Hello, in this ATUS extracted table from well-being module, why were there multiple different responses in a same caseID as well as the other parameters?
Thanks!
It appears that each line in your data extract is a singular activity reported by a sample respondent. For example, the respondent with CASEID 20100101100019 recorded 12 activities during the day that they were interviewed about. This format allows you to analyze each activity individually. For example, it shows that they reported being happier (SCHAPPY) while doing the third activity as compared to when they were engaged in the fifth activity. This FAQ page contains more information about the difference between rectangular and hierarchical data.
You can change the data format on the extract request page (screenshot below). Click Change next to Structure and select between rectangular person level data without activity information, rectangular activity level data (which you currently have), or hierarchical data which includes both person and activity level data as separate entries.
I was trying to find “well-being” data of each industry / occupation. I wanted to see the well-being (happiness, sadness, meaningfulness, stressfulness, etc) of each industry type / occupation in general, but like the screenshot attached, I always got tons of same responses with a same person and a same occupation, and from then I don’t know what I should do next.
Can you please guide me through finding the dataset? Thank you so much!
Based on your screenshot, you have requested a data file that is rectangularized on activity; each line in the data file will be an activity reported by a respondent. You can use the CASEID variable to identify respondents. In your screenshot, the respondent with CASEID 20100101100019 recorded 12 activities during the day that they were interviewed about. For the first 12 lines on your data, the respondent’s industry and occupation will remain the same because this is all data on the same individual. Information collection on a new respondent begins when the CASEID number changes on line 13.
Well-being variables such as SCHAPPY, SCSTRESS, and MEANING report the respondent’s feelings while performing a specific activity and do not describe their overall well-being. Moreover, these variables only report well-being for three randomly selected activities. To what extent they affect the respondent’s overall well-being is unclear. One avenue you might consider is to compare reported well-being for activities across industries and occupations. For example, you can compare the value of SCHAPPY when it is reported for activities that involve reading across industries and occupations. Please be aware however that estimates using small sample sizes will have larger margins of error. For additional guidance, I recommend taking a look at the FAQ page, this video tutorial on how to create an extract, as well as available training materials, and previous workshop presentations.