Hi
I downloaded data including some time use variables (especially ACT_carehh and similar variables I created). I read in a few places that values should be the number of minuets respondents spent on the specified activities, yet they seem to be coded differently.
The values range between 2-9900, and I didn’t manage to find any consistency with actual numner of minuets I summed (manuualy for a number of respondents) from the respondents’ avtivities description.
So- is the data dlawed in some way or are ACT_… variables coded?
many thanks
Hillel
I am not able to replicate the frequency distribution you are describing. I will provide more information below, but please email ipums@umn.edu directly for more targeted troubleshooting assistance.
Time use variables (like ACT_CAREHH) report the number of minutes a person spent in the activities included in that category over the course of the day; values should range between 0 and 1440 (e.g., persons may spend anywhere between no time and the entire 24 hour time diary in a given activity). I submitted an ATUS extract of all available years of data that includes respondents, non-respondents, and household members; I am seeing values that range between 0 and 1230. There are a large number of missing values; these belong to persons other than the ATUS respondent (e.g., non-respondents and household members of the respondent–neither of these categories will have reported time use information). Similarly, in a handful of custom time use variables I have created in the past, I am only seeing values ranging between 0 and 1440.
If you are working with a hierarchical data extract, there will be two types of records in your file: person records (these will include demographic information, time use variables summarized to the person for each diary day, other person-level indicators) and activity records (these will include activity-specific information like the location and duration of the activity). In a hierarchical file, the person-level variables will not be present on the activity records. However, I would still expect the time use variables to range between 0 and 1440 (this is how they appear in my extracts).
You can view more information about which detailed activity codes are included in a time use variable by clicking on the variable name, and then any hyperlinks for related activities in the subsequent pop-up window (see ACT_CAREHH below for reference).
Main ACT_CAREHH information:
Additional detail about specific activities included in the broader “Caring for and Helping Household Members” activities category:
Hi Kari
Thank you very much for the detailed reply.
Unfortunately, and after rechecking, my sample includes peculiar values (for preson record observations), I will try to attach a screenshot for example.
I will contact the trubleshoot team via mail.
Regards
Hillel
1 Like
Thank you for the direct email with additional details. I was mistaken when I stated that I wasn’t able to replicate the issue in a hierarchical extract–my apologies. The issue you are describing is an error in our syntax files; which are currently assigning the incorrect column locations to time use variables in hierarchical extracts. We are working on fixing this error.
To thank you for helping us identify and fix this error, we would like to send you an IPUMS mug. Please contact ipums@umn.edu with a mailing address where we can send this small token of our appreciation.
