COVID-19 unearths several important gender dimensions and implications before, during, and after data collection which Chi-Chi Undie, Nicole Haberland, Sanyukta Mathur, Isabel Vieitez and Julie Pulerwitz of the Population Council explore in this blog.
In many ways, the COVID-19 pandemic has changed the world as we knew it – including the world of data collection. New global realities have extended the world’s experience and enriched its vocabulary, normalizing terms that have new meaning for data collection, such as ‘lockdowns,’ curfews,’ ‘physical distancing,’ and ‘economic downturns.’ There is no question that these new realities have challenged us. In every area of life, we have been compelled to do things differently. As researchers, we have had to re-examine, re-think, and reconstruct taken-for-granted data collection processes. And in a matter of months after the onset of COVID-19, an impressive repertoire of guidance documents on data collection in these times has emerged.
Among these documents are some that lay out the various gender considerations that should inform decisions about if and when data collection should take place – particularly, data collection associated with sexual and gender-based violence (SGBV). But, there is relative silence in regard to gender considerations once a decision has been made to collect data. Yet, COVID-19 unearths several important gender dimensions and implications before, during, and after data collection.
We outline several of these dimensions here as a reminder of just how gendered data collection is during COVID-19 and similar disasters, and in times of normalcy. The gender dynamics that we highlight are important in the first place because gender serves as a reminder of how societies construct and assign roles, identities, and power to different data collection stakeholders (or of how stakeholders do so themselves). These modalities of ‘doing gender’ can occur in ways that could negatively or positively impact respondents’ and data collectors’ participation in the research process, the quality of the data collected, and the reproduction of inequality. In paying attention to gender concerns, the goal is to render data collection as gender-equitable as possible for all concerned (including participants, data collectors, and institutions involved in collecting data).
What to think about when it comes to training
Think about who your data collectors are going to be, and the gender dimensions that would impact training sessions during COVID-19. How do lockdowns and other relevant pandemic characteristics and mitigation efforts impact your data collectors, given their gender roles? For instance, it is likely that parenting data collectors (regardless of gender) will have childcare responsibilities that need to be taken into account. Unlike pre-pandemic times when caregiving responsibilities (whether for children, older people, the ill, etc.) of potential data collectors might not have arisen, these statuses are now important considerations in order to ensure that the training process is sensitive to the realities of data collectors in the unusual circumstances engendered by COVID-19. The goal is not to discriminate against data collectors with a certain status or identity, but rather, to play a facilitative role in ensuring that enabling conditions are in place for all qualified data collectors.
Taking such issues into account would involve pre-training discussions with data collectors to understand their constraints, and to devise workable solutions in advance. Examples of such solutions might include covering or contributing to childcare costs during the training and/or data collection period, and choosing the timing of training sessions with data collectors’ caregiving responsibilities in mind (i.e., selecting times that are most suitable for them). Possible solutions for virtual training sessions could include building in breaks to allow for attention to childcare, while maintaining an informal, flexible atmosphere in which interruptions by children, for example, are anticipated and taken in stride (C. Gould, personal communication, August 7, 2020).
The content of training also benefits from a strong gender lens. In addition to ethical protection of human subjects, and infection prevention guidelines and procedures during a pandemic, devoting training time to help interviewers critically reflect on how gender and power inequalities are affecting their own experience of the pandemic and the research process, provides an opportunity to shift toward more equitable gender attitudes. Space for such critical reflection can also help disrupt the transmission of inequality in ways small and large. For example, training content can include reminders of unacceptable sexist, homophobic, and transphobic language, as well as warning signs of violence experiences (e.g., intimate partner violence) among survey participants that should be followed up with an explicit referral.
Given the intrinsic linkages between gender and power, gender-informed training can also help trainees recognize the inherent power dynamic between interviewer and interviewee, and how this can be exacerbated by crises such as COVID-19. Interviewers need to be particularly mindful of this reality during the pandemic, as well as respectful toward interviewees in a time when many face increased vulnerabilities. Data collection teams can have inhibitive power dynamics of their own (e.g., imbalances in power between data collection team leaders/coordinators and actual data collectors), which training sessions can deliberately unearth and address to ensure that trainees feel comfortable with being honest about these realities, as well as about how to resolve them to ensure a successful and efficient data collection exercise.
Who to collect data from
An abundance of guidance is now available on who not to collect data from (e.g., SGBV survivors), based on some of the important gender-related concerns that COVID-19 brings up. While this guidance tends to take SGBV as its point of departure, many of its facets are helpful for thinking through the ethical issues associated with gender-related research in general. However, it is also important to think about whom you actually can collect data from, based on existing evidence of their level of autonomy, and, therefore, a lower likelihood of facing repercussions due to their participation as respondents.
