TIME-USE OF WOMEN AND MEN IN IRELAND
331
of the interviewer. Additional demographic and satisfaction information was
collected through a self-completion questionnaire attached to the diary.
A total of 585 households participated in the survey, giving a household
participation rate of just under 58 per cent of those targeted. Not everyone
filled in both diaries, and not all diaries which were completed by household
members could be used in the analysis: diaries with more than 15 empty time-
slots (2 hours) were excluded. In total 79 per cent of eligible individuals within
households contributed at least one useable diary.
The sample was re-weighted, controlling for gender, age, household
composition, region, educational attainment and principal economic status, to
represent the national population (see Appendix Table B1).7 All descriptive
tables presented in this report are based on these reweighted data (see
McGinnity et al., 2005 for further details of re-weighting).
2.1 Overall Patterns of Time-Use in Ireland
The survey asked respondents to complete two time-use diaries one for a
weekday and one for a weekend. We generally present results separately for
weekday time-use and weekend time-use as these vary considerably: for
European comparisons we combine these to give an ‘average’ day. As
respondents were permitted to record multiple activities (to reflect the reality
that individuals often carry on more than one activity at a time) the total time
recorded often adds to more than 24 hours. In order to limit the total time to
24 hours, where respondents undertake more than one activity at once, we
impose alternative definitions of which is the ‘main’ activity. For this paper we
impose the following priority order for defining the main activity: 1 childcare
and adult care; 2 employment and study; 3 household work; 4 travel; 5
personal care and eating; 6 leisure and voluntary activity; 7 sleeping and 8
unspecified time-use. If two or more activities are recorded in a time-slot
priority is given to the activity that appears first in the list.8
While prioritising tasks is necessary to reduce very complex data into
something that can be meaningfully analysed, it does have certain
implications for time-use estimates (see Gershuny, 2000 for a discussion). In
7 From Appendix Table B1 we can see from the unweighted data that women and those over 45
years were overrepresented in the sample. Those with highest education Leaving Certificate and
in employment were also over-represented, as were those with no children, married and
respondents from the BMW region. As later analysis will reveal, some of these characteristics
increase the likelihood of being ‘time poor’, some reduce it. However, as the weighting procedure
adjusts the sample to be representative of the population on these key indicators, we expect any
bias remaining after weighting to be negligible.
8 Results using an alternative priority setting - 1. employment and study, 2. travel, 3. personal
care and eating, 4. housework and shopping, 5. caring, 6. leisure and voluntary activity, 7. sleeping
and 8. unspecified time-use - are discussed in McGinnity et al., 2005, Appendix B.