ELECTION FORECAST FROM SURVEY DATA

The project of the Survey Methods Room Budapest.

Project duration: February 2024 - February 2025

Our research group has been awarded a grant from the University Excellence Fund (Egyetemi Kiválósági Alap, EKA) for the project "A new method to improve the accuracy of estimates in empirical social research: accessibility-based post-stratification".

During the research, we use a new method to forecast elections from questionnaire data, using specific proxy questions, and then perform post-stratification based on these questions. Our main question is to what extent we obtain different results when estimating attitudes towards participation in elections by adjusting the responses not only for demographic characteristics but also for accessibility groups.

The project will run until February 2025, a particularly timely research theme in the run-up to the European Parliament and local elections. The method builds on several international and national studies and could lead to a method that can be used by the social sciences.

The research involves Tamás Rudas, Blanka Szeitl, Emese Túry-Angyal, Bernadett Nagy and Anna Kovács.

PUBLIC OPINION RESEARCH METHODS IN A NEW TECHNOLOGICAL AND SOCIAL ENVIRONMENT

The project of the Survey Methods Room Budapest, supported by the National Laboratory for Social Innovation

Project duration: January 2022 – January  2024

The aim of the project is the methodological examination of Hungarian online survey research. Within the framework of the project, we analyzed both the market and scientific aspects of the current Hungarian data collection opportunities: our results show that, contrary to global and regional trends, the Hungarian data collection market has lost significant size and quality has deteriorated drastically: the estimated market size of USD 107 million in 2017 has shrunk to USD 63 million by 2021, and we have the second least developed industry according to the official IMDP (Insight Market Development Index) metric.

Based on Hungarian population parameters (regional internet penetration data, composite respondent attitudes and socio-demographic composition estimates), we will implement the international guidelines necessary for the correct and reliable conduct of research using online questionnaire data collection and data interpretation. Simulation studies will be carried out in order to highlight research questions for which online survey data can be used and the topics for which it may be misleading. Using face-to-face survey data, we define theoretical subsamples of the population available offline and online and evaluate the estimates derived from these. This phase of the research will summarize the simulation results in a complex dashboard and make them interactively available to both academia and civil society.

THE LAYERS OF POLITICAL PUBLIC SPHERE IN HUNGARY 2001–2020

A research project supported by NKFIH (National Research, Development and Innovation Office) (K-134428), the project of the Research Center for Computational Social Science.

Project duration: December 2020 – December 2023

The research involves Renáta Németh (lead researcher), Jakab Buda and Márton Rakovics from the Department of Statistics.

The overall aim of the research is to map Hungarian online political discourse from the 2000s to the present day. The transformation of the political sphere and the public outlines the content framework of our research. We analyze different layers of political discourse, including official communication channels (e.g. parliamentary speeches); different types of political press (e.g. online press, news portals, tabloids) as well as user-generated content (online comments, forums, blogs and public Facebook posts). We want to analyze not only the internal discursive content and dynamics of these layers, but also the interactions between them.

Digital data produced in online public spheres are primarily textual. Such data require analytical tools, which became accessible only recently with the emergence of the field of Natural Language Processing (NLP) capable of processing large-scale textual data in a systematic, automated way. These innovative tools provide suitable depth in results for sociology (Németh and Koltai, 2020). 

More information here.

DISCURSIVE FRAMING OF DEPRESSION IN ONLINE HEALTH COMMUNITIES

A project of the Research Center for Computational Social Science, with the participation of Jakab Buda, Renáta Németh and Márton Rakovics from the Department of Statistics, supported by the National Laboratory for Social Innovation.

Project duration: 2018-2024

We investigate the potential for NLP techniques in understanding individual framing of depression in online health communities. Framing of depression is a social construction, it defines the meaning of depression, gives a causal explanation of it and can even determine treatment preferences. The current clinical explanations of depression point to biological, psychological and social discourses (e.g. Comer, 2015).  

Previous research in this field has been primarily qualitative. Investigators have used qualitative content analysis of offline texts (personal diaries, letters, interviews) to investigate the framing of depression (e.g. Riskind et al, 1989). We believe that there is significant research utility in the application of automated text analysis methods to investigate the framing of depression in online, patient-generated non-clinical texts.

