences, and has, indeed, facilitated recognition of substantive patterns and analytic problems common to the social and other sciences. One of the most lively areas of social network analysis has been the development of suitable methods for applying the network concept in social research. These methods address the three main issues: sampling, measurement, and data analysis. In each of these areas, the problems faced by network researchers are considerably, though not entirely, different from those encountered by conventional attribute-based research.

This session will provide a forum for presentation of new developments in research methods for social network analysis. These papers may be: theoretical, concerning epistemological problems in the use of the concept of the social network; methodological, concerning technical developments in sampling, measurement, or data analysis; or empirical, demonstrating novel applications of social network analytic methods in actual research.

RC33 Business Meeting. Session Organizer: Katja Lozar Manfreda, University of Ljubljana, Slovenia, katja.lozar@fdv.uni-lj.si.

Response Latencies in Survey Research: Methodology and Applications (session in English). Session Organizers: Jochen Mayerl, University of Kaiserslautern, Germany, jochen.mayerl@sowi.uni-kl.de; and Volker Stocké, University of Kassel, Germany, volker.stocke@uni-kassel.de. The time necessary to answer survey questions is an unobtrusive and cost-effective measure to enrich data of computer-assisted surveys. There is a wide range of applications of response latencies in survey research, inter alia to operationalize cognitive accessibility and depth of thought, to identify high predictive social judgments, to detect response effects, to measure implicit attitudes, to identify wording problems and to evaluate survey questions in other respects, and to explore the validity and reliability of respondent’s answers. However, response latency data are proxy measurements and biased by survey mode (CATI, CAPI, CASI), measurement instrument, interviewer characteristics, interview situation, and individual characteristics. Thus, identification of biasing factors as well as measurement and statistical treatment of response latency data is essential to avoid biased interpretation of response-time results and to improve data quality. More research into the factors affecting the predictive validity of response times is needed.

This session has the goal of contributing to solving the above mentioned and other open questions about the validity of response latencies as an indicator for data quality. In particular, we welcome contributions relating to: (a) substantive and methodological applications of response latencies in survey research, (b) analyses concerning data quality and data treatment of response latencies, and (c) data collection issues of response latencies.

Spatial Methods (session in English). Session Organizers: Nina Baur, Technical University Berlin, Germany, nina.baur@tu-berlin.de; Linda Hering, Technical University Berlin, Germany, linda.hering@tu-berlin.de; and Cornelia Thierbach, Technical University Berlin, Germany, cornelia.thierbach@tu-berlin.de. The session aims at exploring which research methods are appropriate for approaching space in the social sciences, seeing space either as a dependent or independent variable. Researchers can ask how people think about space and construct space or they can see space as a relevant frame for social action that influences social life. Papers should address one of the questions below either at a more general methodological level or using a concrete example in a specific research project. Which qualitative and/or quantitative methods are best suited for which kind of theoretical problems? What methodological innovations concerning the spatial can be observed? (How) can traditional sociological or geographical methods be adjusted to address spatial problems within sociology? Which sampling strategies are appropriate for spatial problems? What are the specific data requirements for spatial analysis, and how can these data be collected? Which strategies of data analysis are appropriate for spatial analysis?

Theoretical and Methodological Challenges in the Study of Different Dimensions and Aspects of Inequality (session in English). Session Organizers: Elisabet Nasman, Uppsala University, Sweden, elisabet.nasman@soc.uu.se; Linnea Bruno, Uppsala University, Sweden, linnea.bruno@soc.uu.se; Julijana Angelovska, International Balkan University, Macedonia, julijana.angelovska@yahoo.com; Krystyna Slzny, Jagiellonian University, Poland, krystyna.slany@uj.edu.pl; Ewa Krzaklewska, Jagiellonian University, Poland, ekrzaklewska@gmail.com; and Anna Ratecka, Jagiellonian University, Poland, anna.ratecka@uj.edu.pl. Intersectional analysis has now been used in research on dimensions of inequality for many years, but has also been criticised and contested concerning theory development as well as how such an approach can be applied in empirical research. There are, for instance, different perspectives on intersectionality, mainly concerning the relationship to power. There is also a question about the stress on inequality in terms of financial resources compared to other aspects of inequality that are not easily or meaningfully quantified or sufficiently captured in such studies.

Departing from an approach that focuses on power relations and different social dimensions of inequality, on the different aspects of inequality that impact on lived experiences of oppression or privilege, or on the formation of subjectivities, we would like to invite discussion concerning new theoretical ideas and problematic or enlightening experiences drawn from empirical studies. There is a need for further comprehensive research, especially on experiences of inequality in daily lives.

The organisers of this session are trying to challenge the state of art as far as inequality is concerned. By using the methods produced across different disciplines concerned with inequality, can we really understand the potentials of equality for individual quality of life and social development? Do we have sufficient tools to study inequality on the individual level? Are we able to catch up with the changes in the experienced inequality within life course? Are we able to compare these situations between diverse societies or between those living in different social contexts?

Time Use and Daily Activities (session in English). Session Organizer: Kimberly Fisher, University of Oxford, United Kingdom, kimberly.fisher@timeuse.org. Time is a basic element of daily life. Unlike financial, physical or other resources, time is the one resource shared equally by all people. Understanding how people use their time informs a wide variety of social research, from monitoring human impacts on the environment, to measuring the full scale of economic activities (including those transpiring outside the paid economy), to modelling scheduling of care, voluntary, and free time activities in with changing patterns of paid work, to assessing shifts in well being and the degree to which people lead (un)healthy lifestyles, to assessing integration of minority populations, to examining the scale of gender inequalities generated by daily activity patterns. Interest in time use research has expanded recently, with the UN Statistics Division, UN Economic Commission for Africa, UN Economic Commission for Europe, OECD and Eurostat all releasing new guidance for the use of time use data in policy research in 2013. This session will accommodate a range of papers about the collection of time use data, the use of new techniques in analysing people’s use of time, and new findings arising from time use research.

