In preparation for my research protocol, I’ve identified that I want to combine interviews (what people say happens, with direct observation (what actually happens). We are charged in this journal entry with creating the instrument, and since I intend to use free listing and a pile sort in my interview protocols, I suppose now is as good a time as any to figure out what that looks like.
Here’s my plan:
Interview 1: “Hey there, dancer person, today I’m going to ask you to complete a ‘free listing exercise,’ in an attempt to identify qualities that make up good dance performance. What that means is, you can use this sheet of paper to write whatever comes to mind when you hear the phrase, ‘good dance performance.’ I’m going to leave the room to give you some privacy, and I’ll come back in 5 minutes. After you’ve finished, I might ask for clarification on some of the items that you’ve listed, so that I can understand what you mean. Does that sound alright?”
From this data, I will create a scree plot to identify the attributes that occur most, and visualize those attributes through photos taken of the company being observed. These photos will be used for Interview 2.
Interview 2: “Hey there, dancer person, today we’ve identified some recurring attributes based on an analysis of the free listing you and your colleagues performed a few months ago. For this interview, I’ve created some photos based on those attributes, and I’d like to ask you to look at the photos, and arrange them into categories. You can place the photos into piles however you want – based on how you think they should be organized. I’m going to give you a few minutes to complete this activity, and when I return I’d like to ask you some questions about why you categorized the cards the way you did. Does that sound alright?”
From this data, I will conduct a cluster analysis based on the categories selected and compare them across dancers, genders, role within the company, and between the two companies. This will then be compared across studies who have developed models through different theoretical lenses.
For example, if a dancer has six photos, he/she could categorize them as such:
- Pile 1: Pictures 1, 4 and 5
- Pile 2: Picture 6
- Pile 3: Pictures 2 and 3
I then ask why they’ve categorized them this way. The dancer might say something like, “1, 4, and 5 appear to have really beautiful lines, 6 has amazing feet, and 2 and 3 look really confident and powerful.”
Ok, great. So then, sort of like the NBA playoffs, I narrow those attributes to line, physical characteristics (feet), confidence/power, assuming that, according to this person, those are the things that matter in good dance performance. When I compare those things with the attributes identified by other people at that dance company, and across companies and existing models, well, minds could be blown.
When I read this chapter in the book, it all seemed a bit quantitative-y to me, but really it’s not – I’m still dealing with words here, and feelings and associations. I find this to be of utter importance when it comes to answering the question of how to measure performance, because ultimately, dance, or any artform for that matter, is about dialogue, and interpretation, and perceptions of the world around us. What can p values and chi squares tell me about that?