|Home | About | Journals | Submit | Contact Us | Français|
The field of time use research historically has had an undercurrent of promoting social justice. This dimension dates back to early explorations in the field, from Maud Pember-Reeves use of diaries kept by working class women and George Bevans and use of diaries kept by working class men to debunk myths suggesting the poor lead idle lives without long hours of paid work. Pember Reeves (1913), Leeds (1917), Kneeland (1929), Reid (1934), and others demonstrated that women make significant contributions to the economic output of nations while undertaking unpaid domestic work which official national statistics and economic policies ignored. While it has taken decades for these early observations to gain widespread recognition, recent United Nations reports highlight the importance of collecting surveys of people’s daily routines to promote gender equality, both by making women’s full economic contributions visible and to formulate other gender equality promotion policies (United Nations 2005; UNECE 2013; Calderón Magaña 2013).
While gender equality for women has featured prominently in the time use research literature, the field offers potential to support a range of social justice agendas, as a basic principal of time use surveys is that the all behaviours of all peoples matter. As time use surveys collect data on activities that have received little research or policy attention, and as these surveys collect diaries from of groups with limited social status, some activities which - and people who - have been invisible in policy debates become visible. This principal is not complete. Time use survey sampling methodology tends to leave out institutionalised people and populations whose transient accommodation makes them difficult to sample (homeless people and refugees, among others). Some minority populations, including people with minority sexual orientations and gender identities whom we consider here, have been sampled, but not specifically identified, making analysis of their time use challenging.
Lesbian, Gay, Bisexual, Transgender and Intersex (LGBTI) people recently have made significant legal breakthroughs in some countries, though in others, anti-homosexuality laws and attitudes have become more severe. In those countries where the legal status of LGBTI people has improved, change remains contested and controversial. Few social surveys and official statistics identify LGBTI people as a separate community of research and policy interest. Such large-scale social surveys that are available for LGBTI research mostly concentrate on same-sex couples, and more of these surveys have been collected in the USA than elsewhere (Fisher and Suen 2014).
Visibility in official statistics matters. Public policies cannot cater for unknown needs. Evaluation of the success or failure of programmes requires reliable data on changes in those communities that policies aim to assist. Appearing in routine population statistics confirms regularisation of the legal and social standing of minority groups. Just as recognising the value of the unpaid domestic work of women has played a role in improving the status of women, presence in official population figures will have a role in improving the quality of life of LGBTI people. Carpenter and Gates (2008) reflect a growing number of voices who “strongly urge researchers to more routinely include direct measures of sexual orientation identification on surveys”, not only by collecting whole household age/sex matrixes from large samples, but also asking more specific details about partners and partnership history.
Tracking changes in daily behaviour over time for cohorts of LGBTI people and their majority sexuality contemporaries can reveal the extent to which each of these groups uses the same social spaces at the same (or different) times, the range of regular activities different groups undertake, and whether some groups make compromises (taking longer routes to reach the same destinations or more complicated sequences of behaviours to achieve the same outcomes) to manage the same range of daily experiences. Degrees of different use of spaces, performance of activities and arrangements of days can reflect the degree of social integration (or lack of integration) of any minority population.
Time diary surveys in particular offer the additional possibility of informing the way minority communities which have been ostracised alter their daily routines as they gain social acceptance. As even in their most quantitative and reduced form, time diaries collect narratives, the narrative component of time use surveys offers elements of resonance with the qualitative sexuality studies. On-going experiments with GPS and related devices tracking the location of diarists will enable future time use research to consider the more precise location of activities (in height and well as longitude and latitude), which may prove more useful if measuring differences in uses of social spaces.
Some time use surveys offer the potential to explore the daily activities of same-sex couples. The United States Census and Current Population Surveys collect data on couples in house-holds, including collecting the age and sex of people who identify themselves as being married or living together as a couple. The American Time Use Survey samples a subset of the CPS. Similar possibilities arise in surveys following the Harmonised European Time Use Surveys guidelines, and collect detailed matrixes of household members mapping relationships between members.
As yet, the capacity to identify same-sex couples is incomplete. No currently released national sample time use survey explicitly asks participants about their sexuality (though this will change in 2015). As this research is in the early phase, we adopt a basic definition - people reported to be of the same sex and in a couple on household grid information. We find 415 such people completed diaries (Fisher and Suen 2014) included in the Multinational Time Use Study archive (Fisher and Gershuny 2013 summarise this dataset). We accept that this crude approach may capture some data errors where the sex of one person is recorded in error. Only eight of the over 60 surveys in the MTUS appear to include any same-sex couples, and of these, only three have sufficient numbers of couples for independent analysis – but prospects for analysis do exist for Spain (HETUS surveys and the USA (ATUS). We explore the Spanish surveys in separate research (where we compare the time use of both partners for the same days). In this paper, we concentrate on people living same-sex couples in the USA, where only one of the people in these couples completed one 24-hour diary.
Table 1 displays basic descriptive statistics comparing same-sex and mixed-sex couples in the American Time Use Survey (MTUS version for 2003-2010). People in same-sex couples are slightly younger, a lower proportion of them live in rural areas, and generally they have greater social and monetary capital resources. This most basic comparison suggests sampling bias which we cannot wholly eliminate in modelling. As we have no reliable statistics for the total same-sex couple population, precisely disentangling this potential bias is difficult.