More information here.

TEXT ANALYTICAL MODELS ON SOCIAL SCIENCE CORPSES

Project duration: 2018-2020

Renáta Németh participates in the research as a member of the Research Center for Computational Social Science, with the support of the Higher Education Institutional Excellence Program (Felsőoktatási Intézményi Kiválósági Program).

The research group works on multiple topics:

1. Examination of the robustness of text analytical studies. The robustness of text analytics methods (primarily sentiment analysis of Twitter data) is examined by basic methodological aspects, with the secondary analysis of known and published data, with the goal of formulating methodological recommendations.
2. Discovering sociological knowledge through automated text analytics. Also from a methodological point of view, the new possibilities of automated text analysis methods for sociology are examined.
3. Analysis of online press news related to corruption, using dynamic topic models.
4. Examining the framing of depression using posts published in online forums. The question is what causes do the posts trace back the emerging problems: rather to biological, psychological or even sociological factors.

CAUSAL INFERENCE IN SOCIAL SCIENCE RESEARCH

The project of Renáta Németh supported by the János Bolyai Research Scholarship.

Project duration: 2012–2015.

The classical axiom, according to which only data from randomized experiments are suitable for drawing causal conclusions, has been shed new light by some results of recent decades. In my research, supported by the János Bolyai Research Scholarship, I connect to these new results by representing the point of view of the social sciences. My main question is how new statistical results concerning causality can contribute to social science research.

STRUCTURAL EFFECTS IN CROSS-CLASSIFIED DATA

The project of Tamás Rudas, supported by the National Scientific Research Basic Programs (OTKA).

Project duration: 2012-2016.

We expect to gain a deeper understanding of what aspects of data collection designs determine whether or not causal inference is possible based on the data collected. That understanding will offer methods to assess designs and to tell what extent a causal inference may be valid. Methods will be developed, together with software implementation, to measure the effects of assignment and the effects of treatment, given a particular design. These results will be applied to the analysis of data arising from various data collection procedures that are now identified in the literature as pseudo-experiments, natural experiments, confounded or partially confounded designs, etc. The results will be relevant for sociological research and evidence-based policy making.

MIGRATION – APPROACHING THE REALITY WITH INNOVATIVE METHODOLOGY (120711)

The project of Dávid Simon.

Migration is a sensitive social, economic and political topic that requires careful analysis. To date, only a few researches in Hungary have focused on the issue of emigration, especially its economic and labour market consequences. If we examine the available data from a methodological point of view, we can list the strengths and weaknesses, but it can be seen that the methods used do not exhaust the methodological possibilities. Our research goal related to migration research is the assessment of migrants' losses and gains related to migration. Our further goal is to evaluate innovative sampling procedures for investigating the migrant population, in a survey-statistical context, using the approaches, terminology and statistical tools used in this field. We are planning two independent data collection: a simple online survey with propensity score matching correction (using labour force survey and census data), and virtual online respondent-driven sampling. As a complementary method, we plan semi-structured interviews in order to better understand the reasons for sampling biases and the characteristics of the response.

FURTHER PROJECTS:

• 2014–2017 EURA-NET (Transnational Migration in Transition: Transformative Characteristics of Temporary Mobility of People) international research funded by Framework Program 7, with the participation of our instructor Dávid Simon (Kopint-Tárki Ltd, consortium leader: University of Tampere).
• 2015 Complex evaluation research of innovative blended learning teacher training programs with the participation of our instructor Dávid Simon (Educational Research and Development Institute).
• 2015 Satisfaction with state healthcare institutions in the Central Hungary region, with the participation of our instructor Dávid Simon (State Health Care Center).
• 2015 Assessment of the state of school segregation by linking and secondary analysis of educational databases, with the participation of our instructor Dávid Simon.
• 2015 Survey of abuse in kindergarten, with the participation of our instructor Simon Dávid.
• 2014-2015 Complex examination of school abuse, with the participation of our instructor Dávid Simon (Educational Research and Development Institute).