Home > Uncategorized > Naturalistic Inquiry and Silk Purses

Naturalistic Inquiry and Silk Purses

For me as a social computing researcher, some of the most insightful and useful parts of the research literature are qualitative sociology written thirty to fifty years ago. For example, The Great Good Place by Ray Oldenburg (1989) documents how people need a “third place”—one that is neither work nor home. Within the third places Oldenburg studied, he observed timeless patterns of human behavior like the role of “the regulars” and how people greet a newcomer. Or consider The Presentation of Self in Everyday Life by Erving Goffman, originally published in 1959. Goffman documents how people are always performing for one another, and the impressions they deliberately give may differ from the impressions given off (what people actually infer from someone’s performance). How could we even begin to understand social behavior on sites like Facebook and Twitter without Goffman as a foundation?

When I think about this style of research, I often think about how it would be reviewed if it were submitted for publication today. No doubt Goffman would be told his work was insufficiently scientific—how do you really know there are “impressions given off”? Can you measure them? I doubt he’d make it through the review process.

How do we know that Goffman’s observations are “true”? Honestly, we don’t. The “proof” is in how useful generations of researchers to follow have found his observations. Goffman took messy qualitative data, and used that as a touchstone for his own powers of inference and synthesis. His observations aren’t in the data, but emerged from a process of his examination of the data. It’s probably impossible to make a firm statement about how valuable this type of work is at the time of writing—the evidence is in the value that others find in building on the work later.

Contemporary researchers are still working in this style. I call writing in this mode “silk purse papers.” They say you can’t make a silk purse from a sow’s ear, except sometimes you can. Insights emerge from messy data in an act of interpretation. The first paper of my own that I would put in this category is my paper “Situated Support for Learning: Storm’s Weekend with Rachael,” published in Journal of the Learning Sciences in 2000. In it, I take a case study of a weekend in which one girl taught another to program and draw broader conclusions about the nature of support for learning—support for learning is more effective when it is richly connected to other elements of the learning environment. I can’t prove to you that those observations are valid, but I can say that others have found them useful.

I’m always vaguely embarrassed when I’m in the middle of writing a silk purse paper. A tiny little voice in my head nags, “You’re making stuff up. You could use this data to tell a dozen different stories. You’re saying what you want to say and pretending it’s science.” But then I remember Goffman, and try to cheer up. It may not be “science,” but it’s certainly a valid mode of naturalistic inquiry. Take a second and think about foundational writings in your area of expertise. How many of those papers are silk purses?

In fact I wonder if we are all always making silk purses. Just because a paper is quantitative does not make it any less interpretive. My colleagues and I just finished a complicated analysis of content people post on Facebook publicly versus post to friends only (to appear in CSCW 2017). It’s totally quantitative work. But what we chose to measure and not measure and how we explain what it means strike me as much acts of interpretation (to be evaluated by the passage of time) as the complicated stories we tell with qualitative data.

Where we as a field get into trouble is when people with naïve ideas about “science” start reviewing qualitative work, and get it wrong. Sociology of science and technology should be featured in more graduate programs in human-computer interaction and human-centered computing. And as a field we need to have a more nuanced conversation about interpretive work, and how to review it.

  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: