[Paper Review] Gender Imbalance and Spatiotemporal Patterns of Contributions to Citizen Science Projects: the case of Zooniverse

Khairunnisa Ibrahim, Samuel Khodursky+1 more|arXiv (Cornell University)|Jan 7, 2021

Species Distribution and Climate ChangeEnvironmental Science45 references19 citations

TL;DR

This study analyzes spatiotemporal and gender patterns in contributions to Zooniverse, a major citizen science platform, using 54 million classifications from 340,000 volunteers across 198 countries (2009–2013). It reveals significant geographical inequality driven by socio-economic factors, high evening activity bursts, and a persistent gender gap (30% female), with higher representation in life/nature sciences and potential informal access for women in underrepresented research regions.

ABSTRACT

Citizen Science is research undertaken by professional scientists and members of the public collaboratively. Despite numerous benefits of citizen science for both the advancement of science and the community of the citizen scientists, there is still no comprehensive knowledge of patterns of contributions, and the demography of contributors to citizen science projects. In this paper we provide a first overview of spatiotemporal and gender distribution of citizen science workforce by analyzing 54 million classifications contributed by more than 340 thousand citizen science volunteers from 198 countries to one of the largest citizen science platforms, Zooniverse. First we report on the uneven geographical distribution of the citizen scientist and model the variations among countries based on the socio-economic conditions as well as the level of research investment in each country. Analyzing the temporal features of contributions, we report on high "burstiness" of participation instances as well as the leisurely nature of participation suggested by the time of the day that the citizen scientists were the most active. Finally, we discuss the gender imbalance among citizen scientists (about 30% female) and compare it with other collaborative projects as well as the gender distribution in more formal scientific activities. Citizen science projects need further attention from outside of the academic community, and our findings can help attract the attention of public and private stakeholders, as well as to inform the design of the platforms and science policy making processes.

Motivation & Objective

• To understand the geographical distribution of citizen science contributors on the Zooniverse platform.
• To examine temporal patterns of participation, including daily and weekly activity rhythms.
• To investigate the gender imbalance among contributors and compare it with formal scientific fields.
• To identify socio-economic and research investment factors influencing national-level participation disparities.
• To explore whether citizen science serves as an alternative pathway for underrepresented groups, particularly women, in science.

Proposed method

• Analysis of 54 million classification records from 17 Zooniverse projects between November 2009 and June 2013.
• Geospatial mapping of contributors using two-letter country codes derived from user-provided approximate locations.
• Correlation and regression analysis linking country-level socio-economic indicators (GDP per capita, Internet users, school enrolment) to participation rates.
• Temporal analysis of classification timestamps to identify daily and weekly activity patterns, including circadian rhythms.
• Gender classification using the Gender API based on first names from user registrations, with validation against known gender distributions in formal science.
• Comparison of gender ratios across projects and countries to assess field-specific and regional disparities.

Experimental results

Research questions

• RQ1 What is the geographical distribution of Zooniverse contributors, and how do national socio-economic factors influence participation?
• RQ2 What are the temporal patterns of citizen science participation, and when are volunteers most active during the day and week?
• RQ3 What is the gender distribution among Zooniverse contributors, and how does it compare to formal scientific communities and other collaborative online projects?
• RQ4 To what extent do national disparities in formal science participation correlate with participation in citizen science?
• RQ5 Do certain scientific disciplines on Zooniverse show more balanced gender representation, and if so, why?

Key findings

• Over 80% of the variation in national-level participation rates on Zooniverse can be explained by a country's population, GDP per capita, Internet connectivity, and school enrolment rates.
• National research and development investment explains 73% of the variation in participation, indicating a strong link between scientific infrastructure and public engagement.
• Volunteers are most active during evening hours (typically 6–10 PM), suggesting a leisure-time participation pattern aligned with household routines.
• The gender distribution among Zooniverse contributors is skewed, with approximately 30% identifying as female—higher than in Wikipedia or open-source software but still significantly imbalanced.
• Countries with larger gender imbalances in formal science (e.g., India, Iran, Ukraine) show relatively higher proportions of female contributors to Zooniverse, suggesting it may serve as an informal access point.
• Project-level gender distribution varies significantly: wildlife and nature projects (e.g., Snapshot Serengeti, Whale FM) have over 50% female contributors, while astronomy projects (e.g., Galaxy Zoo, Planet Hunters) have less than 20% female contributors.