Monday, 4 March 2013

"Salsa Dancing into": Participant Observation

Luker (2008) coins the phrase “salsa dancing into the social sciences” to describe her way of doing research. She argues this method strikes a balance between academic rigour of structured scientific methods and a more open ended pragmatic approach. When reading Luker I have derived that the main traits of the Salsa Dancing social scientist are found in contrast against what she terms the canonical social scientist. These basic traits are an ability to review multiple methods and theories looking to the best ones for your specific research as well as a flexibility of thinking. This flexibility that ensures you do not get stuck in a paradigm a theory or method that might cause you to miss important factors in your research.

When reading Shaffir (1999) it strikes me that ethnographic research contains the traits of flexibility of salsa dancing social science. This is demonstrated when he states “Any attempt to codify the process - much less to force it into the rigid protocols of "hard science" - is to miss the point,” later adding “practitioners and teachers came to understand that there was no simple best way of conducting it, that experiences were highly variable, and that approaches had to be adapted to the particular research problem and setting” (pg. 677, 679). As a result of the flexibility of this method I first thought that participant observation should be automatically defined as a method that fit with salsa dancing social science. On review of Luker’s (2008) salsa dancing definition I realized that she considered traditional field research also known as participant observation to be on the opposite end of the spectrum of canonical social science (pg.2). In other words Luker suggests that there is not a concrete enough methodology for participant observation and it is too variable to be automatically considered part of salsa dancing social science. Luker states that participant observation is still useful, but she seems to limit this use to what she terms theory building (pg. 160). For me participant observation is a necessary method to understand actions and behaviors in a social setting. As Mandi mentioned in her earlier post “Hang around, observe, and record [your] observations” ethnographic research is an important way to determine the difference between what people say and what people do. For these reasons participant observation will continue to be important methods, but we should be aware of its limitations.


Luker, K. (2008). Salsa dancing into the social sciences: Research in an age of info-glut. Cambridge, MA: Harvard University Press.
Shaffir, W. (1999). Doing ethnography. Journal of Contemporary Ethnography, 28(6), 676-686.


  1. 'The flexibility of thinking' coupled with the act of ethnographical research; that of cherry-picking from a list of research options the research method one thinks will produce the best results. By naming ethnographic research, and defining what it is not, it could be argued that the authors are indirectly establishing what it is, in effect, 'codifying the process'? It's only my opinion, but it seems to me that the while the flexibility of thinking may support the act of observing, it loses some, if not all of its flexibility once those observations are transcribed into text.

  2. That is an interesting thought that I had not considered. I do agree that when Shaffir describes what not to do it represents a form of codification. I believe he would argue however, that the flexibility lies with no clear statements on what one needs to do to conduct participant observation. Still, if a researcher limits the possible actions to a high degree this will limit the choices and by extension the flexibility that is possible. Thinking back to Shaffir’s woodman metaphor, the woodman could only tell the lost man what roads not to take. When I think about this I wonder if the woodman was able to exclude the majority of the roads, would there not be few possible paths left? In both the Shaffir and Stebbins articles I do not think that they present enough descriptions of what not to do limit the possibilities to a great extent. For me they still leave participant observation to be a flexible and variable enterprise.