The COVID-19 pandemic in early 2020 caused unprecedented disruptions globally, significantly impacting universities and research. While some scientists focused on COVID-19 research, many faced substantial obstacles due to campus closures, limited access to labs and resources, and the shift to online teaching. Previous research, primarily from the US and Europe, highlighted decreased research productivity and increased workload. However, evidence from Latin America was lacking. This study addresses this gap by examining the impact on researchers in Chile and Colombia, two countries with significant research output and stringent lockdowns. The researchers aimed to understand the overall effect on research time, exploring disparities across disciplines and based on individual characteristics. A unique aspect of this research was the investigation of the relationship between time devoted to research and personal beliefs about gender roles, considering the existing gender inequalities in Latin American households and universities. The research questions were: (a) What was the overall impact of the pandemic on research time among Chilean and Colombian researchers? (b) Are there unequal effects at the discipline and individual level? (c) Is there a relationship between losses in research time and personal beliefs about gender roles?

Existing research indicates significant disruption to scientific work due to the COVID-19 pandemic. Studies from the US and Europe showed a decline in total working hours for 55% of scientists, with research activities experiencing the most substantial decrease (24%). Disciplinary differences were also observed, with "bench sciences" (requiring lab work) more affected than "desk sciences". Gender disparities emerged, with women facing more significant challenges, potentially due to their primary responsibility for household chores and childcare. A "motherhood penalty" was noted, where women with young children experienced larger reductions in research time. The impact on early career researchers with less stable contracts was also a concern. However, this literature primarily focused on high-income countries, leaving a gap in understanding the situation in Latin America, particularly considering its unique context of precarious workspaces, comparatively lower investment in research, and stringent lockdowns.

This study used an anonymous online survey administered in July-August 2020 to researchers in Chile and Colombia. Researchers from leading universities in both countries were targeted, resulting in 3257 valid responses. The survey employed items adapted from previous research and the International Social Survey Programme (ISSP) to assess gender-role beliefs. The dependent variable was the percentage change in weekly research hours before and during the pandemic. Independent variables included discipline, gender, age, family composition (children under 12, older adults in the household), partner status, type of contract, and academic position. An egalitarianism index was created using three questions to measure personal beliefs about gender roles in the household. Analysis of variance (ANOVA), Bonferroni tests, and Lasso regression were used to analyze the data. The Lasso regression helped to identify the most significant predictive variables of changes in research time. The study acknowledges limitations associated with the non-probabilistic sampling and reliance on self-reported data, employing the CHERRIES checklist to mitigate these issues.

Overall, 54.1% of researchers experienced a decline in research time, with an average decrease of 16.97%. Disciplinary differences were significant, with "bench sciences" showing the largest declines, and health sciences reporting the lowest. Women experienced a significantly greater decline (22.13%) compared to men (12.42%). Having children under 12 at home was a significant factor, particularly for women. Women with at least one child under 12 experienced a 34.49% reduction, rising to 43.97% for those with three or more children. Lasso regression analysis confirmed the importance of field-level differences, with "bench sciences" still showing significant declines, while some health sciences and arts disciplines showed increases. Being a woman was associated with a 5.35% decrease in research time, even after controlling for other factors. Having children under 12 remained the most significant factor, with having three or more children associated with a 12.89% decrease, while having no children meant an average increase of 16.53%. The study also found that having a partner and two or more adults at home were disadvantages for women. Interestingly, while no teaching constituted a relative advantage, particularly for female researchers, teaching two or more classes negatively impacted research time. The analysis found no significant difference between the type of contracts and the level of impact. Regarding the Egalitarian Index, while no overall statistical difference was found, a significant difference was revealed when analyzing by gender, with conservative women experiencing a substantial decrease in research time compared to conservative men.

The findings highlight the significant and unequal impacts of the COVID-19 pandemic on researchers in Chile and Colombia. The results are consistent with previous studies from higher-income countries, demonstrating the persistence of gender inequalities and the burden of childcare on women's research productivity. The larger impact on women in this study compared to previous research may be due to higher numbers of households with multiple young children. The inclusion of gender-role beliefs provides new insights, suggesting that traditional views on gender roles exacerbate the negative impact on women. This is particularly noteworthy within the Latin American context, known for its relatively traditional gender roles. The study contributes to a growing body of literature emphasizing the need for targeted policies to support researchers, especially women with care responsibilities, to mitigate the effects of similar crises in the future.

This study reveals the disproportionate impact of the COVID-19 pandemic on female researchers and those with childcare responsibilities in Chile and Colombia. The findings underscore the need for universities and research institutions to implement targeted policies, considering gender and family responsibilities, to support researchers and mitigate future disruptions. Further research could explore the long-term effects of these disruptions on research outputs and career trajectories, and explore more effective interventions.

The study's limitations include its non-probabilistic sampling method and the reliance on self-reported data, potentially subject to recall bias. The study's focus on Chile and Colombia limits the generalizability of the findings. Future research using more robust sampling methods and incorporating multiple data sources could address these limitations.