All time references are in CEST
Poster Session |
|
| Session Organiser | Dr Daniel Seddig (Criminological Research Institute of Lower Saxony) |
| Time | Tuesday 18 July, 09:00 - 10:30 |
| Room | |
Keywords: poster
Dr Romana Burgess (Population Health Sciences, University of Bristol) - Presenting Author
Mr Mark Mumme (ALSPAC, Bristol Medical School, University of Bristol)
Dr Anya Skatova (Population Health Sciences, University of Bristol)
Surveys are foundational to longitudinal population studies (LPS), providing self-reported data on health, sociodemographic and other factors. However, self-report is prone to biases—like recall errors and social desirability effects—which can limit reliability. To address this, we piloted the linkage of supermarket shopping data from loyalty cards with survey responses in the Avon Longitudinal Study of Parents and Children (ALSPAC), a UK LPS. We demonstrate the potential of using objective purchase records to enrich survey data, providing a more comprehensive understanding of individual health and wellbeing.
Between September and November 2023, 6170 ALSPAC participants were invited to share loyalty card data, with 511 providing details from at least one of five UK retailers. Participants provided their name, card number, and electronic signature. Retailers are unfamiliar with providing data for research and reluctant to enter data sharing agreements, so participant details were securely sent to retailers as Freedom of Information requests to retrieve the transaction records.
Data of 244 participants from one retailer were linked into the ALSPAC database. Transactions—covering 658,375 item purchases from 2013 to 2023—include item descriptions, date, time, price, and quantity; items are categorised into 82 categories (e.g., produce) and 1,002 subcategories (e.g., bananas), covering 55,176 unique products. These records complement ALSPAC’s extensive existing dataset, which includes 30 years of sociodemographic and health-related survey data; including a 2019 food frequency questionnaire which covers product-level consumption patterns (e.g., dairy, meat). Despite challenges like low consent rates and ethical complexities, combining these data sources shows potential to advance population health research by improving the accuracy and depth of dietary analyses.
Our findings showcase the bidirectional value of linkage: survey data contextualise shopping records, informing representativeness and potential biases, while objective purchase data mitigate self-report biases.
Professor Emilia Rocco (University of Florence) - Presenting Author
Ms Lisa Braito (University of Florence)
Dr Veronica Ballerini (University of Florence)
Professor Chiara Bocci (University of Florence)
Dr Patrizio Lodetti (University of Florence)
This study aims to employ robust methods to derive valid and reliable inferences about firms' behaviors using web scraping data, with a particular focus on the propensity of Italian firms toward technological innovation and sustainable practices. These topics, which are closely aligned with the United Nations Sustainable Development Goals, have garnered increasing attention. However, comprehensive data or probability-based survey samples are often unavailable, especially for specific areas or sectors of interest. While existing business surveys based on probabilistic samples provide extensive firm-related information, they often fail to directly capture sustainable behaviors. In contrast, innovative sources like web scraping enable the collection of data on these topics from large but non-probability samples. Yet, relying exclusively on such data may result in significant bias due to selection bias. Therefore, it is crucial to combine them with traditional probability survey data, provided that auxiliary variables shared by both sources are available and that the two sources refer to the same target population. In this study, to evaluate the possibility of obtaining reliable estimates by integrating the two data sources, we assume a subset of firms included in the ORBIS database as the target population. Non-probability data are collected for all firms with valid URLs, and relevant target variables -such as indices reflecting innovation or sustainability- are constructed from website and social media activity. A smaller probability sample is selected using a stratified sampling design. A double-robust estimation approach is then applied to obtain reliable and unbiased estimates of key outcomes. Finally, some remarks are made regarding the potential of combining traditional and innovative data sources to derive meaningful insights into firm behaviors, as well as the crucial role of auxiliary data from traditional sources in ensuring the external validity of the estimates.
Ms Jessica van de Grint (University of Cambridge) - Presenting Author
Ms Laurel Fish (University College London)
Ms Marialivia Bernardi (University College London)
Professor Pasco Fearon (University of Cambridge)
Numerous studies have highlighted the importance of early language experiences in shaping children's language, cognitive and social development. Audio recordings have become increasingly popular for evaluating young children’s language environments. Recordings provide an objective and precise modality to assess children’s early language development and the linguistic environments that promote it. Manually transcribing and analysing recordings is a very time-consuming process, with transcription taking typically eight times the recording duration. Therefore audio recordings have thus far only been collected on a relatively small scale. AI algorithms have opened up new possibilities for automated speech processing and analysis in the context of large-scale studies. Data collection on a larger scale is crucial for generalisability, detecting smaller effects, and for identifying important at-risk subgroups in the population.
