Despite popular belief that, as an English major, I am fundamentally opposed to quantitative analysis, I love numbers. Last spring I took Bekar’s Environmental Economic course and I truly believe that it was the best class I took at Lewis & Clark (so far). Despite my affinity for beautifully inefficient prose like that of Salman Rushdie a la Midnight’s Children, nothing satisfies me more than a clean graph. Perhaps my attention to numerical evaluation is reflected in my choice of sport: swimming. Success and progress in swimming is evaluated 100% on times. You don’t get points for swimming pretty (although this can sometimes help you swim faster) but only for getting your hand on the wall before anyone else. This is obviously a very reductive way to qualify an emotional and highly variant commitment to a team sport, but nevertheless that is the long and short of it.
The first week of ENVS 350, we experimented with a little quantitative theorizing with SPSS, a statistics program. We conducted several factor analyses on two different surveys, one administered via Facebook friends of ENVS 220 Fall ’15 students and the other a global survey. Both of these surveys sought to identify environmental values of the participants. You can find a more detailed write-up of the results here and here.
Although I came into ENVS 350 expected narratives and big words and language, I was very happy to find quantitative tools to help guide the linguistic and narrative aspects of this endeavor. As I wrote about in my first reflective post, theory seems to be (so far) a method of organization. Our factor analyses were able to identify and organize survey results at a very basic level that allowed us to explore different ways of organizing the responses. However, the biggest hiccup I had in understanding how factor analyses was useful was how I struggled to separate the original language of the survey from the factors we identified. It seems to obvious that the factors are determined by the original questions asked. Therefore, how can a factor analyses reveal anything that one couldn’t determine from simply reading the survey and looking as the results? Upon further reflection, I decided that factor analysis was a more efficient way to organize the survey results and an effective way of communicating them to an audience using tangible, numerical evidence.
Quantitative analysis here is fundamentally based on language and perception. The surveys are administered in language (and in the case of the ISSP survey, many different languages. I wonder how different languages construct environmental values? I imagine the etymologies of the translations varied quite widely. How can language and translation help reveal different and similar environmental values across culture? Across dialect?). What quantitative results cannot detect, however, are flaws or biases in the questions, the language, and the interpretation of these in the survey participants. I accept factor analysis only in conjunction with qualitative analyses of the questions, the sample size and the authorship. While some types of analysis can be fruitful in a sort of vacuum (New Criticism for example, a mode of literary criticism that relied solely on the words on the page), quantitative analysis is not one of these. Despite my affection for the elegance and concision of numbers, I take quantitative analysis with a large, qualitative grain of salt.