Data 2 Homework
This is a collaborative exercise to be carried out by
teams. The teams are of your own choosing.
Obtain a small dataset of text to analyze. There are
two basic types of datasets you could use for this assignment:
- Text corpus. This is a free-form, unstructured
collection of text that is meant to be analyzed as a whole. Examples
include a novel, an interview transcript, a brochure, etc. It can
consist of many separate texts, but in the analysis you will not
really distinguish the source -- it will be as if it is all one
thing. For example, a set of urban legends or myths. To analyze
corpus data you should follow the example of grounded theory.
Ideally, the product of your work should be an Atlas/ti database. Or
you can just put together a word document with some codes, notes and
conclusions
- Text records. The data consist of a set of
comparable records, each one containing a small bit of text. For
example, a set of personal ads, obituaries, responses to an openended survey question, etc. To analyze these data you can either
use grounded theory methods (see above) or (better) content analysis
techniques. If you use content analysis, you will be coding the
data, and creating an excel file with the coded data. The rows will
be cases (records) and the columns will be codes (typically 1/0
variables).
Please post the data to the class wiki site:
Reference:
|
