Here is a downloadable article on
levels of analysis in research. The fact that in social science,
there is not just one level of analysis (the individual) but
potentially many (the individual, the team, the gender, the age
cohort, the organization, the province, the country...), is hard to
grasp for many. This has led to grave errors and misconceptions.
Geert once wrote an article to clarify the difference between the
various levels of analysis. It uses the analogy of flowers,
bouquets and gardens. If you are considering replicating his work,
read it!