Lang and Baehr, Data Mining

Lang, Susan and Craig Baehr. “Data Mining: A Hybrid Methodology for Complex and Dynamic Research.” College Composition and Communication 64.1 (September 2012): 172-194.

Lang and Baehr argue that data mining is a useful research methodology for researchers and administrators in composition and rhetoric because of its inductive nature and its ability to organize and use large sets of data.  Their article defines data mining, explains how current computer technologies make data mining an efficient and useful research tool, describes the process of data mining, gives an example of it in practice (from their work at Texas Tech), and names the limitation of the methodology.  They offer data mining as a tool for researchers to engage in a RAD research agenda, as called for by Richard Haswell and Chris Anson.  They believe that in this age of increased demand for accountability, data mining can help teachers and administrators develop better assessment techniques and argue for their programs.

Notable Notes

data mining allows for categorization, clustering, and the emergence of associations and patterns (178-179).

distinction: data mining is more inductive – the data comes first (not the hypothesis), and the findings emerge (179).

application of data mining to Chris Anson’s taxonomy of six types of research (research categories) (181-184).

example: why do students earn DFW in first-year writing? What are the factors? Data mining study at Texas Tech

limitations: the complexity and scope of the data; longitudinal studies are necessary to increase validity; it cannot completely substitute for other kinds of research methodology; quantitative methods aren’t as accepted in the field (190-191).

data mining process: (185-186)

1. identify the problem(s)
2. select raw source of data
3. decide what measures or criteria to apply to the data
4. develop a formal procedure (a repeatable process) for sifting through the data
5. interpret the results

Quotable Quotes

“Data mining is the iterative process of systematically interpreting, organizing, and making meaning from data sources” (191).

“The increasingly accoutnability-focused climate of higher education demands that we at least begin to explore the use of data-mining technologies” (184).

“Data and text mining extend these activities beyond what is possible for us to do as individuals without the assistance of computer technology, as large amounts of numeric or textual data can be examined for various types of relationships, including classes, clusters, associations, and patterns” (178).