International Day Against DRM
I’d just like to take a moment to post this:
May 4, 2012
· 
Sean · 
No Comments
Tags: drm, ebooks, lockin, software · Posted in: Software
Bibliometric Exploration of Autoethnographic Research
When I was doing coursework in qualitative methods, I read something, perhaps just a side note, about the autoethnographic approach. I still haven’t read any autoethnographic studies, but I’m curious about it and wonder if or how it might useful as a research design for information seeking and use studies.
My own first step in exploring the literature in a given area is to take a gander at its scholarly communication patterns. My understanding is that autoethnography is relatively new but is gaining popularity. To gather data about its publication history. I did a simple Scopus search for “autoethnographic” in the title, abstract, keyword field. This resulted in 518 documents with a publication range from 1993 to 2012. I downloaded the data as a CSV file and imported it into a R workspace (using RStudio).
Caveat: Aside from it being kind of funny that I’m using basic quantitative bibliometric methods to look at this highly qualitative style of research, this is only a cursory exploration — a way for me to see the whole picture before I dive into the details. To keep this from being a 20 page post, I’m not going to be doing much textual interpretation here. I’ll leave that to the reader.
The following is a histogram (using ggplot2) of counts by publication year.
> qplot(Year, data=scopus, geom="histogram", binwidth=1, xlab="Publication Year", ylab="Frequency of Documents (Count)", main="Prevalence of Autoethnographic Research Publication")
The histogram helps confirm that this is a relatively new kind of research design. I’m kind of curious where most of this research is published. Here’s a list of ten journals that have published the most autoethnographic articles. I generated this list using the summary command, where the downloaded CSV file is called scopus and the variable for the publication source is called Source.title:
> attach(scopus) > summary(Source.title, 11)
- Qualitative Inquiry: 125 publications
- Cultural Studies – Critical Methodologies: 46 publications
- Journal of Contemporary Ethnography: 16 publications
- Journal of Research Practice: 10 publications
- Qualitative Report: 10 publications
- Forum Qualitative Sozialforschung: 9 publications
- Journal of Psychiatric and Mental Health Nursing: 9 publications
- Qualitative Health Research: 9 publications
- Symbolic Interaction: 8 publications
- Athenea Digital: 6 publications
Now I’d like to see the overall distribution of publication sources. To do this, I need to manipulate some data. The Source.title variable only includes the list of journal titles. What I need to do is create a data frame with journal titles and frequencies. Then I can create a density plot. Here’s the code:
> x <- table(Source.title) # generate a table from the Source.title variable, recall I've already attached the data set > head(x) # inspect the data (not shown here) > x <- as.data.frame(x) # convert table to a data frame > names(x) # look at the variable names, which are "Source.title" and "Freq" (not shown here) > qplot(Freq, data=x, geom="density", xlab="Frequency", ylab="Density", main="Journal Titles: Density Plot")
As we can see from the density plot, the top ten list from above pretty much captures the majority of autoethnographic research publications.
Let’s look at some other data. Specifically, I want to examine the citation counts. After looking at the data set, I see that any document that doesn’t have a citation count has a data value of NA. I need to replace NAs with 0s (zeros).
> scopus$Cited.by[is.na(scopus$Cited.by)] <- 0
Now I can generate some basic descriptive statistics of the citation counts:
> summary(Cited.by) Min. 1st Qu. Median Mean 3rd Qu. Max. 0.000 0.000 1.000 3.168 3.000 85.000
When the third quartile of data is a smaller value than the mean, we know there’s some skewness in the sample distribution. First we’ll look at a histogram of the entire data, and then we can generate small multiples by document type, which include Article, Article in Press, Conference Paper, etc., and year.
> qplot(Cited.by, data=scopus, geom="histogram", bindwidth=5, xlab="Citation Counts", ylab="Frequency of Documents (Count)", main="Histogram of Citation Counts")
And below we can look at each document type’s citation counts by year. Part of my intention in presenting this in small multiples is to explore ways to present this data. If anyone would like to comment on how readable they think it is, please feel free.
> qplot(Cited.by, Document.Type, data=scopus, facets = ~ Year, geom="jitter", xlab="Citation Counts", ylab="Document Type", main="Citation Counts by Document Type by Year, as of March 2012")
Let’s take a look at the descriptive stats for each document type:
> l <- aggregate(scopus$Cited.by, list(DocType = scopus$Document.Type), FUN=length) > s <- aggregate(scopus$Cited.by, list(DocType = scopus$Document.Type), FUN=summary) > cbind(l,s$x)
DocType x Min. 1st Qu. Median Mean 3rd Qu. Max. 1 Article 424 0 0.0 1.0 2.6250 3.0 71 2 Article in Press 5 0 0.0 0.0 0.0000 0.0 0 3 Conference Paper 8 1 2.5 3.5 9.1250 6.0 47 4 Conference Review 1 0 0.0 0.0 0.0000 0.0 0 5 Editorial 6 0 0.0 0.0 0.6667 1.5 2 6 Note 6 0 0.0 0.0 0.6667 1.5 2 7 Review 65 0 0.0 1.0 6.8770 7.0 85 8 Short Survey 3 0 0.0 0.0 0.0000 0.0 0
Now let’s take a look at the authors in this data set. Since the Authors variable is similar to the Sources.title variable, that is, because there are so many names, we have to convert the Authors variable into a data frame with author names and frequencies as our variables:
> x <- table(Authors) > x <- as.data.frame(x) > names(x)
[1] "Authors" "Freq"
> head(x) # examining the first five entries of the new data set, sorted simply by alphabet
Authors Freq 1 Adams, T.E. 4 2 Adams, T.E., Holman Jones, S. 1 3 Aguirre Armendáriz, E. 1 4 Alexander, B.K. 1 5 Alexander, B.K., Moreira, C., Kumar, H.S. 1 6 Allen-Collinson, J. 1
Now I want to see how much each author is publishing in this area. I’m not analyzing coauthorship at this time because the CSV file I downloaded from Scopus contains multiple authors in a single cell. It’s not too hard to break that up and to put each author in its own vector space when there is coauthorship. For now, if two or more authors published a single paper, I’m counting that as single authorship, just to keep things simple. However, if co-authors have published multiple times but have different naming orders, it’ll be counted as two separate authors.
> m <- qplot(Freq, data=x, geom="bar", binwidth=1, xlab="Authorship Frequency", ylab="Document Type Count", main="Authorship") > m + xlim(0,10)
As we can see, most “authors” have only published one paper in this area although we do see that some “authors” have published multiple times. It suggests to me that there’s a core group of researchers in this area (where Author Frequency >= 2), but that for many, it’s perhaps their first step into this research design.
Well, that’s enough for now. I’ve gotten enough of a bird’s eye view of the field to help me put it in some context. I think it’s now time to retrieve an autoethnographic article I’d like to read.
April 5, 2012
· Sean · No Comments
Tags: autoethnography, bibliometrics, R programming, research designs, scholarly communication, visualization · Posted in: Scholarly Communication
Scholarfy, OpenLibraryfy, WorldCatify, Merlinify
Johan Ugander, a PhD student at Cornell, posted a nice hack that converts any Web search that uses a query string into a Google Scholar search (Ugander notes that his Scholarfy hack is good for a Google or Bing search conversion into a Google Scholar search, but it should be good for converting any query string into another). It’s a great hack because, as Ugander points out, Google recently, and unfortunately, demoted Google Scholar to a third tier search service.
I explained in a post on Google Plus that this can be used for your local library’s online catalog. So below are three modifications of Ugander’s hack, or bookmarklet. To use them, simply drag them to your bookmark bar. The first one is a bookmarklet that converts a query string (ex: a Google search) into an OpenLibrary keyword search. The second one does the same thing for a WorldCat search. The third one is a bookmarklet for those of us at the University of Missouri who may want to convert a query string (ex: a Google search) into a keyword search of our own online catalog. Either of these can be modified to suit your oft-used needs by replacing the query string to match another’s. The strings may also be modified so they’ll conduct advanced searches. For example, for the online catalog, the string can be modified to convert a search into a subject search.
