Survey data for sustainable development research: a call for improved data quality

High-quality data are essential for studying sustainability, development and socio- environmental transitions, yet many influential institutional datasets – particularly large official cross-national surveys – present structural weaknesses that can compromise empirical analyses and misinform policy design. These limitations are especially consequential in sustainability science, where public attitudes toward climate change, energy transition and environmental behaviour are prone to normative pressure and rhetorical consensus. This paper examines the reliability, limitations and common biases affecting survey data used in sustainable development research, with a particular focus on official cross-national surveys. Drawing on critiques of international indicators and survey-method research, we discuss how methodological choices – question wording, translation, interview mode and self-reporting biases – shape the quality of data on sustainability attitudes and behaviours. We then propose a methodological agenda to reduce measurement artefacts through enhanced transparency, mode adjustments, improved questionnaire design and methodological triangulation (anchoring vignettes, list experiments, behavioural validation). These recommendations aim to support a more bias-aware sustainability science that moves beyond rhetorical consensus toward empirically grounded inference.