Revealing the Hidden: The Pitfalls of Conventional Surveys in Authoritarian Contexts

Description

Understanding what people think about sensitive political and social topics in authoritarian societies is always challenging. In free societies, public opinion surveys conducted via phone or in-person interviews are a rigorous method for capturing attitudes and preferences within the population. However, in repressive societies, expressing opinions on sensitive topics can be dangerous, as such views need to remain "non-public" for safety. The challenge, then, is to develop methods for measuring and revealing "non-public opinions" in these societies, allowing people to express genuine attitudes without fearing repercussions from the regime.
The demand for insight into non-public opinions in influential authoritarian countries has driven institutions and researchers to rely on conventional sampling methods, often without considering their limitations or potential for misleading results. Recognizing this demand, many authoritarian regimes either invest in conventional polling as a propaganda tool or appreciate the results of international pollsters, using indicators like high electoral turnout and high approval rates for their leaders or policies to bolster legitimacy.
This study examines the results of three conventional surveys conducted in Iran: the World Values Survey in 2020, Gallup in 2021, and IranPoll, a Canada-based organization, in 2021. The paper presents and compares these findings with those from large-scale, well-designed, and carefully adjusted anonymous online surveys by GAMAAN among Iranian residents in 2021–2022, replicating questions from the aforementioned surveys. The findings reveal a significant discrepancy—about 50 percentage points—for various sensitive questions, while showing strong consistency on non-sensitive topics.
Exploring whose public opinions are actually reflected in conventional surveys, we found that these results closely align with the attitudes of the pro-regime sub-sample from the online surveys. The study’s findings have important implications for survey research in authoritarian contexts.

Period	21 Jan 2025
Event title	DIGITAL METHODS IN RESEARCHING AUTHORITARIAN CONTEXTS: TOOLS, RULES, AND LIMITATIONS
Event type	Workshop
Location	Amsterdam, NetherlandsShow on map
Degree of Recognition	International