Using two different on-going large-scale birth cohort studies, we have tested the feasibility of two methods of remotely collecting language audio recordings. Data from both studies were processed using the Voice Type Classifier and Whisper. In study 1 participants of the Early Life Cohort + Feasibility Study (ELC+-FS) received a wearable audio recorder at the end of a home-visit for the child (6-12 months) to record from waking up to as close to evening bedtime as possible (N = 38). In study 2, a sub-group of participants from the Children of the 2020s longitudinal birth cohort study recorded a parent-led, 5-minute video recording task with their three-year-old via the study’s smartphone app, Baby-Steps (N = 277 thus far).
In the presentation we will present our findings on the quality and validity of the automated audio records and then discuss lessons learned and future directions.
Dr Jonas Beste (Institute for Employment Research) - Presenting Author
Dr Corinna Frodermann (Institute for Employment Research)
Professor Mark Trappmann (Institute for Employment Research)
Panel surveys provide particularly rich data for implementing adaptive or responsive survey designs. Paradata and survey data as well as interviewer observations from all previous waves can be utilized to predict fieldwork outcomes in an ongoing wave. In the German panel study PASS, data from the first 14 waves were used to develop a machine learning-based model to prioritize households at high risk of attrition. This model enables targeted interventions to reduce panel attrition effectively.
An experiment evaluated the impact of increased monetary incentives, raising prepaid incentives from 10 to 20 euros. The results showed that this increase significantly reduced attrition among households with low response propensities. Additionally, the intervention mitigated biases in key target variables, thereby enhancing overall data quality.
Building on these findings, the increased incentives for the half of the cases with the highest predicted dropout probability were applied in PASS waves 16, 17, and 18. Although no experimental design was implemented during these waves, we investigate whether the positive effects of increased incentives persist over time. Furthermore, we explore potential unintended side effects, such as increased dropouts when additional incentives are omitted in later waves. The goal is to assess both the long-term benefits and potential risks of this incentive strategy, providing valuable insights for the design of future waves in the PASS panel study.
Miss Meike Scholz (TU Bergakademie Freiberg) - Presenting Author
The widespread adoption of social media for data collection has created unprecedented opportunities for researchers. Prior studies use social media data to investigate a variety of different occupational, individual but also societal matters like organizational attractiveness (e.g., Carpentier et al., 2019), human behaviour (e.g., Kaya & Bicen, 2016), or public opinion (e.g., Gorodnichenko et al., 2021). However, collecting and analysing social media data for research purposes raise significant legal challenges. A key concern is compliance with data protection regulations such as the General Data Protection Regulation (GDPR) in Europe. Researchers must navigate complex questions regarding the distinction between public vs. private data, as social media platforms often blur these boundaries. Additionally, terms of service agreements set by platforms like Instagram need to be considered and, thus, may restrict data collection practices. Also, ethical implications of using social media data further complicate legal considerations. Issues of (informed) consent, respect for people or beneficence need to align with regulatory frameworks (Legewie & Nassau, 2018). Furthermore, social media data can be fugacious as they do not underlie the control of the researcher. Just as sudden as an Instagram-post appeared, it can be deleted. This stands in contrast with obligations to store research data for a certain amount of time to align with the principles of FAIR (findable, accessible, interoperable, reusable) and to prevent scientific misconduct (del Pico et al., 2024). Overall, the accordance of legal and ethical considerations can impede academic research by imposing significant barriers to data access, collection and analysis, even when research serves the public interest. Collaborative efforts between legal experts, ethicists, and researchers are essential to create guidelines that balance legal and ethical accountability. Future research must focus on developing procedures that comply with legal standards while maximizing the utility of social media data for scientific inquiry.
