My question is about activities in the category of “education” in the ATUS data from 2015-2019: How does the presence of others get coded in education activities?
For more detail: I am looking at a sub-sample of teenagers (15-17 years old) in 2015-2019 who report some time in education, including the code “Taking class for degree, certification, or license” (activity=60101). In the variable “relatew” to identify who else is present in this activity, some respondents report having another person with them during this time - e.g., “Other non-household adults 18 or older” and “Friends.” However, other respondents report “Blank.”
Is there any logic to when respondents can report others present during education activities such as class time, or when they report it as blank? I have not been able to find anything in the ATUS Coding Rules documentation (from 2011) that sheds light on this, nor in any other documentation.
Please let me know if I can clarify!
This is unfortunately a result of messiness in the data. This kind of disconnect occurs with other variables in the ATUS such as travel activities and locations as mentioned in this blog post. Please note that blank responses do not imply the individual performed the activity alone. There is a separate code that’s reported for this case (100) as opposed to when a response is blank (9998). It seems that the presence of blanks for this activity indicates missing data. The 2019 ATUS codebook marks these cases using the variable TRWHONA. There are 35,288 such entries in the 2019 BLS microdata file, which matches the number of blanks in the 2019 IPUMS Time Use sample. The ATUS user guide notes that “ATUS collects information (“who” codes) on the person(s) in the room with or accompanying the respondent during his or her activities, yet several activities are ineligible for a “who” code assignment.” However, education does not fall into one of these ineligible activities. The guide further states that “no imputation is performed on missing “who” codes”, suggesting these are simply “missing” codes.
To address the issue of people reporting being alone while taking classes, I checked the location of the activity to see if these respondents were remote students. However, the majority reported being at school while taking classes. A likely scenario is that some respondents implicitly excluded their classmates and instructors while others included them. That would explain why you’re seeing a mix of answers. I wish I could have given a more definitive response, but this is often the nature with survey data.
Thank you, Ivan, for your in-depth answer! I appreciate this. This is exactly the kind of information we were looking for to be able to contextualize the blanks in our research.