Time series survey data analysis - beginner questions

[Christopher Tracey]

I am working on a data set with 70 variables and 440 observations of different types, categorical (ordinal and nominal) and numerical. The data set is from two knowledge, attitude, and practice surveys conducted in three villages (one control, one with a single intervention, one with two interventions), the first conducted in 2016 and the other in 2018. The 2018 survey contained a few repeat questions, but also collected variables that were not included in the first survey (demographic variables and others). The data was in the long format with the first survey respondents on top of the second survey respondents with the surveys identified by a time variable that is 0 if conducted in 2016 and 1 if conducted in 2018. The same participants were interviewed both times and a case ID was generated that identifies the participants as being the same in both surveys.

The question is, how do I approach analysis of this data? Do I reshape the data to wide with i(caseid) and j(time)? Is it possible to compare the same individuals through time while comparing the villages to each other?

Any guidance would be appreciated.

[Nick Cox]

Cross-posted at https://www.reddit.com/r/stata/comme...ysis_beginner/ Please note our policy on cross-posting, which is that you tell us about it. https://www.statalist.org/forums/help#crossposting

That question has attracted no answers in 5 days, and I have to guess that's largely because "how do I approach analysis of this data?" is too broad a question. What do you want to know? What are your goals in analysis? Comparison of villages and individuals too is fair enough, but that means that the whole of [XT] and much else could be relevant.

My only specific comment is that reshape wide is usually the wrong way to go in Stata. From what I can gather, your data layout is fine.

[Christopher Tracey]

Thank you and my apologies.

I am a novice to Stata and data analysis in general, I guess what I was trying to get at was how to format the data within Stata in order to run [XT] analyses, but with your response here and searching the forums I have found more than enough information to format the data and explore [XT] on my own.

I will be sure to come with a more specific question and applicable code next time!