Mrs Kristin Williams (RTI International) - Presenting Author
Mrs Kate Burdekin (RTI International)
Mrs Hope Davis-Wilson (RTI International)
Mrs Carolyn Huitema (RTI Internaional)
Mr Joel Montavon (RTI International)
Introduction: Nutrition for Precision Health, powered by the All of Us Research Program is a multi-site clinical trial using a discovery science approach to explore physiological responses to different diets, and use artificial intelligence/machine learning to develop predictive algorithms. Nutrition for Precision Health, funded by the NIH Common Fund, is the first ancillary study to the All of Us Research Program. Data collected via wearables, surveys, physical measurements, dietary assessments, and biospecimen collection are sources for the diverse dataset. This presentation will focus on operational challenges and lessons learned in collecting activity, sleep, and heart rate data via wearable devices.
Methods: The ActiGraph GT9x device was chosen to collect activity and sleep data due to its ability to generate the full accelerometry dataset for analysis and maintain data masking. A Polar Chest Strap Heart Rate monitor was chosen to collect heart rate data. Devices are initialized by coordinators at the study site and placed on each participant. Data is then transferred from clinical sites via a secured file transfer (SFTP) process once returned by the participant.
Results: The study identified several operational challenges with the data collection. These included the burden of resolving misidentified files, Bluetooth pairing connectivity issues, and lack of battery life for certain aspects of the protocol. Important steps to resolving these issues included maintaining a good relationship with the vendor, discussions with coordinators, generating troubleshooting documentation, and working to identify automated methods to file cleaning. These steps were completed to improve the quality of the data and the efficiency of processing it.
Conclusions: Wearable data can provide key insights into the activity of individuals but needs thoughtful implementation to scale for large multi-site studies. Lessons learned from this study can aid future researchers in considering potential limitations and how to overcome them.
Dr Sara Scholtens (Public Health Agency Sweden) - Presenting Author
Mrs Karin Valentin Asin (Public Health Agency Sweden)
Mrs Katrin Bjerhag (Public Health Agency Sweden)
The Public Health Agency of Sweden is assigned to promote being outdoors to achieve health benefits. The agency currently uses survey questions in the Environmental Health Survey to measure how often the public are in nature. The survey is population-based and has been carried out every four years since 1999. The results are widely used; however, little is known about how the questions are understood by respondents. Here, we aimed to gain insight to people’s understanding of the current survey question regarding how often they were in nature and the concept of “outdoor activities” in order to give potential recommendations for further question development.
Method
We sent two different questionnaires to separate samples of 200 adults belonging to the agency’s nationally representative web panel Hälsorapport. One questionnaire was the fixed-answer question used in the Environmental Health Survey followed by two open-ended probes. The other questionnaire consisted of three open-ended questions. Both questionnaires had a response rate of 50 percent, with acceptable distributions of age, sex, education level and region. An inductive content analysis of the qualitative data was performed by grouping the qualitative data into categories reflecting general themes in the material.
Results
The results point to severe measurement issues with current survey question and respondents felt that the question reflected a markedly urban perspective. Descriptions of “being in nature” and understanding of the survey question varied greatly. Physical, social, and existential aspects of the concepts were highlighted in the responses.
Discussion
The methodology used conserved resources while allowing for the collection of rich qualitative data. The complex mental models that were manifested in the answers to the open-ended questions point to a need for more specific operationalisations of “being in nature”.
Mr Alexander Seitz (GESIS Leibniz Institute for the Social Sciences) - Presenting Author
Dr Silke L. Schneider (GESIS Leibniz Institute for the Social Sciences)
European societies today are characterized by a high degree of cultural and ethnic diversity. Yet, survey research on the construction and negotiation of socio-cultural and ethnic differences and their societal and political consequences, including ethnic inequality and discrimination, has often been hindered by a lack of available data on the ethnicity of respondents. The European Social Survey (ESS) has asked respondents to self-identify their socio-cultural ancestry within the framework of the European Standard Classification of Cultural and Ethnic Groups (ESCEG) in the past five survey rounds, but this variable remains poorly understood and is rarely used in published research. In this paper, we contribute to existing research by re-assessing the validity of the ESS ancestry variable as a measure of ethnic origin and/or immigration background using a larger dataset which for the first time includes data from both face-to-face and self-completion modes (web and paper). We investigate the measurement properties, validity, and results of the ancestry variable across countries in ESS Rounds 7-11 using descriptive statistics and multilevel regression, with a focus on immigrant groups. We find that the ancestry variable yields high-quality data that mostly overlaps with similar measures, such as country of birth or self-identified minority status, but offers a clearer and more detailed picture than these. However, processes of assimilation and lower salience of ethnic identity may result in underreporting of minority ancestries, especially among later generations of immigrant descendants. This underreporting may lead to case numbers too small for robust inference at higher resolutions of the ESCEG classification. It is weakly but significantly related to assimilation indicators, sociodemographic characteristics, political ideology, and survey mode.
