Qualitative content analysis?

Question

Sorry for all the questions on content analysis I'm asking!

I want to do a content analysis of the images of women in men's magazines to find out how sexualised they are. Could this be a qualitative approach? As the book I got off of amazon states it is a positivist quantitative research method. If it is qualitative - how? What approach should I be taking? Help!

Does anyone know any good basic content analysis books? As I just want it to be a small part of my research.

I'm so confused so far that I feel I have no idea what I am trying to do, so any advice would be appreciated :-). So far I have only used qualitative methods such as focus groups and interviews so this is all a bit intimidating to me!

Answer 1

look at a film by Margret Mead on youtube.. She will give you an idea of some  qualitative and quantitative material... their are also a couple of lectures by the sociological  school in chicago, that has some great examples on Youtube good luck...

Answer 2

Have you taken a statistics course....???

Ok, you need to first find a scale that has been used and validated that assesses some of the variables you are interested in.  In this case "sexuality"...there are many on "attractiveness".  Always try to find a validated instrument rather than creating your own.  (library research using professional journals,..can't access on the internet).  You might try looking at apa.org  (you won't be able to read the article without membership, but you can find the source to go to the library and find that).  Essentially you need a scale that turns a "qualitative opinion" into a number.  If all else fails use a  Likert scale, from 1-5 or 1-7 (your choice...always an odd number so people cannot use a "middle point")....associate the ends with "words"....very s**y = highest number, not all all s**y to the least.  You can have several scales and associate the ends and some anchor points (all would be good) between the extreme's.  You can then use statistical analysis to see if men vs. women rate them differently.....you need to define your independent variables,....the sexuality ratings are the dependent variable.  If only you want to rate them, you need another to do so, then you need to go through inter-rater reliability analysis, or your research is meaningless.  Hope this starts you off.  What you are proposing to do has been done several times by psychologists, and is used extensively by advertisers to sell just about everything (especially cars)..  Good luck!  (college psych prof)