Disaggregating data by gender can help to understand who is more or less likely to be at risk, and exposed to harm (e.g. violence) in the time of COVID-19 by isolating and comparing the experiences of various gender categories. To this end, where possible, gender disaggregation should go beyond ‘female and male’ to include the various identities that these broad gender categories encompass. To safely generate data on gender issues during COVID-19, populations of focus could include female household heads, emancipated minors, and LGBTI+ communities, for instance. Sample sizes should be determined to allow for gender disaggregation – for both quantitative and qualitative studies.
How to collect data
Data collection modalities are naturally dependent on the contextual issues that define the data collection site, including movement restrictions; lockdown situations; and literacy levels, technological device access, and electricity and internet availability where the targeted respondents are concerned. Online/virtual methods are increasingly relied upon during COVID-19 in response to these contextual factors, and service provision sites have been highlighted as contexts that are amenable to in-person data collection with service-seekers during the pandemic. In thinking through appropriate data collection sites (which inform the modalities for collecting data), key questions to ask include: In what ways are possible data collection sites gendered (e.g., in some refugee camp sites, nutrition centers are viewed as settings for women alone, making them safer for data collection involving women)? Which sites are likely to assure privacy for all, regardless of gender (e.g., HIV pre-test and post-test counseling settings)?
Data collection plans should consider that potential participants’ time has been impacted – sometimes radically so – due to their shifting or additional gender roles and responsibilities in the time of COVID-19. As such, data collection tools should be as brief as possible. Given the therapeutic and empowering potential of sharing nuanced information about personal stories, concerns and priorities (which are key features of qualitative data collection methods), ways of incorporating such methods into data collection tools should be taken into account. This can be achieved by incorporating open-ended questions into quantitative tools, for instance. Furthermore, to allow for a detailed understanding of the intersections between COVID-19 and gender, data collection instruments themselves would benefit from being less gender binary, with questions and response categories framed with inclusivity (rather than heteronormativity) as their point of departure. This could entail, for example, employing non-presumptuous language for exploring respondents’ lives: providing a wider range of gender response options beyond ‘female’ and ‘male’ alone, or using terms such as ‘partner’ instead of ‘wife/husband/girlfriend/boyfriend’ to examine participants’ intimate partnerships.
Which kinds of support and services are part of the process?
The disruptions caused by COVID-19 make it more important than ever before to draw on ‘compassionate’ research methods – methods that are ‘not only driven by caring for others and a desire for improving their lot,’ but that ‘actually immerse the researcher in compassionate work.’ As part of an ethos of compassion, more thought needs to be given to incentives that participants might receive for their time or trouble. In times of crisis, researchers must be more willing to build incentives into their data collection processes. Where incentives are not monetary, ensuring that the items received by participants are tailored to their gender-determined needs during the pandemic should be regarded as a critical part of data collection. Study participants that have children, for instance, can almost always benefit from some form of childcare support. Personal protective equipment, such as face masks and hand sanitizer would also be appropriate.
Beyond incentives, free services that might be required to attend to participants’ needs (dietary, economic, psychosocial, etc.) as a consequence of COVID-19 merit attention and should be built into the data collection process. As service-seeking is often gendered, it would be beneficial to consider (and to provide) tailored support. For example, men and LGBTI+ people might be less likely than heterosexual cis women to seek services, for various reasons. Consequently, men might require information about and referrals to male-friendly services, while LGBTI+ people might need to be connected to service sites that have had their capacity built to avoid stigmatizing this population.
This applies to data collectors, too, who might need psychosocial support after being immersed in the realities of marginalized populations during fieldwork in COVID times. Similar to participants, the ability of data collectors to seek such support might be gendered, while also being informed by age, class, etc. Ensuring that support for data collectors is tailored according to these realities would be important.
At the most basic level, pre-data collection scoping of available referral options is essential. During crises such as COVID-19, communities suffer devastating job losses and food shortages, and an increase in violence rates, often exacerbated by pre-existing inequalities due to gender, age, class, caste, and geography, for instance. Connecting respondents to available relief efforts or appropriate services can help. Confirming referral destinations’ operation, accessibility, and full contact information, and providing this information to respondents is fundamental to ethical data collection.
The bottom line
Crises rarely affect all people equally. Their worst impacts are often gendered, and felt more by the vulnerable and less by the privileged. Trying to redress this inequality is always a good strategy and the focus of many innovative programs. It is no less urgent to do so in research and data collection. All of us who are involved in the latter are not merely objective observers, but active participants in a gender (and gendered) system.