Friday, April 07, 2006

bibliometrics 1

Some unusual uses of normally boring bibliometric methods:

. Create sets of terms that tag to different concepts. Count and compare.

. What are the most common verbs (nouns, pronouns, adjectives) in a work?

. What are the terms excluded from the index? Create an index of these terms.

. How does the concordance of the work at hand compare to the concordances of the works it cites?

. For a work widely regarded as a Feminist text, what is the number and type of correspondences between "male" and "female" terms in the text?

. What is the Lexile count in comparable stump speeches for competing politicians?

. What type of emotional states appear in advertisements in English? In Spanish? In Russian? How many different terms are used for similar emotional states in a given ad?


Bibliometrics can be used for a heck of a lot more than simply tracking who-cites-who in scientific journals. As a method, bibliometrics seems old fashioned in an age of computer indexing. But the fact is: most content in most media hasn't yet been indexed or treated in any way with any bibliometric methodology.

Concerned about class struggle in Charles Dickens? Do yourself a favor and don't take the weight of "Marxist literary criticism" to the texts. Don't take any school of literary criticism to any text. Instead, let the text speak for itself -- what type of terms are there? How often? What are their relationships to other terms? If there is class struggle in Dickens, let the text tell you for itself. Texts can be lessened by the weight of imposed theory, and we, as readers, can be too.

Scientometrics is but one use of a powerful method.

Librarians and Information Scientists, please, do something new with bibliometry.

No comments: