Here are my reading notes for week 4 of the Engaged Intellectual. Superficially these papers seemed oriented around the three cultures of social science, the humanities and the physical sciences. But there were some interesting cross-currents between them.


Crotty, M. (1998). The foundations of social research : Meaning and perspective in the research process. London ; Thousand Oaks Calif.: Sage Publications.

Crotty starts out by outlining a set of questions that must be answered when embarking on a research project:

  1. What methods do we propose to use.
  2. What methodology governs our choice and use of methods.
  3. What theoretical perspective lies behind the methodology in question.
  4. What epistemology informs this theoretical perspective.

He then goes on to define this terminology, while pointing out that that we often talk about them all together:

methods
the techniques or procedures Participant Observation
methodology
the strategy or design of the chosen methods that fit the desired outcomes (e.g. Ethnography)
theoretical perspective
the philosophical underpinnings for the methodology (e.g. Symbolic Interactionism)
epistemology
the theory of knowledge that is embedded in the theory ; how we know what we know (e.g. Constructivism)

The pragmatist in me wants to pause here to reflect that the real value does not lie in the truth of this picture of research, but that it’s a useful way of distinguishing between all the concepts that are flying around when learning about research. A firm understanding of these different levels helps ground the decisions made about what methodologies to use, and how to interpret the results.

The theoretical perspective is often assumed as part of the methodology and needs to be made explicit. Epistemology (how we know) is distinguished from Ontology (the study of being). They often get muddled up too. Corry points out that we often start with particulars: a specific problem that needs to be solved, a research question or set of research questions.

We plan our research in terms of that issue that needs to be addressed, a problem that needs to be solved, a question that needs to be answered. We plan our research in terms of that issue or problem or question. What, we go on to ask, are the objectives of our research? What strategy seems likely to provide what we are looking for? What does that strategy direct us to do to achieve our aims and objectives? In this way our research question, incorporating the purposes of our research, leads us to methodology and methods. (Crotty, 1998, p. 13)

The methodology needs to be defended, so that people will understand the results. Methodologies are created based on need, and it can help to understand the menu of methodologies that are available. Methodologies can also be combined, and of course new ones can be created.

The discussion has helped me partially unravel my own muddled thoughts about what I want to research from how I want to research. It at least helped me feel OK about being muddled! I have come into the program wanting to study Web archives, specifically the ways we decide what to archive also known as appraisal. At the same time I am interested in looking at these decisions as an expression of individual and collective notions of values in particular contexts. How are these values arrived at? What’s the best way to study them? As I’ve mentioned in the past reading Steven Johnson’s work about repair was an inspiration for me to enter the PhD program: so I’m hoping to use his ethnographic approach, possibly in combination with Geiger’s trace ethnography (Geiger & Ribes, 2011). But what this means as far as methods go I’m still not sure a) what methods this approach suggests, and b) if it they are a good fit for the problem I’m studying (appraisal in Web archives). Hopefully mixed methods will grant me some license to use several methods in a coherent way.

This was a lot to pack into a book introduction. I might have to return to read more Crotty (1998) when there is time since he was able to explain some pretty complicated things in a very clear, compelling and useful way.


Uzzi, B., Mukherjee, S., Stringer, M., & Jones, B. (2013). Atypical combinations and scientific impact. Science, 342(6157), 468–472.

It was really quite interesting having read (Crotty, 1998) beforehand, because it made me think about this paper a little bit differently in terms of method, methodology, theory, and epistemology. They are trying to measure the novelty of scientific ideas, in an attempt to show that high impact articles combine conventional and unusual ideas. At first it seemed like it would be extremely hard to try to measure novelty. The methods were firmly from descriptive and inferential statistics (the methodology), from statistical theory. Wikipedia tells me that statistics are from the Formal Epistemology school which uses model and inference to answer questions.

One of the epistemological foundations of the paper also seems to be that Science (writ large) is measurable by the body of scientific literature and the citations found within it. I guess the authors would agree that it is possible that non-documentary factors could determine the creativity and impact of scientific ideas as well. But this does nothing to necessarily invalidate their own findings. Since the Philosophical Transactions in 1887 the very idea of science has been explicitly tied to publishing. But this is a relatively new occurrence, and the history of science extends much further back in time.

