A modified policy has was put in place recently to promote best breastfeeding practices. The researcher seeks to assess if the targets have been able to exhibit the intended behaviour and if not what barriers still exist.
At a basic level, measure % attempting breastfeeding; % breastfeeding at 1, 3, and 6 months; and % exclusively breastfeeding at those ages. Does that vary by (a) first-time mother, (b) education, (c) income level, (d) minority/non-minority, (e) spatially (eg, rural/urban)? And of course compare to the same stats pre-intervention, using a logit model controlling for the above factors. :-)
I support Ted Miller's suggestions. This is quantitative monitoring and evaluation, very important but if you don't have baseline statistics it can be tricky.
If you are able to, I think you could add qualitative results as well; this would make the evaluation a lot stronger, e.g. quotes from people who were exposed to the intervention and who did/didn't change their behaviour as a result.
As regards theoretical frameworks, these are usually set up before an intervention takes place. It might be worthwhile to read up what arguments and motivations were used to justify setting up the policy (perhaps interview a couple of the people who were involved in preparing and deciding to roll out the policy). There may be explicit and/or implicit frameworks in there.
You can certainly quote and/ or use the framework supplied by Asma, I find it really clear and helpful. But I think you should make it clear if you are introducing a framework at the evaluation stage or if the framework was already present beforehand.
i agree with Derek, on the need for baseline data. However, I doubt very much there is baseline data, especially since it's specific to a particular work force. Is it then possible to collect data, retrospectively from those who did not benefit from the policy and compare to those who benefited or are benefiting from the policy as Ted suggested?
A new theory, Multi-theory model (MTM) of health behavior change may interest you in this regard.
http://www.webmedcentral.com/article_view/4982
You need to decide whether you want to evaluate this as policy change or as health behavior change. However based on what you state the researcher wants to know I am not sure that you really want to do a theory-based evaluation. Ted's suggestion uses a descriptive epidemiology approach a and will answer whether the behavior change has occurred. For the second part you might want to find a theory that addresses barriers to breastfeeding (look at evaluations of breastfeeding interventions to see what they used); then use that theory to develop a survey to assess barriers. Or you could do an open-ended survey to assess barriers among those who did change and those who did not change.
I agree with Ted's method for behavior change but a framework or theory needs to be determined before the research takes place as stated by some respondents. Also evaluating the policy including its development process, reach, and evaluation is a different process. For the behavior change, triangulation of qualitative and quantitative findings (with generation of themes) would allow for a stronger research design and more answers.
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude