Community Informatics and the Case Study

I'm going back again, but this time to case studies. As I work on my research proposal and look ahead to another assignment for another class, I realise that they are both well suited to case study comparisons. In my SSHRC proposal I didn't mention case studies because they weren't really on my radar even though I referenced a number of other case studies within the field of development as well as community informatics. I had also read a lot of case studies in my research for my SSHRC proposal, this final research proposal and my other assignment (which is a report on the evaluation and sustainability of CI driven projects in three Pacific Island Countries). Looking back through my readings, probably 90% were case studies. In one of the main texts I am using: Community Informatics: Enabling Communities with Information and Communications Technologies edited by Michael Gurstein has a whole section about CI case studies. There are no shortages of case studies in the CI field. Is this a bad thing? At first I thought perhaps, but then I checked myself and thought absolutely not (in my humble opinion). Yin (1981) states that the "distinguishing characteristic of the case study is that it attempts to examine: (a) a contemporary phenomenon in its real-life context , especially when (b) the boundaries between phenomenon and context are not clearly evident" (p. 59). In a field emphasises the importance of context, it is no wonder that CI researchers tend to create case studies.

Yin, R.K. (1981). The case study crisis: Some answers. Administrative Science Quarterly 26(1), 58-65.

Last week's class on ethics was fascinating both as a researcher but also as someone interested in the "field of information."  Because in a sense, ethics is all about information.  Do the participants have the information on possible health risks? privacy issues? possible risks to stress, embarrassment and all such other possibilities?  I think Dean said something to the effect that the benefits needed to outweigh the risks.  Who makes this judgement?  We heard about the composition of the review board.  While I think it would be fascinating to review all sorts of proposals, the responsibility to see beyond the words, to imagine the implications for all but especially for those who are not able to imagine (intelligence, ignorance..) is a great one.  One should not go through the motions when thinking about an ethics approval.  While it is always easier to imagine the benefits to oneself, it is harder to see some of the unintended or hidden drawbacks for the participants.
Mary

Identifiers

Like others before me (Karen, Jacqueline, and Victoria) I have been thinking a lot about participant anonymity. In his guest lecture, Dean Sharpe talked about delinking identifiers. Identifiers can be anything from audio or video recordings to name or e-mail addresses, etc. that can identify the participant. Even search histories.

Earlier this semester in my information policy class we were discussing the 2006 AOL search history debacle. An AOL researcher released over 36 million search queries and IP addresses of hundreds of thousands of AOL users collected over a three month period. The information was released intentionally but was intended for the research community for research purposes. People were able to identify the individuals based on their search queries. AOL had "de-identified" each user by assigning them a number, however, all of the search queries associated with an IP address were assigned the same number.

• Here is a New York Times article about one of these individuals: http://www.nytimes.com/2006/08/09/technology/09aol.html?_r=2. 
• And here is the Wikipedia entry about the scandal: http://en.wikipedia.org/wiki/AOL_search_data_scandal. 
• And here is an Ars Technica article about it: http://arstechnica.com/uncategorized/2006/08/7433/.

Lernert Engelberts and Sander Plug created a series of short films about user #711391 title "I Love Alaska." Here is the trailer:

I Love Alaska - Episode 1/13 from Submarine Channel on Vimeo.

You can view them all online here: http://www.minimovies.org/documentaires/view/ilovealaska. This is another Ars Technica article, but this time about "I Love Alaska": http://arstechnica.com/business/2009/01/aol-search-data-spawns-i-love-alaska-short-films/.

How de-identified is de-identified enough? It also makes a very strong case for encrypting data.

I was really intrigued by yesterday's discussion of doxxing on the web.  I felt really conflicted over what happened to the people's whose lives were ruined by having their online persona "outed".  People's online anonymity is important to them.  As someone who frequents Reddit I see a lot of people who will create temporary, throw-away accounts before posting something sensitive.  These throw-away accounts are usually created when a fellow Redditor poses a question to the Reddit community.  People want to answer but they don't want anyone to be able to trace the answer back to them.  There's a need to share experiences with people online but preserving anonymity is crucial.  I think the same could be applied to interviews.  People might want to share their experiences or opinions but are scared that their identities might be revealed.  I think that if a researcher approached the same Redditors and asked them to do an interview there would be hesitation.  An in-person interview feels less anonymous than replying to questions in an online forum even though the researcher probably has a much stricter confidentiality  agreement.    

