Time-use diary analysis in Stata: averaging unpaid care work by gender

[Shamsudini Amidu]

Dear Family,

I collected time-use diary data covering a 24-hour period for couples (male and female) and recategorized all reported activities into four broad groups:

• Productive
• Reproductive (unpaid care work)
• Leisure
• Other

The main objective of the survey is to estimate, on average, how much time men and women spend on unpaid care (reproductive) activities.

Each respondent recorded the number of hours spent on different activities within a 24-hour period.

During data exploration, I observed that:

• In some households, females reported more activity episodes than males but with different amounts of time spent.
• In other households, males and females reported the same number of activity categories, but with different amounts of time spent.
• Each of the activities for bother gender summed up to 24 hours.

This raised a methodological concern for me. I initially thought that, since I am comparing time spent on reproductive activities between males and females, I might need to restrict the analysis to couples who reported the same number of activity categories, in order to avoid bias.

However, in my final analysis, I did not impose this restriction. Instead, I calculated the average hours spent on reproductive activities by gender, regardless of whether the total number of reported activities differed between partners.
I would like confirmation on whether this analytical approach is methodologically sound and if there is a best way to go about this.


My do-file and example dataset is elaborated below for your review.


* Example generated by -dataex-. For more info, type help dataex
clear
input str36 _uuid byte Activity_Cat double duration_hr str6 gender
"003b21e4-d5db-494c-b4ee-5a1f513135cb" 3 12.583333387970924 "Female"
"003b21e4-d5db-494c-b4ee-5a1f513135cb" 1 5.583333343267441 "Female"
"003b21e4-d5db-494c-b4ee-5a1f513135cb" 2 4.583333343267441 "Female"
"003b21e4-d5db-494c-b4ee-5a1f513135cb" 4 1.25 "Female"
"003b21e4-d5db-494c-b4ee-5a1f513135cb" 3 16.166666768491268 "Male"
"003b21e4-d5db-494c-b4ee-5a1f513135cb" 1 6.500000014901161 "Male"
"003b21e4-d5db-494c-b4ee-5a1f513135cb" 4 1.1666666865348816 "Male"
"003b21e4-d5db-494c-b4ee-5a1f513135cb" 2 .1666666716337204 "Male"
"00a32c70-16e8-4930-8636-22f076bf3751" 2 5 "Female"
"00a32c70-16e8-4930-8636-22f076bf3751" 4 1.5 "Female"
"00a32c70-16e8-4930-8636-22f076bf3751" 1 6 "Female"
"00a32c70-16e8-4930-8636-22f076bf3751" 3 11.5 "Female"
"00a32c70-16e8-4930-8636-22f076bf3751" 2 1.5 "Male"
"00a32c70-16e8-4930-8636-22f076bf3751" 4 3 "Male"
"00a32c70-16e8-4930-8636-22f076bf3751" 1 9 "Male"
"00a32c70-16e8-4930-8636-22f076bf3751" 3 10.5 "Male"
"011804cd-99ad-4089-81ca-279ba2e8d770" 1 4 "Female"
"011804cd-99ad-4089-81ca-279ba2e8d770" 2 5.833333373069763 "Female"
"011804cd-99ad-4089-81ca-279ba2e8d770" 3 14.166666686534882 "Female"
"011804cd-99ad-4089-81ca-279ba2e8d770" 2 2 "Male"
"011804cd-99ad-4089-81ca-279ba2e8d770" 4 2 "Male"
"011804cd-99ad-4089-81ca-279ba2e8d770" 1 6.5 "Male"
"011804cd-99ad-4089-81ca-279ba2e8d770" 3 13.5 "Male"