Mr Matthieu Olivier (CDSP (Sciences Po/CNRS)) - Presenting Author
Mrs Emmanuelle Duwez (CDSP (Sciences Po/CNRS))
ELIPSS (Étude Longitudinale par Internet Pour les Sciences Sociales) is a French, probability-based, web panel for social sciences research surveys.
Set up in 2012, thanks to fundings granted by the French National Research Agency (ANR), the panel has been a self-financed infrastructure since 2020, that has evolved in light of past experiences while maintaining the fundamental principles of the project: providing the scientific community with the means to produce data to the highest methodological standards (randomly drawn sample, longitudinal nature), to develop methodological research (innovative projects, experiments), and to document and disseminate the data produced (reuse of data, cumulative research processes).
This poster will provide an opportunity to review in particular the different recruitment methods and their effectiveness (face-to-face, mixed methods, individual sampling, etc.), panel management to limit attrition (hotlines, incentives, newsletters, etc.) and changes in the devices used to respond to surveys and the impact on survey response times (over 50% of respondents now use their smartphone). Based on these findings, what are the challenges and opportunities for the coming years?
Ms Jennifer Kasabiiti/Asiimwe (makerere University) - Presenting Author
Background:
Participant and Public Involvement and Engagement (PPIE) is critical in aligning research with community priorities, fostering inclusivity, and maintaining ethical standards (Esmail et al., 2015). However, implementing PPIE in low-income countries like Uganda presents distinct challenges, including managing participant expectations, ensuring diverse representation, and providing timely feedback (Abimbola et al., 2021). These barriers often limit meaningful engagement, particularly for marginalized populations (Cornwall, 2008).
Objective:
This study examines the challenges of integrating PPIE in Uganda and identifies practical strategies to foster sustainable and inclusive research practices.
Methods: A mix of policy, programmatic, and research documents will be reviewed. Thematic analysis will identify key themes such as barriers to inclusion, ethical concerns, and mechanisms for feedback while gap Analysis will Compare existing policies and practices with global best practices in PPIE.
Results:
Preliminary findings reveal three major challenges: unrealistic participant expectations driven by financial constraints and perceptions of immediate benefits, difficulties in achieving diverse representation due to cultural, linguistic, and logistical barriers and limited feedback mechanisms caused by resource constraints. Proposed strategies include transparent communication to manage expectations, forming inclusive community advisory boards developing sustainable feedback loops, and capacity-building initiatives for researchers and participants
Conclusion:
Integrating PPIE in LICs requires context-specific strategies to overcome systemic and cultural barriers. Transparent, inclusive, and capacity-driven approaches are pivotal to fostering meaningful participant engagement, ensuring that research is ethical, relevant, and impactful.
Keywords:
PPIE, Low-Income Countries, Participant Engagement, Inclusive Research, Challenges and Solutions.
References
• Abelson, J., et al. (2003). Deliberations about deliberative methods: Issues in the design and evaluation of public participation processes. Social Science & Medicine, 57(2), 239-251.
• Abimbola, S., et al. (2021). The uses of participatory engagement in global health research: Lessons from the field. BMJ Global Health, 6(5), e006407.
• Cornwall, A. (2008). Unpacking 'Participation':
Ms Sigurbjörg Björnsdóttir (University of Iceland)
Dr Vaka Vésteinsdóttir (University of Iceland) - Presenting Author
Studies have shown lower participation rates among young adults in surveys, and efforts have been made to identify factors that can increase the likelihood of their participation. The aim of this study was to explore strategies for recruiting young participants into a research panel using insights gathered from interviews with young adults. Fifteen interviews were conducted to inquire into possible methods for encouraging participation in research. Interview participants primarily mentioned phone calls or emails as effective contact methods, though several also suggested social media. Traditional methods were considered appropriate but not necessarily likely to succeed. Sending an SMS before a phone call was generally thought to improve response rates. The vast majority believed that the chance to win a prize would increase participation. More participants preferred higher chances of winning smaller prizes over lower chances of winning larger ones, but there was considerable variation in what prizes were considered desirable. Due to difficulties in identifying people on social media, only traditional methods were used during the recruitment process. A random sample of 1,000 individuals aged 18–31 was drawn from the Icelandic national registry. Of the total sample, only 406 individuals had an active, registered phone number. Attempts were made to contact these 406 individuals and invite them to join a research panel under four conditions: 1) No SMS, no prize; 2) No SMS, with a prize; 3) SMS, no prize; and 4) SMS, with a prize. Among those reached (n = 291), approximately 50% agreed to participate. Neither sending an SMS before the call nor offering a prize for participation had a significant effect, contrary to the expectations of interview participants. In 110 out of 133 cases, a reason for refusal was provided, with the most common being lack of interest (49%) and lack of time (35%).
