'Time use during the UK national lockdowns' (AsPredicted #61386)
Author(s) This pre-registration is currently anonymous to enable blind peer-review. It has 2 authors.
Pre-registered on 2021/03/19 - 04:43 PM (PT)
1) Have any data been collected for this study already? It's complicated. We have already collected some data but explain in Question 8 why readers may consider this a valid pre-registration nevertheless.
2) What's the main question being asked or hypothesis being tested in this study? a) How did the amount and quality of time spent in broad activity categories change as a result of the COVID-19-induced national lockdowns in the UK?
b) How do these changes differ across socio-demographic groups?
3) Describe the key dependent variable(s) specifying how they will be measured. 1. Amount of time spent in broad activity categories: respondents will fill in 24-hour time-use diaries, detailing the duration and sequence of activities (e.g. sleeping, eating, paid employment) they engaged in during a particular day. Across the two survey waves, respondents will fill in up to 6 time-use diaries, corresponding to the most recent non-working day and working day during each of the three time periods of interest (February 2020, May 2020, and February/March 2021). We use the same broad activity categories as the UK Time Use Survey, which are almost identical to the categories used in other national time use surveys. We calculate the amount of time spent on each activity category by adding up the reported duration of time spent in that activity across all episodes in a diary day.
2. Quality of time use: We examine three variables that are likely to affect quality of time use - multitasking (whether a secondary activity was done concurrently with the main activity), co-presence (who the respondent reported doing the main activity with), and fragmentation of time use (number of times the respondent did the activity in a given day, divided by the total number of activities in that day).
4) How many and which conditions will participants be assigned to? N/A
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. 1. T-tests for differences in means in the amount and quality of time use across broad socio-demographic groups (age, income, gender, education, household composition).
2. Multivariate regression where the dependent variable is time use in a given activity (expressed as a percentage or proportion of total time in a day), and the main independent variables of interest are
a. Socio-demographic variables - age (continuous variable), gender (dummy=1 if female), education (dummy=1 if BA degree or higher), ethnicity (dummy=1 if white).
b. Household composition - living alone, living with at least one child under 5, living with at least one child aged 6-11 (all dummy variables).
c. Employment situation - income (categorical variable), currently work from home (dummy), flexible working hours or location pre-pandemic (dummy=1 if respondent selected 'agree' or 'strongly agree' to statements of having freedom to choose working hours or location), contract type (categorical or dummy=1 if zero hours), Hours worked (categorical, continuous, or dummy=1 if full-time).
Where appropriate, we will use Bonferroni corrections to account for multiple comparisons.
3. From the distribution of time spent in a given activity, we will calculate measures of inequality (the Gini coefficient and/or ratio-based measures) to complement the standard summary statistics.
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. For each time period (before lockdown, first national lockdown, third national lockdown), we will include all participants that filled in at least one time-use diary that contains at least 3 entries (a 'complete' time use diary) for that period. We will use all available data provided by these participants and not exclude any observations.
7) How many observations will be collected or what will determine sample size? No need to justify decision, but be precise about exactly how the number will be determined. Sample size was determined by the amount of funding we received to conduct this study. We used the online survey platform Prolific to recruit 1000 working-age adults living in the UK at the time of the first national lockdown. Deviations from this number are entirely due to Prolific's software and are beyond our control. We then re-contacted these participants in February and March 2021 to complete the second wave of our survey.
Our sample is demographically diverse and designed (by Prolific) to resemble the composition of the UK adult population in terms of age, gender, and ethnicity. All recruitment and sampling processes were done by Prolific and were beyond our control.
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) Regarding data collection: The survey design was approved by the Oxford Department of Economics Ethics Review Board on 5 May 2020 and has not been altered since that date. Our survey contains two waves, timed according to the national lockdowns in the UK. We have already finished data collection for the first wave (first national lockdown from March-July 2020) but are still collecting data for the second wave (third national lockdown from January 2021-April 2021). We have not conducted any of the analyses specified in (5) on any of the data yet.
Depending on the analyses in 5), we also plan to:
1) Run multivariate regressions with additional control variables, including
a. Occupation (categorical variable) and employment history (categorical variable or dummies for each employment status)
b. Quality of work environment at home - personal workspace at home (dummy), shared workspace (dummy), frequency of interruptions by family members and social media (categorical or dummy), feelings of being able to work better at home (Likert scale),
c. Other activity- or episode-specific characteristics - used a device during the activity (dummy), time of day (categorical), where the activity was conducted (categorical or dummy), episode-specific enjoyment (Likert scale of 1-7).
d. Preference rankings over all activities listed.
2) Conduct the regression analysis in (5) with additional interaction terms between the main independent variables and/or the variables listed in this section.
3) Calculate a quality-weighted measure of time use to examine the overall effect of changes in the extensive vs intensive margin.
