This week's lesson focused on segmentation of the audience to identify clusters (segments) of a population that differ on variables of interest. Aside from getting the requisite data, perhaps the hardest part of the segmentation process is determining which variables are of interest. The example given in class of food types-- red meat, poultry, fish, green leafy veggies, root veggies, milk, yogurt, etc -- and clustering them by diet style -- carnivore, pescatarian, lacto-ovo-vegetarian, vegan-- helped me conceptualize the process better. It also made me realize how easy it would be to create a homogeneous group just from diet alone. Imagine all the different reasons someone might be vegetarian: religion, animal rights, environmentalism, health concerns, general hippie status, relationship with someone who is vegetarian, lack of available meat, lack of resources to purchase available meat, allergy, etc. Also, is vegetarianism defined as not-meat? Perhaps vegetarians should be further separated into healthy and junk-food vegetarians by not-meat + fruits, veggies, whole grains, healthy non-animal proteins (i.e., healthy diet) and not-meat + few healthy foods and lots of unhealthy carbs and dairy products.
I thought the short-names applied to the groups presents an interesting dilemma. The PRIZM project provides a great example with catchy, but potentially offensive, names like "God's Country" and "Money & Brains." These may help as mental-shortcuts but it would also be easy to misconstrue the characteristics of the group because it overly generalizes. For instance, I imagined "God's Country" to be a highly religious, middle income family, but it's the name for upscale, highly educated, white, childless couple aged 35-54 who live in towns, buy from zappos.com, go on golf vacations and read about skiing. The Maibach (1996) article suggested this was a risk, stating "names will oversimplify the healthstyles which are, in fact, quite complex and rich with subtlety" (p269). I noticed this in one of the articles presented, particularly for the group called something like "Disinterested Males" which was composed of 85% males. It would be easy to forget that 15% are female, and if the communication campaign was heavily gendered, this might leave out a large number of the segment audience, particularly if done on a large or national scale.
My thesis, which is currently in the planning stages, will (probably) involve an intervention with a group of college men aimed at primary prevention of sexual assault. Initially I thought I'd focus on a single fraternity or fraternity men in general, but I worried that men who would not be receptive to the message would be included in the group and cause the whole thing to fail due to side-tracking or derailing the conversation. Now I'm wondering if I should include screening questionnaires to ensure that I have a more homogeneous and appropriate group. My best bet might be to simply conduct the formative research rather than be too ambitious and put together a program rather blindly. I've been pouring over the existing literature to figure out what strategies have been tried and what seems to have worked and ran across an interesting article in the journal
Trauma Violence Abuse by Casey and Lindhorst (2009) called Toward a Multi-Level, Ecological Approach to the Primary Prevention of Sexual Assault: Prevention in Peer and Community Contexts. It brings in examples of successful prevention programming in other areas (alcohol abuse, bullying, and HIV) to make suggestions for how to tackle sexual assault. I'm excited to be in this class and learning the public health perspective because otherwise I likely would have overlooked these important considerations in planning my thesis.
-Melissa