Mr Thibaud Ritzenthaler (French National Institute of Demographic Studies (INED)) - Presenting Author
The Generations and Gender Programme (GGP) is committed to producing high-quality metadata for its cross-national longitudinal survey, which covers important demographics topics such as partnerships, fertility, and work-life balance.To achieve this task, the GGP team has developed a workflow that aims to produce metadata following FAIR best practices and that is shared with the Data Documentation Initiative framework.
The poster will present the GGP's metadata documentation process, from raw data to explicit and comprehensive metadata, using a web platform based on DDI Lifecycle.The poster will also present the GGP's workflow, technical stack and quality control pipelines used to automate the process, ensuring high-quality metadata that meet international standards. These include the studies context and the thought process that led to the establishment of the organisation.
Mr Benjamin Gröbe (University of Mannheim/University Library) - Presenting Author
Ms Carolin Bahm (University of Mannheim/University Library)
Mrs Anne Balz (University of Mannheim/University Library)
Mr Tobias Rettig (University of Mannheim/University Library)
A large body of interdisciplinary literature has clearly demonstrated the negative effects of lengthy surveys. A shorter questionnaire, by contrast, could reduce the burden on respondents and again contribute to better data quality. Starting from this assumption, we investigate how the use of a modular survey design, in which a self-administered survey is split up into several parts that can be completed at the respondent's convenience, affects participation and module completion rates, order of starting modules and respondents’ evaluation of the questionnaire. For this purpose, we embedded a randomized experiment over three survey waves in the German Internet Panel (GIP), a long-standing probability-based online panel of the German population. We split the questionnaires into three to four topic modules, each of which took between three and 15 minutes to complete and randomly divided respondents into five groups, each receiving different information about the topic, length and incentives of the modules. While the respondents in group 1 received the questionnaire as a single survey in the usual way, group 2 was additionally shown the topics of the individual survey modules. Participants in groups 3, 4, and 5 were given an overview of all survey topics and were free to choose the order and timing in which they completed the different survey modules, but were given different additional information about the length and incentives of the survey modules. Our results indicate that respondents did not complete the survey more often when they could choose the order in which they answer the modules, even when they knew the length and incentive next to the topic of each module. Rather than determining the order in which modules should be answered, respondents showed a clear preference for answering all survey modules in the given order at one time.
Mrs Caroline Poppa (SHARE BERLIN Institute GmbH) - Presenting Author
Dr Désirée Nießen (GESIS Leibniz Institute for the Social Sciences Mannheim)
Dr Jessica Daikeler (GESIS Leibniz Institute for the Social Sciences Mannheim)
Professor Henning Silber (University of Michigan, Ann Arbor)
Dr Bernd Weiß (GESIS Leibniz Institute for the Social Sciences Mannheim)
Professor David Richter (SHARE BERLIN Institute)
Publication bias is the prioritized and selective reporting of scientifically significant results and stems from the assumption that significant research findings are rated as “better”, more “valuable” or “publishable” than non-significant or null findings. This assumption may lead researchers to either a) change their hypothesis/expectations post-hoc (HARKing), b) publish only their significant results without mentioning other findings below standard statistical thresholds (selective reporting), or c) not publish any of their results at all (dissemination bias). Ultimately, the scientific evidence across entire fields becomes distorted, with a focus on significant findings, whilst lacking the valuable insights gained from accepting null or insignificant results. This can be problematic because time and effort are invested in repeating research that has already been conducted, but due to the nature of its results, has never been shared. In psychology, the consequences of this selective publication practice became poignantly evident during the 'replication crisis' of the past decade.
