Collecting time use data

1
Collecting time use data

• Social and Housing Statistics Section
• United Nations Statistics Division
• Time Use Statistics workshop for Arabic speaking
  countries, Amman,25-28 April 2011

2
Collecting time use data

• Goal of my presentation
• Overview of the different survey frameworks.
  Practices which prevail amid member countries.
• Survey framework?
• Type of survey
• Mode of data collection
• Type of survey instrument

3
Collecting time use data

• Goal of my presentation
• Overview of the different survey frameworks.
  Practices which prevail amid member countries.
• What do I mean by survey framework?
• Type of survey
• Mode of data collection
• Type of survey instrument

4
Mode of data collection

• 2 main modes
• Recall interview
• Self-reporting

5
Mode of data collection

• 2 main modes
• Recall interview
• Self-reporting

6
Recall interview

• Interviewing about time use during a previous
  period
• Face-to face or over the phone

7
Recall interview

• General limitations
• Memory recall errors (Robinson, 1985)
• Normative editing (Robinson, 1985)
• Underreporting of quick activities

8
Face-to-face recall interviews

• Advantages
• Most common technique when low literacy rates
• Higher response rate
• Limitations
• More costly

9
Computer-assisted telephone interviews

• Advantages
• High response rate compared to self-completed
  questionnaires
• Lower costs than face-to-face
• Controls from CATI software
• Limitations
• Limits of random digit dialing
• Low response rate when more than one
  questionnaire or more than one respondent in the
  household
• Variations in the classification
• Development of cell phones

10
Mode of data collection

• 2 main modes
• Recall interview
• Self-reporting

11
Self-reporting by respondents

• Respondents are asked to complete a
  leave-behind diary on an assigned day (or days)

12
Leave-behind diary

• Advantages
• Suitable for high literacy rates
• May be considered less intrusive
• Higher quality than recall data ( 10)
• Limitations
• Not feasible when low literacy rates
• Costs greater than recall because 2 interviews
  are needed
• Data quality may be affected if the respondents
  are asked to fill in more than one day (Harvey
  and Taylor, 2000)

13
Type of survey instrument

• 2 main instruments
• 24-hour diary
• Stylized questionnaire

14
Type of survey instrument

• 2 main instruments
• 24-hour diary
• Stylized questionnaire

15
24-hour diary

• 2 main issues
• Full versus light diaries
• Recording of time

16
Full diary

• Writing verbatim descriptions of activities that
  are coded later on to an activity classification

17
Light diary

• Restricts activity descriptions to a limited
  categorization of pre-coded activities

18
Full versus Light diary

• The choice depends on
• Analytical objectives
• Cost
• Literacy concerns
• Misclassification of the activities

19
24-hour diary

• 2 main issues
• Full versus light diaries
• Recording of time

20
Recording of time

• 2 main choices
• Fixed intervals non-overlapping segments of
  uniform length
• Open recording the respondent reports start and
  finish times of each activity

21
Recording of time

• Issues to take into account
• The open method yields large variations in the
  quality of data and complications for editing and
  processing (Fleming, 1999)
• But fixed intervals not suited to short duration
  activities
• Length of fixed intervals

22
Type of survey instrument

• 2 main instruments
• 24-hour diary
• Stylized questionnaire

23
Stylized questionnaires

• The respondents need to recall the amount of time
  spent on activities.
• May target specific activities or be designed to
  be as exhaustive as possible as a 24-hour diary

24
Stylized questionnaire

• Advantages
• Less expensive
• Preferable for a specific and short time period
• Used to supplement 24-hour diaries

25
Stylized questionnaire

• Limitation
• High degree of error (Hofferth, 2000)

26
Conclusion

• Analytical purposeguiding principle
• Advantages and limitations are essentially known
• How multiple options in a single study?
  Combination of methods?