The final version WordStat v1.0 has now been released. A fully
functional shareware version of the program can be downloaded from the
following URL:
http://ourworld.compuserve.com/homepages/simstat/wordstat.htm
WordStat is a content analysis module specifically designed to analyze
textual information stored in SIMSTAT for Windows data files (a
comprehensive statistical program running under Win 3.1, Win 95 and Win
NT) such as responses to open-ended questions, interviews, titles,
journal articles, public speeches, electronic communication, etc.
WordStat may be used both for automatic categorization of text using a
dictionary approach and manual coding. The program can apply existing
dictionaries or categorization schema to a new text corpus. It also
facilitates the development and validation of new dictionaries. When
used in conjunction with manual coding, the program can provide
assistance for a more systematic application of coding rules, help
uncover differences in word usage between subgroups of individuals,
assist the revision of existing coding using KWIC (Key-Word-In-Context)
tables, and assess the reliability of coding by the computation of
inter-raters agreement statistics.
The numerical results of a categorization or word count may be combined
with information stored in other numeric fields, allowing the user to
compare how different subgroups differs in vocabulary, topic, or
category. They also may be used to uncover existing relationships
between specific categories or words and numeric variables such as age
of the respondent, year of publication, etc. In addition, the program
also allows comparison of word, keyword or category occurrences between
different text fields and the assessment of word co-occurrence by the
display of words by words frequency matrix.
Its close integration with SIMSTAT, a comprehensive statistical program,
facilitates further quantitative analysis on numerical results obtained
from the content analysis (ex.: factor or cluster analysis on word
frequencies, multiple regression, etc.).
Program capabilities
~~~~~~~~~~~~~~~~~~~
- Analysis of text stored in alphanumeric fields containing short
textual information (up to 255 characters) such as responses to
open-ended questions, titles, descriptions, etc. as well as on
longer texts (interviews, etc.) stored in memo fields (up to about
64K of text)
- Optional exclusion of pronouns, conjunctions, etc, by the use of
user-defined exclusion lists.
- Categorization of words using existing or user-defined dictionaries.
- Interactive development and validation of dictionaries and
categorization schema.
- Word substitution and scoring using wildcards and integer weighting.
- Frequency analysis on words, derived categories or concepts, or
user-defined codes entered manually in a text.
- Univariate word frequency analysis (word count & record occurrence).
- Cluster analysis and dendrogram of words or categories.
- Word by word co-occurrence matrix.
- Word by record data matrix.
- Bivariate comparison between any text field and any nominal or
ordinal variable (such as sex of the respondent, specific
subgroups, years of publication, etc.).
- Choice between 12 different association measures to assess the
relationship between word occurrence and nominal or ordinal
variables.
- Computation of inter-raters agreement statistics (percentage of
agreement, Kappa, Scott's pi, Krippendorf's r and R-bar, intraclass
correlation, etc.)
- Key-Words-in-Context table with ability to directly jump to
the original text in order to modify, annotate, or enter codes.
- Numerical results and KWIC tables may be saved on disk in new data
files in order to be submitted to further analyses.
- Full integration with SIMSTAT (including ability to perform more
advanced analysis such as cluster analysis, multiple regression on
word or category frequencies or record occurrences, etc.), ability
to filter records, perform transformation, recoding, etc.
SPECIAL INTRODUCTORY OFFER AND DISCOUNTS
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The WordStat module is currently available at an introductory price
of $29 (plus S&H). But please note that the program requires a copy
of SIMSTAT for Windows in order to run (from $40 to $129).
Feel free to contact me if you need further information.
-----------------------------------------------------
Normand Peladeau
Provalis Research
[log in to unmask]
http://ourworld.compuserve.com/homepages/simstat/
-----------------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|