In line with Franco et al. (2014), we assessed the prevalence of publication bias in 189 studies conducted in two representative German academic panels from 2012 to 2021. Our findings revealed that, after at least three years of data availability, over half of the studies (60.1%) had not yet resulted in a journal article. Among those studies that led to journal articles, less than half (33.8%) included hypotheses from the original study submission, while the remaining 66.2% published only new, previously unmentioned hypotheses. This poster will cover first insights on risk factors for publication bias that we have assessed in our subsequent author survey, as well as a comparative analysis of publication bias prevalence across different scientific disciplines (psychology, economics, political science and sociology).
Professor Rory Fitzgerald (European Social Survey ERIC (City, St George's University of London))
Dr Rene Bautista (NORC)
Dr Gianmaria Bottoni (European Social Survey ERIC (City, St George's University of London)) - Presenting Author
Dr Leah Christian (NORC)
Professor Ben Edwards (Australian National University)
Professor Matthew Gray (Australian National University)
Dr Eric Harrison (ESS ERIC (City St George's University of London))
Professor Jibum Kim (Sungkyunkwan University)
Ms Lorean Ma (ESS ERIC (City St George's University of London))
Many grand challenges require a cross-national perspective to be fully understood, such as climate change, health, immigration and others. Many of these social and economic phenomena transcend borders, requiring collaborative solutions to address them among multiple nations. In addition, variation across different countries provides unique opportunities to study individuals’ behaviour and attitudes within different political, institutional and cultural contexts.
There are already several established web panels (e.g. NatCen panel in GB, Life in Australia Panel, AmeriSpeak panel) but these are designed for national use. Meanwhile the European Social Survey has been developing the world’s first, input harmonised, cross-national ‘web-first’ panel (CRONOS) to complement the main ESS, and has successfully included 12 European countries thus far (Bottoni and Fitzgerald, 2021).
The ESS team are now investigating the possibility of adding countries outside of Europe to the panel. The aim of this cross-national panel will be to provide infrastructure that facilitates comparison between Europeans and those in other developed countries outside Europe. Initially it is being investigated if Australia, Korea and the USA could be added. However, the team are also seeking to identify and engage other potential countries that could be part of a larger panel. The scoping covers a wide range of considerations related to the survey life cycle, as well as organisational, governance and funding issues.
Adding countries outside of Europe would provide a unique resource in the form of an academically led, probability based, input harmonised cross-national social survey, including countries across multiple continents. Such a resource has the potential to provide data faster than current international survey programmes allow and facilitate longitudinal data collection at the cross-
national level which is rather rare. In this paper the authors will outline the challenges and opportunities from expanding the panel beyond Europe.
Miss Mette Kurtzhals (Department of Nutrition, Exercise and Sports, University of Copenhagen) - Presenting Author
Miss Paulina Sander Melby (University of Southern Denmark, Department of Sports Science and Clinical Biomechanics)
Mr Peter Elsborg (Center for Clinical Research and Prevention, Copenhagen University Hospital – Bispebjerg and Frederiksberg)
Mr Peter Bentsen (Center for Clinical Research and Prevention, Copenhagen University Hospital – Bispebjerg and Frederiksberg)
Mr Glen Nielsen (Department of Nutrition, Exercise and Sports, University of Copenhagen)
Purpose: Many important aspects of wellbeing and quality of life are subjective experiences and therefore require self-report. The KIDSCREEN-27 questionnaire is widely used for this purpose. However, the self-report versions have mainly been validated for children aged 12 to 18 years. This study aims to develop a video-assisted format the KIDSCREEN-27 that enable self-report of wellbeing by children aged six to ten years and to test its psychometric properties. Methods: The Danish-translated version KIDSCREEN-27 was slightly adapted in wording and items (N=12) and a video-format, including audio, illustrations, and smiley-supported scales, was developed, and tested. Next, a psychometric investigation of this version (KIDSCREEN-VIDEO) was conducted on 788 Danish children aged six to ten years (49.8% girls). Results: Confirmatory factor analysis showed an acceptable to good model-fit: ꭓ2 =727.053; df =242; P <0.001; root mean squared error of approximation=0.05; the comparative fit index=0.98; and the Tucker-Lewis index=0.98, and factor loadings ranged from 0.40 to 0.88. Cronbach’s alpha values ranged from 0.65 to 0.89, suggesting acceptable to good internal reliability of the scales. Linear mixed model analyses, and Pearson’s r correlation coefficients showed positive associations with the global and physical self-worth scales, indicating convergent validity. The test for measurement invariance indicated the model fit for the five-factor model was consistent across sex and age groups. Conclusion: Based on our results, the KIDSCREEN-VIDEO provides a promising self-reported measure for wellbeing among children aged six to ten.