Given the long history of limiting access to adoption and fertility treatment and the very recent legal recognition of same-sex couples in many US states, the lower percentage of couples living with young children is not surprising. Qualitative research comparing same-sex and mixed-sex couples suggests that the absence of established cultural narratives defining roles for domestic arrangements means same-sex couples enjoy more freedom to experiment and invent roles their roles (Shipman and Smart 2006; Smart 2008). The possibility that more people in LGBTI couples work full time may reflect a more widespread interest in domestic equality (some literature on this topic summarised in Fisher and Suen 2014). Other stark differences in the basic distributions are more difficult to explain, though higher levels of education in same-sex compared to mixed-sex couples has been observed before (Shipman and Smart 2006; Smart 2008). It may well be that Lesbians and Gay men with more social standing (reflected by the employment status, income and education) may feel more able to openly acknowledge their sexuality and to choose to form partnerships in keeping with their identities.
The way LGBTI people structure daily routines is a new area in the time use research field. In a small-scale survey of parents, Chan et. al. (1998) observed no difference in the time investments Lesbian and straight parents devote to raising children. While conducting in-depth interviews with ageing gay men in the UK, Suen (2012) recorded many instances where these men recalled needing to use caution with the timing certain activities as well as taking care in choosing the place of some common-place activities, like eating out, in order to avoid trouble.
The MTUS harmonises time use surveys post-collection. This process involves translating original activity codes into a set of 69 harmonised time use activities (Fisher and Gershuny 2013). As a point of initial exploration, we selected those ATUS (MTUS version) diarists who completed good quality diaries and who live in a couple, and ran simple 1-way Anova tests of time in all 69 of these activities comparing the mean daily minutes spent in each activity by same-sex and mixed-sex couples; then by same-sex couples with and without children, and mixed-sex couples with and without children (full tables in Fisher and Suen 2014). Of these 69 activities, only 21 showed significant to marginally insignificant variation in total minutes spent in each activity per day across the two then the four couple groups (Fisher and Suen 2014), though as we have very small samples across pooled years of the American Time Use Survey of same-sex couples, these numbers are not necessarily meaningful. We collapsed these 21 activities into 15 categories for further analysis.
The MTUS offers blunt location information as such detail is not readily harmonised. We tested three additional basic concepts regarding location and timing in the same way as we initially examined the 69 activity categories – total minutes of leisure time in the day spent with the spouse or partner; total minutes per day away from home, and total minutes per day away from home after 18:00. In simple one-way Anova tests, the four groups of couples appear to vary, but again as the numbers of same-sex couples are small, these variations are not necessarily meaningful.
We then followed up the total minutes per day spent in these 18 groups of activities which appeared to show some differences using a simple OLS model2. We opt for this basic model, in part as this paper offers an initial overview of what might be possible in this area, and in part as the numbers of same-sex couples is too small to permit many more sophisticated approaches. While there are some diarists in couples who have no time recorded in one or more of these activities, as people do not undertake every potential activity every day, these 0 observations reflect real behaviours over the 24-hour diary observation windows. More details and results appear in Fisher and Suen 2014.
Once some consideration is made for basic person and household demographics, same-sex couples appear to undertake only four of these activities differently from mixed-sex couples. Same-sex couples spent roughly four extra minutes per day walking dogs; 10 more minutes per day using the internet as well as going out to cinemas, theatres or concerts; and half an hour additional time visiting and in conversation with others. The extra internet and cultural performance time may reflect what appears to be a sample bias, as those same-sex couples with more income, education and higher status jobs may well be over-represented. We suspect that the higher social time same-sex couples enjoy, both with pets and with other people, may prove note-worthy in follow-up research.
The non-significant results, however, also have meaning. We replicate the finding of Chan et. al. (1998) that same-sex couples make the same time investments in their children. As time with children, time contributing to wider social good through organisational and voluntary activities, and time in religious activities reflect some of the contested ground in policy debates over the legal rights and social status of people with minority sexualities, finding no difference between same-sex and mixed-sex couples behaviour in such activities reinforces arguments that protecting the civil rights of minorities poses no threat to the majority population.
Such contemporary concerns as work-life balance or the impact of behaviour on the environment matter for people of minority as well as majority sexualities. Minority groups can face particular circumstances requiring policy attention, and while this is not always the case, policy research should consider the prospects for such differences. As visibility in social statistics affects the representation of minority social groups in policies promoting fairness of opportunities and access to services, time use surveys have particular relevance for collecting some of this baseline social data.
The huge gap in knowledge relates to minority sexuality people who are not in couples. This case has a parallel in the wider time use research literature, as comparatively few articles consider the time use of single people separately, and many which do are relatively recent.
The 2015 Canadian General Social Survey, which will include a 24-hour mixed-method interview time diary, also will ask all participants a basic question about their sexuality (Fisher and Suen 2014). This question will add clarity (or perhaps open new research investigations) into the number of same-sex couples and the suitability of using household matrices to find such couples. More intriguingly, this question will identify some LBTBI people who are not in couples, albeit using categories covering a limited range or orientations. Though not all people will feel empowered or inclined to answer this question, leaving some concern with sampling bias, this question nevertheless represents an advance.
Having a larger sample of non-straight diarists holds out the prospect for more detailed consideration of the timing of activities and structuring of days. The daily routines of LGBTI people merit further research. We hope more future surveys may build on the 2015 Canadian example.
2controlling for sex, age, age2, citizenship, whether the diarist undertook post-secondary education; living in a household in the highest 25% income band, holding employment in a managerial or professional job, working full-time or not, living with a child aged <13 in the household, renting accommodation, living in rural area or not, and (appears to) live in a same-sex couple.