[UPDATE: WordPress truncates the bookmarklets, so I'll add them to my academic homepage: http://web.missouri.edu/~csbc74#misc]
March 2, 2012
· Sean · No Comments
Tags: bookmarklet, google scholar, query string · Posted in: Search
The future of libraries is librarians?
T. Scott Plutchak[1] has written a very nice article addressing the critique that libraries (specifically academic libraries) are on the decline. Of particular interest for me, given our ALISE presentation this January (with Matthew Altman) is the confusion about agency in our field — specifically, the concern about the use of ‘libraries’ as if libraries are agents, which entail they can act. Matt and I address this issue in relation to our field’s stated ethics[2]. We pick on the Library Bill of Rights for its wording — ex: “Libraries should provide …”, “Libraries should challenge …”, and so forth, and argue that more accurate (theoretically and not just grammatically) ethical propositions should be re-formed. That is, the term “librarians” should be used — ex: “Librarians should provide …”, etc.
Plutchak says what he learned as editor of the Bulletin of the Medical Library Association (BMLA) was that
throughout this varied catalog of articles [articles submitted to the journal and that he was reading as the editor], I noted a consistent thread. Whatever it was that was being done, it was the library that was reported as the agent. It was “The Library” that had initiated this new program or was engaged in this particular interesting project. The more that I noticed this, the more it began to trouble me. “Libraries,” I thought, “are just buildings, or gatherings of objects, or an abstract diagram on an organization chart. Libraries don’t do anything—people do.”
Interestingly, in an early view article from the Journal of the American Society for Information Science and Technology (JASIST), Vincent Larivière, Cassidy Sugimoto, and Blaise Cronin use bibliometric measures to chronicle the first hundred years of library and information science. There’s a good tie in between their article and Plutchak’s (and to some other work that the author from the JASIST article, Cassidy Sugimoto, has done with others; see Finlay, C.S., Sugimoto, C.R., Li, D., and Russell, T.G, 2012). Under the section “Stable and Declining Words,” Larivière, Sugimoto, and Cronin find that the truncated term, librar*, as well as the term, librarian, have both declined (proportionally) over the time period as apparent subject and research topics in library and information science literature. For the latter term, it’s a fairly striking decline. They write that
the term librar* was found in nearly 70% of all titles in the early part of the century but the corresponding figure today is 16%, the most dramatic decline of all (although this still represents a large proportion of the overall literature). However, this is not necessarily an indication that the absolute amount of library literature has declined, but rather that the proportion of such literature has declined as information science journals were founded (Donohue & Karioth, 1966). Librarian has also decreased, from 7% in the first part of the century to less than 1% today. The most precipitous decline happened between 1960 and 1980, when many information-related terms entered the literature (early view edition, p. 11).
Of course, this is a bird’s eye historical view, but it’s certainly worth further investigation — and action. My hope is that Plutchak’s argument can gain some momentum and reverse the tide a bit in terms of research focusing on librarians. His is a very nice article about the potential great age of librarians versus the previous great age of libraries (there’s certainly an interesting corrollary in the literature between the shift in focus from libraries to users).[3]
- Finlay, C.S., Sugimoto, C.R., Li, D., and Russell, T.G. (2012). LIS dissertation titles and abstracts (1930-2009): Where have all the librar* gone? The Library Quarterly, 82(1), 29-46. http://www.jstor.org/pss/10.1086/662945
- Larivière, V., Sugimoto, C.R., and Cronin, B. (Early view). A bibliometric chronicling of library and information science’s first hundred years. Journal of the American Society for Information Science and Technology. doi:10.1002/asi.22645
- Plutchak, T. Scott. (2012). Breaking the barrier of time and space: The dawning of the great age of librarians. Journal of the Medical Library Association, Â 100(1), 10-19. Available here: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3257492/
[1] I happened to come across Plutchak’s paper from a Scholarly Kitchen blog post by Kent Anderson, located at http://goo.gl/p9JDH
[2] I defend my dissertation proposal next week, which happens to to address the marginalization of academic libraries. After that, Matt and I will finish and submit the paper we presented at ALISE. In the meantime, our abstract is located at (search for either
Altman or Burns to locate it) http://goo.gl/pcICt
[3] I originally posted this on Google Plus. This is a slightly edited version of that post.
