I agree with Clarke & Primo's perspective on social sciences' "physics envy," but I wonder if the issue they raised is ultimately one of ontology, namely -- maybe the problem lies in social scientists' zealous idolization of positivism in their quests to find "The Truth" when they primarily deal with the most fickle, confused, and questionably rational subjects on this planet: human beings. After all, whether as the observer or the participant in a scientific study, who is to say what either party sees or reports is "objective" from a positivist lens?
In other words, it is not about the "method" of choice, but the ontological drive that the researcher employs to examine the phenomenon at hand. Whereas in physics there may be ultimate "answers" (e.g., E=MC^2), in social sciences, quoting one of my favorite lines from Solaris -- "there are no answers, only choices [in how we see, measure, and interpret things]."
To me one of the most exciting things about social sciences lies in its attempt to understand the intricacy, or 'grayness' of human world. And so who is to envy whom? Hard to say...
Social scientists need to embrace the unique unpredictability of humanity in their quests to investigate the world from a systematic, replicable manner, and to see in every gray difficulty great opportunities rather than the other way around.
p.s. On an unrelated note, maybe it's because I've just taken Dr. Lasorsa's Theory Building class last semester... but if the purpose of theories is to explain AND predict things, then I think 'theories' a priori need to be empirically testable and replicable, contrary to what Clarke and Primo suggest. I'd be interested to hear what others think about this.
The example of candidates having the same platforms used in the article is interesting in considering your question about explaining and predicting. The theory explains AND predicts, even though the prediction will never come completely true (that candidates will have exactly the same platform). Rather, it describes a force and explains how it acts, then predicts in which direction things flow, rather than predicting a precise outcome. This broader conception of "predict" could be useful in developing theories that describe, explain AND predict.
ReplyDeleteGood point Logan. Broader, directional predictions are indeed used often in quantitative research if we consider one-tailed versus two-tailed tests for statistical significance.
Delete