Better data do not necessarily produce better knowledge. Time Use Surveys and an innovative sequence method, Dynamic Hamming Matching (DHM), have transformed our capacity to examine the organisation of work time in far greater detail, recording work hours across the day and workdays across the week rather than relying on people’s estimates. Yet, despite these advances, women’s working time remains unclear. I came to this literature with what seemed like a simple question: when do women work? The problem was not simply one of data availability. It was also a set of analytical conventions that continue to position men’s working time as the standard against which women’s is compared, and treat pooled, gender-comparative analysis as the neutral starting point. This piece examines three moments from my own research on women’s work schedules, where those unreflective habits become visible, and argues that focusing analysis on women – as economic actors in their own right – is not a narrowing of enquiry but what the question demands.
‘A typical female pattern’
A study using DHM to identify a typology of work schedules in Belgium labelled one schedule a ‘typical female pattern’ because 68.5 per cent of those working it were women. The label sounds straightforward, but it is not.
A schedule can be majority female without being typical among women, and these are entirely different claims. The label tells us about the composition of the cluster, not about what women characteristically do. It tells us who works that schedule without telling us which schedule is most prevalent among women. Despite incorporating gender into the analysis, the study leaves a critical empirical question unanswered: when do women actually work?
This reflects a recurring feature of much time use research. Gender is included as a variable in models, and noted in findings, but it does not necessarily shape the questions being asked or how results are interpreted. Including gender and being attentive to the gendered organisation of work time are not one and the same.
‘Women continue to catch up with men’
A second study concludes that women are catching up with men in the amount of time spent in paid work. The phrase sounds neutral, even positive. But it subtly reveals assumptions that require scrutiny.
This interpretation implicitly frames men as the standard against which women are measured and towards which women must progress. Women’s work time is evaluated in relation to men’s. The question of what women’s work time looks like on its own terms is not entertained.
Comparison is a powerful analytical tool and much valuable research on gender inequality has followed from this approach. But comparison is not neutral. It directs attention towards gaps and convergence and away from variation within groups. When the comparison is always women against men, the analytical frame risks reproducing the very hierarchy it intends to critique. Comparison often means the male norm is left intact and the language is used to imply progress.
‘One might wonder whether a one-hour difference is that important?’
A reviewer queried whether workdays ending at 5pm were meaningfully different from those finishing at 6pm. From a modelling perspective, the question made sense. Parsimony is a defensible aim. But from the perspective of anyone responsible for collecting a young child from nursery and then doing the tea, bath and bed routine, the question had an obvious answer.
Subsequent analysis showed that this small difference in clock time corresponded to a significant difference in how work and care can be organised. Women with young children were more likely to have a workday schedule that finished at the earlier time than women without children. One work schedule makes the evening care routine possible; the other, for many women, does not.
Decisions about which distinctions to preserve and which to collapse when settling on a final typology rely on a tacit sense of what differences matter, and that sense is grounded in social position and experience. Who the knower is shapes what is identified as a meaningful distinction. Claims to neutral analysis can obscure the gendered standpoints built into apparently technical decisions. In the name of parsimony, we risk erasing precisely those distinctions that matter most for understanding how work is organised and for whom.
The politics of the pooled sample
Taken together, these three examples share a common characteristic of quantitative analyses of working time. In each, a methodological choice reflects a standpoint that passes as neutral: pooled data is often assumed to be the natural and neutral starting point, comparative analysis often leaves men’s work time as the implicit standard and positions women’s as deviations, and gender is treated as addressed once it has been included as a variable.
My response was to analyse women workers only. In many ways, this was a straightforward methodological decision: to define a population and examine how work is organised for that group. Subsetting a nationally representative survey to women does not undermine validity or generalisability: it produces a nationally representative sample of women and defines the population to which the analysis speaks.
However, I was warned that departing from convention would mean repeatedly having to answer the questions ‘why just women?’, ‘does this not narrow the scope of the research?’, ‘does this not paint a partial picture of the organisation of work time?’, and ‘isn’t it a bit too feminist?’. Tellingly, time use research on unpaid work routinely examines mothers and fathers separately, yet the parallel move in the study of paid working time still must answer for itself. After all, a ‘worker’ is assumed to refer to a rational, neutral economic actor. However, restricting analysis to women – or to men – within a nationally representative survey is not a partial or lesser form of enquiry. It is often exactly what is needed to understand how work is organised for that group.
Time Use Survey data and more sophisticated methods are necessary but will not deepen our understanding of the organisation of work time if the questions guiding their use continue to take men’s working patterns as the norm. I argue that the problem is not that focusing on women is too political, but that analysing pooled data is treated as if it were not political at all.
Jennifer Whillans is Senior Lecturer in the School of Sociology, Politics and International Studies at the University of Bristol. She is a mixed methods sociologist interested in the temporal organisation of people and practices in daily life, with a particular focus on gendered use and experience of time. She is concerned with inter-practice connections such that participation in any given practice shapes, and is shaped by, participation in a repertoire of other practices.
Image credit: Andrey Foley via Unsplash
