Is it possible to combine datasets from different years from the exact same survey?

Question

I have difficulties with the sample size of my targetgroup. I am doing an analysis based on the dataset of the European Social Survey (ESS). This is a survey that is conducted every year with structural modules and changing modules. My research focuses on the structural modules. I am wondering whether it is possible to combine datasets from different years from the ESS to create a greater sample size. And if this is possible, what the possible disadvantages are?

Answer 1

Is it possible? Sure. You can do it. There's lots of things you can do and, often, in statistics it's not a matter of "right" and "wrong" but "justifiable" vs. "not justifiable".

What are possible disadvantages? Without knowing a lot more detail, it's hard to be specific, but you mush together things that could change over time, which is just about everything in a survey like the ESS - that's why they do it every year. How the answers might change depends, of course, on what the questions are.

And things might change at a deeper level than just the answers to the questions. For instance, the correlation between two variables might change over time.

If you can make a reasonable case that nothing you are interested in will change much over the years you are combining, then go ahead. But someone might question you and you have to be able to defend your choice.