Mrs Marie Caroline Vermund (Haver til Maver & University of Copenhagen & Center for Clinical Research and Prevention, Copenhagen University Hospital) - Presenting Author
Mr Glen Nielsen (University of Copenhagen)
Mr Peter Elsborg (Center for Clinical Research and Prevention, Copenhagen University Hospital)
Background: The World Health Organization (WHO) has declared climate change as one of the greatest health threats to humanity. Current trajectories in climate change thus indicate that chil-dren of today, regardless of their residence, face a future in which they will need to be prepared to navigate the increasingly complex challenges related to climate change. This emphasizes a need for future generations to be willing, able, and motivated to engage in individual and societal actions to mitigate climate change. Enhancing children’s climate change literacy could affect the whole process from individual awareness to public engagement. There is thus a growing call for educational activities to fully enable schoolchildren to acquire the knowledge, skills, values, and attitudes needed to contribute effectively to the mitigation and adaptation of climate change. Scales to measure related constructs exist, such as environmental attitudes or climate knowledge. However, to our knowledge, there exists no validated survey integrating a holistic range of ele-ments specific to children’s climate change literacy.
Purpose: We aim to move beyond constructs assessing climate change knowledge or climate change attitudes by itself and develop and validate a holistic and age-appropriate instrument to measure all domains of climate change literacy in schoolchildren between 9 and 12 years.
Methods: An electronic survey was developed, and pilot tested on convenience sample of the target group. The questionnaire was then administered to a large sample of Danish schoolchil-dren aged 9 to 12 years. Tests of dimensionality, reliability and validity will be performed.
Expected results: We aim to contribute with a validated survey measuring a broad range of dimen-sions relevant to climate change literacy in children aged 9 to 12 in a Danish context. The study will thus provide a new and needed tool to measure children’s climate literacy.
Dr Susanne de Vogel (Data Science Center, University of Bremen) - Presenting Author
Dr Lena Steinmann (Data Science Center, University of Bremen)
Professor Rolf Drechsler (Data Science Center, University of Bremen)
The requirements for data collection and analysis in survey research are becoming increasingly complex. Alternative data sources, such as social media and administrative records, as well as AI-based methods, are transforming the field by offering new opportunities while posing significant challenges. Researchers must navigate diverse data sources, ensure data quality, and address ethical and legal concerns.
Surveys are conducted across various disciplines, each with unique requirements. As part of the U Bremen Research Alliance – a network of the University of Bremen and 12 non-university research institutes – and in collaboration with other regional partners, the data competence center “DataNord” equips researchers with the data literacy skills needed to tackle the evolving challenges of modern (survey) research. One key component is the Data Science Center (DSC) of the University of Bremen, where we built up an interdisciplinary team of data scientists to develop trainings and consultation offers as part of the DataNord help-desk. There, we provide tailored support to survey researchers with diverse methodological backgrounds and all skill levels.
Our contribution to DataNord comprises:
• Workshops and training: Topics include survey design, (survey) data management, integration of alternative data sources, reproducible research, and AI-based analysis.
• Individual consultations: Customized support for methodological challenges in (survey) research along the whole data life cycle.
• Interdisciplinary Collaboration: Facilitating networks to share innovative approaches and best practices.
The poster showcases examples and best practices of how the interdisciplinary DataNord help-desk supports researchers across disciplines, from beginners to advanced practitioners, in addressing the complexities of modern survey research. This contribution invites the survey research community to share insights, exchange experiences, and explore collaboration opportunities to strengthen data literacy as a core competency across the social sciences and beyond.
Dr Gamze Özel (Hacettepe University) - Presenting Author
Teaching statistics in nature provides an innovative approach to reinforcing theoretical knowledge through practical applications. This study presents the outcomes of a project designed to enhance students' statistical thinking skills by engaging them in field-based activities. Participants collected data on biodiversity, water quality, and air measurements in forested areas and analyzed these datasets to learn statistical methods in a hands-on environment.
In the second phase of the project, a survey was developed to assess students' attitudes toward statistics, their experiences with nature-based learning, and their overall satisfaction with the program. The survey development process included expert reviews for content validity and a pilot study for refinement. Findings indicate that learning statistics in natural settings increases students’ interest in the subject and makes the learning process more tangible and meaningful.