Using Survey Data to Describe Societies at a Global Level – Is There Still a Hidden Treasure? (session in English). Session Organizers: Markus Quandt, Leibniz Institute for the Social Sciences, Germany, markus.quandt@gesis.org; and Christof Wolf, Leibniz Institute for the Social Sciences, Germany, christof.wolf@gesis.org. With the advent of long-standing comparative programs such as the International Social Survey Programme, various international “Barometer” projects, the European Values Study, the World Values Survey, and the European Social Survey, the option of deriving society level, comparative (and even trend) information from academic survey data has come into the world. Researchers working for example on value change or on the analysis of “cultural zones” have used this opportunity extensively; and their substantive work has been accompanied by a growing literature on the cross-national equivalence of attitude measures based on multiple indicators.

A similar potential for deriving society level measures seems to exist when we turn to non-attitudinal concepts: e.g. behavioural variables such as church attendance or socio-demographic characteristics such as migration background. Potentially, survey data from the generic programmes mentioned above hold a wealth of information in areas where official statistics data fall short of research needs due to limitations in thematic scope and cross-nationally harmonised measures. However, we observe a lack of systematic research on the validity and reliability of such survey-based measures in comparative settings. Often comparability is implicitly assumed to be a trivial issue here.

This neglects problems that can affect all comparative research: translation and question format issues, lack of validity due to different national contexts, variation in sampling designs, problematic ex post-harmonisation requirements, etc. A similar lack of research appears when we look at a limitation of survey data that is rather obvious: Small sample sizes (in comparison to official statistics) reduce estimation reliability and hamper the analysis of sub-populations. But where exactly are the practical limits?

This session strives to invite new research on the quality of aggregate level estimates. We welcome presentations that demonstrate and analyse the robustness or vulnerability of aggregate measures derived from comparative survey data to the aforementioned and related problems. Examples are comparisons of survey-based estimates to census-like data, or research into quality effects of particular measurement approaches that different surveys might pursue for the same concepts. We particularly invite research that strives to overcome limitations of given data sources by combination with other sources; e.g. by pooling data across surveys to increase final sample sizes, by estimating correction factors for known biases, etc.

Well-being and Quality of Life. Methodological Challenges for Cross-National Surveys (session in English). Session Organizers: Wolfgang Aschauer, University of Salzburg, Austria, wolfgang.aschauer@sbg.ac.at; and Martin Weichbold, University of Salzburg, Austria, martin.weichbold@sbg.ac.at. Research on subjective wellbeing has gained enormous importance during the last years. It seems that we witness a fundamental turn from measures of economic progress to multidimensional measures of quality of life and societal wellbeing, as well as a shift from objective indicators to a higher relevance of subjective measures. Thirdly, major efforts in the development of new concepts for cross-national comparisons are clearly visible (e.g., National Accounts of Wellbeing of the NEF foundation, “Your better life index” of the OECD). Despite this boom of subjective wellbeing measures, also with regard to cross-national research, the analysis of the comparability of these approaches still remains in its infancy.

It seems to be common to take the cross-cultural equivalence of the concepts for granted, to neglect the use of statistical methods of equivalence testing and to widely ignore critical approaches questioning the cross-cultural validity of wellbeing in general. Culture-specific concepts – representing a counter-trend in wellbeing-research – highlight that there is no universally accepted general theory on wellbeing and certain components of the construct are culturally sensitive. Locally emerging concepts (such as the GNH approach of Bhutan or the ONS concept of GB) are sometimes considered more valid than approaches with the intent of being universally relevant.

Picking up these contemporary developments and debates in subjective wellbeing research, this session addresses three main research questions. If we decide to conceptualize comparable wellbeing measures, how should we address the methodological challenge of achieving cross-cultural validity? How can we assess the cross-national equivalence of existing concepts of wellbeing? Is it possible to construct new comparable indicators of quality of life and societal wellbeing based on survey data and subjective perceptions of the citizens? The session aims at participants who work in the field of wellbeing research and use cross-national survey data. We welcome speakers who try to develop new conceptual frameworks of wellbeing and quality of life which can be applied to cross-national research or who highlight the challenges of comparability of wellbeing approaches.

Joint Session: Panel Data Analysis of Families Worldwide, Joint session of RC06 Family Research (host committee) and RC33 Logic and Methodology in Sociology (session in English). Session Organizers: Junya Tsutsui, Ritsumeikan University, Japan, junya_tsts@nifty.com; and Michiko Nishino, Toyo University, Japan, mnishino@toyo.jp. In almost every discipline in social sciences, the power and effectiveness of panel (longitudinal) data has been recognized in the analysis of human behaviour. Panel analyses are able to achieve stricter estimations than ordinary cross-sectional analysis. Panel based family studies can provide a more accurate estimation of effects, such as the effect of family support programs on marriage formation or fertility, or the effect of shorter working hours on husband’s contribution to housework, etc. The US has a long history of panel data accumulation. Now panel data are ready to be analysed by researchers in Japan and other regions.

The session welcomes papers analysing family issues using panel data from every region of the world. The session will provide up to five presentations, followed by discussions where researchers are expected to exchange methods of analyses or contemporary conditions regarding the accumulation of panel data on family behaviour. To facilitate this purpose, session organizers will pursue the maximum diversity in regions represented by presenters.