Still the authors have done a nice job of using the structure of the literature to infer novelty. It’s pretty cool that they were able to use all the literature (no need to sample), and it was already structured and easy to process. Hats off to them for writing this article without mentioning Big Data once. I guess they must have had a contact at Thomson Reuters to get access. One caveat here is the biases built in to Web of Science when it comes to what is indexed and how. WoS is a black box of sorts. Also, only data for 1980-2000 was examined, and the scientific endeavor has been going on for a while.

I liked how they cited Darwin and Einstein so that they could increase their own z-score. The Ggogle-colored cubes in Fig 2 are also cute in their suggestiveness. What if this were applied to the Web graph, where a journal was replaced with a website, and an article was replaced with a web page?

At any rate, their idea of z-score is seemed like an interesting technique


Mayer, N. (2004, October). Reclaiming our history: The mysterious birth of art and design. American Institute of Graphic Arts.

Professor Kraus asked us focus on the rhetoric that Mayer uses when reading this piece. Immediately I was struck by a) the number of large colorful and really quite beautiful images, and b) the sparse layout, which read almost like poetry in places. I guess it helps to know that this was a presentation at a conference, presumably slides with reading notes. It feels like it was written to be heard and seen rather than read, although it reads well. Her format underlines her essential point that reading words is ultimately looking at symbols and imagery.

Mayer seems focused on helping us see how dependent we are on culture for interpreting symbols. I liked how she humorously used the the perceived penises and vulvas in the cave paintings to deconstruct the anthropologists who were studying the paintings. I also liked how she ultimately grounded the analysis in art, and the multiple ways of seeing as subjects, and as groups. I was reminded of Neil Gaiman’s Long Now talk How Stories Last, in which he talked about the effort by Thomas Sebeok and the Human Interference Task Force of the Department of Energy to figure out ways of warning people in the future about the nuclear waste stored underneath Yucca Mountain.

How Stories Last is a very entertaining talk about stories and information if you have 45 minutes spare. here is the specific segment about Sebeok’s work if you want to quickly listen. As Gaiman so artfully points out, Sebeok’s advice in the end was to use to use culture to create an information relay across the generations. Here are Sebeok’s words in the conclusion of his report:

It follows that no fail-safe method of communication can be envisaged 10,000 years ahead. To be effective, the intended messages have to be recoded, and recoded again and again, at relatively brief intervals. For this reason, a “relay-system” of communication is strongly recommended, with a built-in enforcement mechanism, for dramatic emphasis here dubbed an “atomic priesthood”, i.e., a commission, relatively independent of future political currents, self-selective in membership, using whatever devices for enforcement are at its disposal, including those of a folkloristic character. (Sebeok, 1984, p. 28)

This in turn reminds me of Janée, Frew, & Moore (2009) which talks about how digital preservation systems can model this type of relay…but that’s a rabbit hole for another post. I only brought this up to highlight Mayer’s essential point that our understanding of information is mediated by shared culture and context. We are lost without it. But understanding is always imperfect. All models are wrong but some are useful. Maybe learning the art of feeling good lost has its uses. Yes, I’m just seeing if you are still awake. Hi!

References

Crotty, M. (1998). The foundations of social research : Meaning and perspective in the research process. London ; Thousand Oaks Calif.: Sage Publications.

Geiger, R. S., & Ribes, D. (2011). Trace ethnography: Following coordination through documentary practices. In 44th hawaii international conference on system sciences (pp. 1–10). IEEE. Retrieved from http://www.stuartgeiger.com/trace-ethnography-hicss-geiger-ribes.pdf

Janée, G., Frew, J., & Moore, T. (2009). Relay-supporting archives: Requirements and progress. International Journal of Digital Curation, 4(1), 57–70. Retrieved from http://www.ijdc.net/index.php/ijdc/article/view/102

Sebeok, T. A. (1984). Communication measures to bridge ten millennia. (6705990). United States Department of Energy. Retrieved from http://www.osti.gov/scitech/biblio/6705990