The results of this study can guide educators seeking innovative approaches to teaching statistics. Furthermore, the analysis of project outcomes lays the groundwork for evaluating the broader impacts of nature-based statistics education on learning motivation and skill development. The developed survey is recommended for use in diverse contexts to enable comparative analyses of its findings.
Miss Meenakshi Sarasa (National Institute of Pharmaceutical Education and Research (NIPER), Hajipur, Vaishali-844102, Bihar, India.) - Presenting Author
Miss Triveni Bahekar (National Institute of Pharmaceutical Education and Research (NIPER), Hajipur, Vaishali-844102, Bihar, India)
Dr Nitesh Kumar (National Institute of Pharmaceutical Education and Research (NIPER), Hajipur, Vaishali-844102, Bihar, India)
Dr Ved Prakash (Indira Gandhi institute of medical sciences (IGIMS) Bailey road, Sheikhpura, Patna- 800014, Bihar, India)
Dr Krishna Murti (National Institute of Pharmaceutical Education and Research (NIPER) Hajipur, Vaishali-844102, Bihar, India.)
Objective
Fluoride exposure is a global public health concern. Understanding the knowledge, attitudes, and practices (KAP) of affected populations is essential for effective community management. This study aimed to develop and validate a KAP questionnaire to assess fluoride and its risk in general population.
Methodology
An extensive literature review and focus group discussions were conducted to construct the questionnaire. Content validity was assessed using the Content Validity Index (CVI) based on expert feedback. Factor analysis was performed for final tool validation, and item characteristics were analyzed using IBM SPSS v. 27 and IBM AMOS v. 26.
Results
A total of 300 responses were collected. Initially, 41 items were included in the questionnaire, which were reduced to 25 after expert review. The final version included 19 items, with an I-CVI ranging from 0.80 to 1.00, indicating no issues with item difficulty or discrimination. Cronbach's alpha ranged from 0.88 to 0.90, demonstrating good internal consistency. The Kaiser-Meyer-Olkin (KMO) value was 0.848, and Bartlett’s test (χ² = 6860.978, df = 156, p < 0.01) confirmed data suitability for factor analysis. Three constructs were extracted with factor loadings greater than 0.5. Confirmatory factor analysis demonstrated a good model fit.
Conclusion
This study developed and validated a robust 19-item KAP questionnaire for assessing knowledge, attitudes, and practices related to fluoride exposure. The tool demonstrated excellent reliability, validity, and internal consistency, supporting its use in guiding effective community-level management and public health interventions in fluoride-endemic areas.
Professor Yoosung Park (The Institute of Statistical Mathematics) - Presenting Author
Professor Tadahiko Maeda (The Institute of Statistical Mathematics)
Professor Daichi Mochihashi (The Institute of Statistical Mathematics)
In Japan, response rates have declined significantly due to the deteriorating survey environment, particularly in the traditional survey method of face-to-face interviews. Meanwhile, many empirical studies in Europe and the United States have proposed survey methodologies to improve response rates and encourage a departure from the existing framework (Couper 2013; Dillman 2017; Dillman et al. 2014; Groves 2011). The current survey situation in Japan undeniably lags behind the global standard, as few researchers in Japan have conducted experiments on survey methodologies (e.g., Kojima 2010). In other words, traditional survey methods are at a crossroads in Japan owing to decreasing response rates and survey participation.
Therefore, this study developed a novel survey method utilizing a Web-based voice recognition survey app as “interviewer” that replaces humans to conduct large-scale surveys for probability samples following the same procedure as conventional interview surveys. This study entails the development of the survey process and evaluating the practical feasibility of the new survey methodology. Specifically, to examine, we conducted a large-scale survey for probability sampling similar to the conventional interview survey procedure based on the Survey on the Japanese National Character. The Survey on the Japanese National Character is one of Japan's leading large-scale social surveys that has been conducted nationwide every five years since its launch in 1953. The survey is conducted through face-to-face interviews using stratified multistage random sampling.
Overall, the response rate of the new survey method was not as high as that of the usual interview surveys. Furthermore, this study shows that the response rate was low compared with the login rate. Consequently, more attractive survey participation features should be developed to retain participants’ responses from the time they log in until they complete the survey.