I'm all for ethics but ...

The Stanford prison experiment has always intrigued me - I guess not only the Stanford experiment but also the Milgram shock experiment. Lacking ethics on so many levels, it is disconcerting to consider that these experiments took place in the 20^th century and baffling further that they occurred in the 1960’s and 1970’s. North America was amidst the me decade - the peace and love era was still going full force. Feminism, environmentalism, anti-war protests, and importantly the civil rights movement defined the times. Of course, it was all about to change. All that touchy feely stuff was about to go out the window. Did the Stanford experiment signal the beginning of the end? Did this experiment reflect the beginning of a time of disconnected research in which humanity would be viewed as secondary to science? People in the Stanford experiment were clearly distressed and deeply affected in extremely negative ways. That being said, I have to admit that I am still intrigued. The scientific findings of this unethical experiment would never have been discovered otherwise. In fact, ethics in science have likely evolved more fully because of these kinds of experiments. Does this make it right? Probably not - But still, we cannot erase their history and must honestly consider whether it is really entirely regrettable when we think about the important insight that such an experiment has revealed to us about human behaviour? Scientifically rigorous data is often revealed despite corrupt ethics. I have a really hard time reconciling this because we can all look back on these archaic experiments with hindsight. Yet somehow, I do think that some experiments need to be constructed a little less so ethically stringent. Sometimes proposals for good experiments are rejected because control groups may not receive actual therapy or medication because the expectation of obtaining treatment may be present. This can be despite a clear directive by the scientist that not everyone will receive therapy. Is this really an ethical dilemma or can some ethical rules be so stringent that they more about being politically correct than doing what is ethically acceptable? Are there grey areas of ethics and experiments that warrant more consideration and flexibility in order to be explored more comprehensively within research?

Knight: Doing it

Chapter 7 of Knight in part discusses, sensitive or distressing issues. He brings to light how social research can be harmful, for example "resurrecting distressing memories or by exploiting them with flattery so as to get disclosure without any commitment to give them something back". He also explains how harm can come to the researcher, physical if they put themselves in dangerous situations to get information, or even emotional, as "finding out more about hurt, misery and terror can corrode the spirit". Both of these observations I found very interesting. I never thought about the fact that by asking people to participate freely we are really not giving them anything in return. Also I have never considered the risk that could come to a researcher (maybe because non of my research interests are "risk" related). Yet I think these are very important points to keep in mind. Knight includes a part where he suggests having a friend or even professional council to talk to (when doing this kind of research). Which I think is a wonderful suggestion, however I am not sure if anonymity is maintained if you then go on an talk about what has upset you!

For building trust to get real answers Knight gives 12 specific points:

1. Getting potential participants’ informed consent by telling them clearly what the research is about.
2. Explaining why they are interested in exploring something that is normally private and why it is valuable.
3. Promising confidentiality and anonymity.
4. Where this is both possible and ethical, inviting potential participants to reassure themselves by talking to other people who have participated.
5. Doing the inquiry in places where participants feel safe. In interview studies it is usually important to find private space.
6. Being a good listener and observer. Good interviewers know that that is nothing like being a good interrogator: humility is needed.
7. Being ready to disclose information about similar experiences they have had. For example, it helps that I can talk with teachers about how we have felt when we have taught classes that have been more intent on larking than on learning.
8. Being able to empathize, which means being able to show informants that they understand and maybe sympathize too.
9. Not persisting if informants signal that they do not wish to go further.
10. Where it is feasible, inviting participants to edit transcripts, to add to them as well as to remove sections that, on reflection, they do not wish to be used.
11. Blending in. There are things that researchers cannot affect (their age) and things that they can (clothing).
12. Becoming an insider. This is not always possible but participant researchers can establish their claim to be trustworthy through becoming close to the people they want to study. On the other hand, Hughes (1996) tells of a participant researcher studying police in Amsterdam who only found at the end of the study that he had missed all the evidence of police corruption.

Do you agree with all of these steps? or find any of them problematic?

Throwback: Critical Discourse Analysis

I have been percolating this post for the past two weeks, so it is a little out of context now. To be frank, critical discourse analysis makes me want to pull my hair out. I also found that after reading the van Dijk, Fairclough and Knight I was left wondering how to go about critical discourse analysis. I have definitions to the brim of what it is, why it is useful, and it’s importance to understanding social inequalities.

