need help with independent survey data

[Shazia ciel]

Hi,

I am working on a research paper that uses data of three survey waves. It is a cross country survey that is published every three years and currently have three waves, year 2011,2014 and 2017.The survey is independent and the participants are individuals.The data set has weight given and in the methodology of the survey they mentioned that this weight take care of the stratification used.Since I am new to this kind of survey analysis, I am wondering how should I set my svy command.I want to uses a probit model since my dependent variable is binary. In this case what command should I use to declare that its a survey data set? also, do I need to use country and time fixed effect since the data are cross-country?

Please advice.

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a helpful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

I don't do survey analysis, but you need to start by reading the Survey Data manual that is one of the manuals in the pdf documentation that comes with Stata. The introduction and early pages handle these issues. It is extremely good practice to search the Stata documentation (there is a subject index available under Index in the pdf documentation and there are manuals for different kinds of analysis) and google the issue before posting a question.

Whether you need fixed effects and the level at which you set fixed effects if you need them, are substantive questions which have no general answer. Be aware that such effects do really change what you're estimating. I'd look at practice in your area.

[Shazia ciel]

Thank you, Phil.