Close Home Forum Sign up / Log in

Cross Sectional / Longitudinal Analysis

N

Hi people!

I am writing up my proposal for my MA dissertation and essentially what I am doing is tracking the changes in labour market activation policies in various European countries over the past 20 years, trying to determining which changes in what countries were the most effective in reducing rates of long term unemployment.

I know this isn't a longitudinal analysis because it isn't the same people being followed throughout the twenty years...so would this mean it's a long-term cross sectional analysis??

I'm sure it's obvious but brain dead has settled in.

Hi,

Could it be a retrospective analysis that you are doing?

H

You're right it isn't longitudinal as you're not following up a cohort of specific people. Similarly it's not retrospective as that's like a backwards longitudinal.

It does depend on the data you have, and whether your exposure (being on the receiving end of policy) is measured at the individual level or the population level. Ditto the outcome (employment status).

I would say that if you are looking at individual people and what policies were targeted to them, and what their employment status was, it's a serial cross sectional study (i.e. repeated snapshots, different people at each measurement point).

If, however, you are just looking at population level changes in policy, and population level changes in employment at different time points, then I would say it was an ecological/correlational study (http://en.wikipedia.org/wiki/Ecological_study). Be careful what you infer from the associations as you do not know the 'exposure' and 'outcome' at the individual level.

I suspect the latter is more likely.

H

======= Date Modified 20 Mar 2011 21:05:06 =======
deleted because of multiple posting issues:$

H

======= Date Modified 20 Mar 2011 21:05:18 =======
deleted because of multiple posting issues:$

17803