Chaya used Fairclough’s definition to guide her lecture. He describes critical discourse analysis as a:

[D]iscourse analysis which aims to systematically explore often opaque relationships of causality and determination between (a) discursive practices, events and texts, and (b) wider social and cultural structures, relations and processes; to investigate how such practices, events and texts arise out of and are ideologically shaped by relations of power and struggles over power; and to explore how the opacity of these relationships between discourse and society is itself a factor securing power and hegemony (Fairclough, 1993, p. 136).

I am taking a class on information policy and I think a critical discourse analysis approach or perspective would be interesting when reviewing copyright law. We were recently discussing copyright law how it privileges a Western context and lacks consideration of traditional knowledge. And the week before last we were discussing censorship. However, it is tricky because within this class we have been warned not to let our personal values or biases affect our research. Yet according to vanDijk (1993) “critical discourse analysts (should) take an explicit socio-political stance: they spell out their point of view, perspective, principles and aims, both within their discipline and within society at large” (p. 252). For example, during our discussion of censorship we focused on social media in China. In 2005, Microsoft complied with Chinese censorship laws and took down edgy journalistic blogger Zhao Jing’s blog. In response Western activists, researchers and even society in general cried foul. However, the blogging community in China understood the sacrifice, because if Microsoft hadn’t complied then they would have lost the communication platform entirely. "Powerful socio-political change can be expected to emerge as a result of the millions of online conversations taking place daily on the Chinese Internet: conversations that manage to stay comfortably within the confines of censorship" (MacKinnon, 2008, p.45). So when I think about this example with a critical discourse analysis perspective, choosing to take up the specific agenda of censorship in China will my research be well met? This is where I get a bit confused. I am so focused on censorship that I fail to see Chinese bloggers pushing boundaries everyday within the censorship framework. I also want to clarify that I am not saying I think everything is peachy keen in China in regards to censorship, I just wonder about checks and balances in critical discourse analysis.

In relation to my research project, I am examining how the governments of three Pacific Island Countries are using ICTs to deliver services and promote social and economic development. I am using Community Informatics as the framework for my research. What does this have to do with critical analysis discourse? Well, according to van Dijk (1993) critical discourse analysis “is primarily interested and motivated by pressing social issues, which it hopes to better understand through discourse analysis” (p. 252). In an alternate version of my project I could explore the discourse of social and political development in the South Pacific. I think.

Beyond approaching topics with a critical discourse analysis perspective, how do you conduct critical discourse analysis? As I mentioned above, I was left wanting. van Dijk in a way addresses my questions in his section on “Reproducing racism in the British House of Commons.” I really wanted someone to come out and say you could use Spotfire to preform sentiment analysis and glean new insights—or something of the like.

Is anyone using (or used in the past) critical discourse analysis?

Case Studies

I love case studies. Ok - maybe not as much as the authors of the Zimbabwe Bush Pump study love their subject, but close. I get the analogies. Yes they are studying without a critical distance. There is fluidity in the production of knowledge. The bush pump is immersed and somewhat inseparable from its environment. All that is what can be argued of case studies. That being said, I am far more critical of case studies than so-called evidence based studies. The other day, a visiting scientist in my other research course was presenting her study proposal on a smoke free public housing legislation study. I felt very much like I had to be critical of the study. What do you mean you triangulated data results? Where was the definitive data on the matter? What the hell kind of scientific study was this? But then, isn’t it better to be critical? At the end of the day, the case study she described was amazingly complex. Both qualitative and quantitative, it examined almost every angle possible. It provided clarity and explored nuances that would have been impossible to see from a purely experimental or quantitative perspective. Isn’t that the value of case study? Aren’t case studies every bit as evidence based as experiments, particularly when they demonstrate knowledge that is unique and unavailable from every other type of study? I guess I am just sick of the diversity of science argument. I should love that word but to me it is just a word that begs for acceptance based on the "we are different but we are ok too" mentality. Perhaps case studies are just good because they are good - because they find stuff that nothing else can. Perhaps they are scientifically better in many ways. Isn’t that why we should love case studies?

In the Spirit of Luker

Today's news (and blogging frenzie) about the conservative government trying to "muzzle" librarians reminded me of something Luker said earlier in her book. She calls librarians the "pit bulls of democracy - as our government increasingly tries to hide things from us, librarians are among the few sounds fighting back" (Luker, 85). Today, more than ever, I can't help but see the truth in this statement.

