Unlike the household questionnaire, the activity diaries required an intensive clerical process.
Processing of the diaries involved sorting the reported activities into episodes, editing where necessary and recording episodes into a data entry system. An episode can contain the following elements:
- start and finish time;
- primary activity;
- secondary activity;
- person or group for whom the activity is done;
- location, both physical and spatial;
- mode of transport for travel items;
- technology/communication code where relevant;
- who the respondent was with; and
- the age and health details of any household people present.
A change in any of these elements created a new episode.
Up to two simultaneous activities were captured (e.g. the main activity and a secondary activity). The main coding principle, confirmed in user consultation, was to remain as close to the respondent's reporting as possible. In practice this meant that:
- the respondent's decision about whether an activity was main or secondary was accepted, except in the circumstances specified below;
- a respondent's combination of activities was accepted if seen as possible; and
- omitted activities were only imputed under the conditions listed below.
Certain activities were seen as overriding, or natural main activities because they created the environment in which other things happened. Travelling, for instance, limits the range of things that can be done, and was kept as a main activity. Socialising or entertainment that involved going to a venue was treated as an overriding activity. Paid work and education were coded globally, that is, there was no breakdown of different kinds of activities within these areas. Activities that occurred at work or school, but were not related to work or school, however, were coded as main activities. Sleeping was normally a main activity but could be secondary to travel.
As one of the main purposes of the survey is to measure the time spent on unpaid work certain activities were seen as 'priority activities'. That is, in most cases, even if the activities were recorded as secondary activities in the diary they were coded as a primary activity so that they could be captured as part of unpaid work. This is because most analysis only uses the primary activities when determining a measure of unpaid work. The priority activities were any activities that involved domestic work, child care (except passive child minding), caring for adults, doing favours or voluntary work. Activities that occurred simultaneously with these activities were recorded as secondary unless they were one of the overriding activities listed above.
Certain activities, particularly the passive care of children or adults requiring monitoring, were not consistently reported. Because the diaries and questionnaires contain information that suggests these activities might be happening even where they are not reported, and because the caring activities are of interest to many users, consideration was given to imputing episodes of care. This did not happen, in the interests of accuracy and of remaining faithful to the respondent's reporting. However, the supporting information was captured in the 'who with' codes for household members, which can provide measures of time spent with children under 12, people who are sick or have disabilities, and older people. Caution should be applied to treating these items as proxies for caring items - only a small proportion of people with disabilities or people aged 60 and over are in need of care. However, combinations of some of these with other items might suggest constraints on the way in which time can be used.
Many activities are likely to have been omitted from the diaries. Constant background activities, for instance, such as passive child care, and others such as smoking and drinking could be under-reported. Whether or when this happens is unknown. There are cases however where reported activities imply or suggest activities that are not reported, and provide the time at which they are likely to have happened. The most common of these is travel. For example, a person is at home and then is at work, or shopping, without an intervening travel episode. The methods for dealing with this were, in order:
- to check for comparable travel episodes elsewhere in the diary;
- to check for similar travel in another diary in the household;
- to use any available knowledge of the areas concerned; or
- to impute travel time, usually ten minutes; this would normally be taken from the beginning of the new activity, unless this began at a standard starting time, for work or school for instance.
The other common case relates to eating. Respondents sometimes wrote 'made a cup of tea' or 'got lunch' without specifically mentioning drinking or eating. If there was little doubt that eating was implied, the activity was imputed, and the time split between preparing and eating. If it was possible that the preparation was for someone else, then it was coded as reported.
Where nothing was reported for a stretch of time, the main activity was coded as 'undescribed', unless it was clear that the previous activity was continuing. Diaries with episodes of undescribed time were examined, and the undescribed time reconstructed where possible by reference to the other diary day or other diaries from the household.
The final average measure of time with an undescribed main activity episode is 7 minutes or 0.49% of the day.
People present during the activity
Information collected in column 5 of the diary (Who was with you at home or with you away from home?) describes:
- who was with the person when they were at home. This referred to all present in the house and grounds, whether belonging to the household or not; and
- those for whom the person is responsible, as well as those involved in the same activity when away from home (e.g. at a picnic, the person helping the respondent prepare the food, the others conversing with them and the associated children nearby).
Information collected from this column referred to the other people present, it did not include the respondent. Family refers to all relatives. There is a distinction between family living within the household and family from another household. The relevant household and family type data items described the family structure within the household. Age status and health status were also coded for household members only. The age group of the household members was identified from the household form. Health status was obtained from information in the diary as well as information from the questionnaire about whether the household members had a disability. Disability was recorded only where the other person present had a long-term restricting condition or impairment reported on the questionnaire. Short-term sickness or injury was only reported when the diary information indicated this.