"021e0c02-e329-481c-82c6-b5c64b54928d" 2 5 "Female"
"021e0c02-e329-481c-82c6-b5c64b54928d" 1 3 "Female"
"021e0c02-e329-481c-82c6-b5c64b54928d" 3 15 "Female"
"021e0c02-e329-481c-82c6-b5c64b54928d" 4 1 "Female"
"021e0c02-e329-481c-82c6-b5c64b54928d" 1 5 "Male"
"021e0c02-e329-481c-82c6-b5c64b54928d" 3 17.5 "Male"
"021e0c02-e329-481c-82c6-b5c64b54928d" 4 1 "Male"
"021e0c02-e329-481c-82c6-b5c64b54928d" 2 .5 "Male"
"0359a1ea-a472-4de0-baf7-835422ba2660" 2 8 "Female"
"0359a1ea-a472-4de0-baf7-835422ba2660" 3 7.333333373069763 "Female"
"0359a1ea-a472-4de0-baf7-835422ba2660" 1 8.666666686534882 "Female"
"0359a1ea-a472-4de0-baf7-835422ba2660" 1 5 "Male"
"0359a1ea-a472-4de0-baf7-835422ba2660" 2 5.333333343267441 "Male"
"0359a1ea-a472-4de0-baf7-835422ba2660" 3 13.666666731238365 "Male"
"0393e43f-4ea1-4477-a28a-294776b9cde1" 1 2.6666666865348816 "Female"
"0393e43f-4ea1-4477-a28a-294776b9cde1" 3 10.500000014901161 "Female"
"0393e43f-4ea1-4477-a28a-294776b9cde1" 4 3.58333333581686 "Female"
"0393e43f-4ea1-4477-a28a-294776b9cde1" 2 7.25 "Female"
"0393e43f-4ea1-4477-a28a-294776b9cde1" 4 4.583333343267441 "Male"
"0393e43f-4ea1-4477-a28a-294776b9cde1" 3 11.250000074505806 "Male"
"0393e43f-4ea1-4477-a28a-294776b9cde1" 1 6 "Male"
"0393e43f-4ea1-4477-a28a-294776b9cde1" 2 2.1666666865348816 "Male"
"03e7e418-752c-4d6e-bd00-a73c4fe77513" 3 11.250000059604645 "Female"
"03e7e418-752c-4d6e-bd00-a73c4fe77513" 1 6 "Female"
"03e7e418-752c-4d6e-bd00-a73c4fe77513" 2 6.75 "Female"
"03e7e418-752c-4d6e-bd00-a73c4fe77513" 1 7.5 "Male"
"03e7e418-752c-4d6e-bd00-a73c4fe77513" 2 1.5 "Male"
"03e7e418-752c-4d6e-bd00-a73c4fe77513" 4 1 "Male"
"03e7e418-752c-4d6e-bd00-a73c4fe77513" 3 14 "Male"
"04b9e68e-bebb-401e-ad83-305a9b33adbe" 2 6 "Female"
"04b9e68e-bebb-401e-ad83-305a9b33adbe" 3 17 "Female"
"04b9e68e-bebb-401e-ad83-305a9b33adbe" 1 1 "Female"
"04b9e68e-bebb-401e-ad83-305a9b33adbe" 3 20 "Male"
"04b9e68e-bebb-401e-ad83-305a9b33adbe" 2 2 "Male"
"04b9e68e-bebb-401e-ad83-305a9b33adbe" 1 2 "Male"
"063bf8d3-1853-4199-8c35-dab38b11376e" 2 2.5 "Female"
"063bf8d3-1853-4199-8c35-dab38b11376e" 1 7 "Female"
"063bf8d3-1853-4199-8c35-dab38b11376e" 3 14.5 "Female"
"063bf8d3-1853-4199-8c35-dab38b11376e" 4 3 "Male"
"063bf8d3-1853-4199-8c35-dab38b11376e" 3 13.5 "Male"
"063bf8d3-1853-4199-8c35-dab38b11376e" 2 2 "Male"
"063bf8d3-1853-4199-8c35-dab38b11376e" 1 5.5 "Male"
"067724c6-7316-4ea5-9faf-649758504750" 3 13 "Female"
"067724c6-7316-4ea5-9faf-649758504750" 1 2 "Female"
"067724c6-7316-4ea5-9faf-649758504750" 2 9 "Female"
"067724c6-7316-4ea5-9faf-649758504750" 1 4 "Male"
"067724c6-7316-4ea5-9faf-649758504750" 4 1 "Male"
"067724c6-7316-4ea5-9faf-649758504750" 2 1 "Male"
"067724c6-7316-4ea5-9faf-649758504750" 3 18 "Male"
"07053c83-a619-4c0d-9e40-609e5cfb2682" 2 5.75 "Female"
"07053c83-a619-4c0d-9e40-609e5cfb2682" 1 5.666666686534882 "Female"
"07053c83-a619-4c0d-9e40-609e5cfb2682" 3 12.58333334326744 "Female"
"07053c83-a619-4c0d-9e40-609e5cfb2682" 3 16.08333333581686 "Male"
"07053c83-a619-4c0d-9e40-609e5cfb2682" 1 7.916666686534882 "Male"
"08a5f2d1-10bb-4f82-8bbe-9cc711ceca24" 2 3.6666666865348816 "Female"
