A survey among filmmakers: making sense of numbers

I designed a survey to investigate the experience of an international group of documentary filmmakers. A survey might not seem the most obvious way to research experiences related to ethics, but it definitively has its advantages. You can not only include many respondents from many places in the world (not least thanks to our modern communication tools) but also execute some complex analyses and find out more than the dutiful mean score and percentages.

For example, I conducted an Exploratory Factor Analysis. A factor analysis groups variables that are correlated. These variables together measure a latent variable. This technique is often used to test a model or theory. But I didn’t really have a model or theory, so I explored whether any meaningful factors, or patterns, could be yielded from my data. I found some, and then I had to interpret them, that is, think of what these groups of variables could measure. So for each group I created a new variable and all respondents got a score (based on their scores on the original variables).

Then I saw that some of these patterns were correlated, but one was not. Just like variables can be correlated. So I decided to investigate how these pattens might form additional patterns but on a more general level. I conducted a so-called higher order Factor Analysis.

In addition, I wanted to know to what extent specific other variables could predict scores on these new variables. To do this, I used a technique called Multiple Regression. With this technique I could basilcally find out which variables predicted scores and also what percentage of the diffences between respondents they explained.

I will tell you a bit more about this later.

I was very lucky that Andy Field had just published a thick book on statistics – the bible! It is so well written that even I understood more or less what he said – or at least I think I did. It’s simply called Discovering Statistics using SPSS (I think it sounds more exciting than statistics merits though). There is also a website (which probably represents a more general feeling about statistics). So thanks Andy, for helping out here!

A more concise introduction to Factor Analysis which I also found very helpful is Uses of factor analysis in counseling psychology research by Tinsley and Tinsley (1987).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s