March 1, 2012
· Sean · No Comments
Tags: action, agency, bibliometrics, ethics, professions · Posted in: Librarians
ALISE 2012 Conference
My friend, Matt Altman, a philosopher at Central Washington University, and I have been working on a paper entitled, “Who or what is constrained by the Library Bill of Rights ? A theoretical analysis of intentionality and collective action for library and information science.” Matt will not be able to attend, but the paper has been accepted by ALISE and I’ll present it next January at the annual conference.
September 18, 2011
· Sean · No Comments
Tags: alise, conferences, ethics, librarians, library bill of rights · Posted in: Studies
Scholarly Communication: MU Output
To help my students see an example of scholarly communication growth, I compiled some MU scholarly publishing stats. See http://t.co/F2myLtJ.
The most interesting aspect: in 2000 there was a 1:1 relationship between the number of faculty and the number of publications. In 2010, it increased to 1:1.5.
September 12, 2011
· Sean · No Comments
Tags: academic publishing, bibliometrics, citation databases, scholarly communication, scholarly publishing, university culture · Posted in: Scholarly Communication
UK photos
I’ve added a selection of photos from our trip to UK. Here’s the link:Â https://picasaweb.google.com/sean.csb/UKPhotos?authuser=0&feat=directlink
August 20, 2011
· Sean · No Comments
Tags: international librarianship, photos, public libraries, UK · Posted in: Studies
Edinburgh Central Library
The other morning we met with librarians at the Edinburgh Central Library. In the last 18 months, they’ve launched an aggressive digital initiative, pushing e-books, digital libraries, social media, and the like. It’s still a young initiative, but it seems to be working well. They have a local community site: yourlibrary.edinburgh.gov.uk. Will write more on return from UK.
August 13, 2011
· Sean · No Comments
Tags: international librarianship, public libraries, social justice, UK · Posted in: Studies
In the UK
I’m in the UK for a couple weeks with a group of SISLT graduate students and a couple professors. We’re out here talking about social justice and librarianship, meeting librarians, visiting libraries, attending conferences, and touring London, York, Leeds, and Edinburgh. It’s a glorious time, and the library part is very eye opening. I don’t have much Internet time. It’s expensive and/or annoying to get, so I’ll have to post about the trip when I return. I will say that today was quite wonderful. We had a behind the scenes six hour tour of The British Library. Amazing, brilliant place. See you soon, figuratively.
August 5, 2011
· Sean · No Comments
Tags: international librarianship, public libraries, social justice, UK · Posted in: Studies
Early Analysis of CiteULike.org Data
The data has been collected and I’m in the process of cleaning it up for analysis. Although cleaning is going to take a while, I can reveal a few details. First, this sample contains 999 units and represents a systematic random sampling from about 2.5 million bibliographic references pulled from CiteULike.org last year. From the CiteULike.org bibliographic record, I collected data on the following variables: bibliographic type (e.g., article, book, etc.), the year the reference was posted to CiteULike by some user, and the year of publication. Then I used Google Scholar to search each reference and gathered the following information: whether Google Scholar found the reference, the citation count, whether there was a link to a full text copy, and if so, the source of the full text copy (e.g., institutional repository). I have two years of data for these variables: 2010 and 2011. Below are some descriptive statistics for the entire sample—that is, these are not broken down by document type. Since I’m still in the process of cleaning the data, some things may change. For the most part, it should be pretty accurate.
Did Google Scholar provide a link to a full text copy?
Yes: 449 (2010), 476 (2011)
No: 423 (2010), 433 (2011)
NA: 127 (2010), 90 (2011)
What are the highest ranking sources for full text access in 2010:
- psu.edu (CiteSeerX): 94
- arxiv.org: 37
- nih.gov: 36
- oxfordjournals.org: 12
- pnas.org: 11
- Other: 264
- psu.edu (CiteSeerX): 94
- nih.gov: 40
- arxiv.org: 38
- oxfordjournals.org: 13
- pnas.org: 11
- Other: 276
July 22, 2011
· Sean · No Comments
Tags: bibliographic references, citeulike, dissertation, google scholar, institutional repositories, open access, social · Posted in: Dissertation