"08a5f2d1-10bb-4f82-8bbe-9cc711ceca24" 1 9 "Female"
"08a5f2d1-10bb-4f82-8bbe-9cc711ceca24" 3 11.333333373069763 "Female"
"08a5f2d1-10bb-4f82-8bbe-9cc711ceca24" 2 2.6666666865348816 "Male"
"08a5f2d1-10bb-4f82-8bbe-9cc711ceca24" 1 7 "Male"
"08a5f2d1-10bb-4f82-8bbe-9cc711ceca24" 3 14.333333358168602 "Male"
"08f31c66-15b6-4b07-ae5b-c042b6762c0e" 3 11.500000089406967 "Female"
"08f31c66-15b6-4b07-ae5b-c042b6762c0e" 1 6.333333373069763 "Female"
"08f31c66-15b6-4b07-ae5b-c042b6762c0e" 2 6.166666746139526 "Female"
"08f31c66-15b6-4b07-ae5b-c042b6762c0e" 3 13.166666761040688 "Male"
"08f31c66-15b6-4b07-ae5b-c042b6762c0e" 1 7.5 "Male"
"08f31c66-15b6-4b07-ae5b-c042b6762c0e" 2 3.333333343267441 "Male"
"09dd50fc-8d34-4703-ac04-8832d8a5bc39" 2 6 "Female"
"09dd50fc-8d34-4703-ac04-8832d8a5bc39" 3 12 "Female"
"09dd50fc-8d34-4703-ac04-8832d8a5bc39" 1 4 "Female"
"09dd50fc-8d34-4703-ac04-8832d8a5bc39" 4 2 "Female"
"09dd50fc-8d34-4703-ac04-8832d8a5bc39" 3 16 "Male"
"09dd50fc-8d34-4703-ac04-8832d8a5bc39" 1 8 "Male"
"0c81144a-ad84-497a-a194-f816ef339503" 4 1.3333333730697632 "Female"
"0c81144a-ad84-497a-a194-f816ef339503" 3 12.166666746139526 "Female"
"0c81144a-ad84-497a-a194-f816ef339503" 2 4.833333432674408 "Female"
"0c81144a-ad84-497a-a194-f816ef339503" 1 5.666666686534882 "Female"
"0c81144a-ad84-497a-a194-f816ef339503" 1 7.416666656732559 "Male"
end
label values Activity_Cat Activity_Cat
label def Activity_Cat 1 "Productive", modify
label def Activity_Cat 2 "Reproductive", modify
label def Activity_Cat 3 "Leisure", modify
label def Activity_Cat 4 "Other", modify


///**************************STATA COde
***Count the number of unique genders per UID WITHOUT dropping activities
bysort _uuid gender: gen tag_gender = 1 if _n==1 // tag first row per gender per UID
bysort _uuid: egen gender_count = total(tag_gender) // sum tags per UID → gives 2 if both male & female exist

*Keep only UIDs with both male and female
keep if gender_count == 2

drop tag_gender gender_count

***Compute average duration per activity by gender
collapse (mean) duration_hr, by(gender Activity_Cat)

**Show results
list, noobs

I have also attached the dofile and the entire dataset for more details.
Thank you.
Amidu Shamsudini.

[John Mullahy]

If I am understanding correctly, here is one opinion.

You write

I might need to restrict the analysis to couples who reported the same number of activity categories, in order to avoid bias.

Is it appropriate to assume that someone who does not report a particular activity category has spent zero hours performing that activity?

If so then I think coding such activity durations as zero and including the full menu of activity categories for each subject would be reasonable. All subjects would then have their time allocations sum to 24 hours across the activity categories, with some of the summands being zero for some subjects.

[Shamsudini Amidu]

Dear John,

My apologies for the late response. I was away on fieldwork in a rural area with very poor network connectivity.

Thank you very much for your input. It is well noted and appreciated.
Shamsudini