As librarians, and more generally as information professionals, we need to take government cut-backs and the decisions made by LAC director, Daniel Caron (a non librarians/archivist!!) seriously to advocate for our right to disseminate information to the public and stand-up for our profession.

Ra Ra Ra!!!

Check out these links:
http://news.nationalpost.com/2013/03/15/library-and-archives-canada/
http://boingboing.net/2013/03/19/canadian-government-muzzles-li.html

Odiousness of our craft--
I was reading "Thinking About Research..." group blog and came across Howard's point of ethics and his concern that we don't discuss the ethical implications of what we as researchers do.  A few weeks ago when we talked about interviews, someone mentioned how that either before or after the interview, the researcher engages in some small talk with the subject and sometimes the subject lets down her guard and says something that she doesn't want to share.  What does one do with this?
I'm likely dating myself but I used to watch a TV detective show entitled "Columbo."  Columbo had a method of acting dumb and confused and just as the interview ended, he would ask a "clarifying" question.  Usually, he caught someone off guard and was able to get some information that he would use to solve his case.  I can see how this technique worked for Columbo but in a research setting, it seems to be deceitful.
Mary

With my research proposal's strategy being the case study, I looked forward to this week's readings.  Now, I'm looking forward to the class discussion because I'm not quite sure of myself.  Yin's article assured me of the value of the case study as a research tool.  I liked his clarification about the types of evidence, the methods for data collection and the research strategies.  He argues that the case study does not imply use of a particular data collection method and that what distinguishes the case study is that it attempts to examine a "contemporary phenomenon in its real life context."  It is at this point in the article that I start to question my own proposal.  I think I need to think about this so-called "phenomenon."  Yin argues that the case method is used when the boundary between the phenomenon and context is not clearly evident ( unlike an experiment where the boundary is made.)  I need to think ( and write more about this.)  The last point that I like to ponder is Yin's classifying the types of case studies--explanatory, descriptive and exploratory--I'd like to think more about these differences.
Mary

Zombies, Higher Education and Research: Increasing University Relevance and Making Research Count

I really enjoyed reading Naylor’s address to the Empire Club of Canada and thought that I would share some of the highlights of this address that are relevant to us as participants in this course, as members of the university and as future graduates of the iSchool:

•  On Page 11, Naylor distinguishes between unfettered research funding and fettered research funding which is funding tied to specific partnership and programs.  While the following data was gathered only for NSERC funding for UofT from 1983 to present, I think it can also be applied to the SSHRC funding:  about $230 million of federal funding has been reassigned from unfettered funding to fettered funding over the past 5 years.  According to Naylor, this has happened nationally as well, but, and to me this is a huge but, the innovation and competitiveness indicators didn’t improve.  Naylor states, “In fact, the real problem was never the type of research that universities were doing. Wrong diagnosis; wrong prescription. It was business related R&D spending that lagged” (11).

•  Applied research is always great, but Naylor also addresses “why great basic, disruptive, fundamental research matters” (13), and I like his third argument the best.  Naylor states, “The third and final reason why serious fundamental research matters is that the distinction between fundamental and applied research is somewhat misleading.   As Nobel laureate Sir George Porter famously pointed out, there is applied research and yet-to-be applied research” (12)  He provides examples of researchers who began researching one thing but ended up finding awesome things that weren’t within the scope of their study, and their research led to advances across many disciplines. At least once per class, it is mentioned that  good research crosses disciplines and boundaries, and the same is said about successful blogs.  Naylor makes the same argument:  “One needs excellence in research and scholarship across disciplines because no one can predict how disciplines will collide.  So much of the best innovation is convergent” (13)

• There are a few more points that I would have liked to address, but I am already way over the word limit.  But quickly, last week we discussed interpreting data and how quantitative approaches to interpretation can be used as a blunt instrument, and how this misuse should propel readers to question who is speaking, who conducted the research, who was interviewed as part of the research process; anyhoo, the information provided on page 15, 16 and 17 of Naylor’s address deals with a lot of the themes and issues we have dealt with in this course.  Naylor discusses the different kinds of data that are collected for university rankings, how the different factors are measured and who was interviewed to gather the stats. 

http://www.president.utoronto.ca/speeches/zombie-vs-zombies