Qualitative Data Analysis
Developer: Matthew Weinstein
Freeware
Version: 3.34b4
Release Date: 2006-09-27
Last Updated: 2007-02-06
Pros: Powerful and full-featured qualitative data analysis tool; Offers analysis methods unavailable in commercial software packages; Integrated transcription tool; Exports reports in XML; Very robust; Free!
Cons: Interface non-intuitive; Requires steep learning curve.
Description
To those who do not have any experience with qualitative data analysis software or even with qualitative data analysis at all the following might seem a bit cryptic. If you would like to learn more about the topic please look at the selection of websites in the respective section of the Links page.
TAMS stands for Text Analysis Mark-Up System, describing
this software's most important, but not only purpose.
Basically what TAMSAnalyzer allows you to do is to
structure and categorise any text into chunks and pieces
using a customisable set of codes, and make observations on
the occurrence of a code or the co-occurrence of a
combination of codes throughout any number of texts,
typically interview transcripts, newspaper articles etc.
Those who have experience with computer-assisted
qualitative data analysis will recognise these features
from similar software packages such as NVivo, Nud*st,
Atlas/ti, Transana and others. Hidden behind a somehow
quirky interface, TAMSAnalyzer provides most of the
capabilities found in the commercial programs, and even
offers a number of features not found anywhere else, for
free.
At the software interface's centre is the so-called
Workbench, where one can add files (by typing or importing
texts), define codes (hierarchical and associative code
structures are supported) and conduct searches for codes.
Each file can be opened and edited in a separate window
where also the coding is taking place. Here one can use the
implemented and very functional transcription tool in order
to listen or watch to audio and video files while typing
and editing the text. The coding itself is done by using
html-styled tags representing the codes or one of a number
of pre-defined commands (e.g. telling the software to
structure the text into sections). This can be quite simple
and straightforward, if one wants to leave it at simply
coding the text with a fixed number of codes, or it can
become more complex by using commands or so-called context
codes, the values of which can be seen universally in any
analysis report (A typical context code is "Speaker",
indicating the name of the person "speaking" at a certain
point in an interview transcript. A search for a certain
code would return the results together with the names of
the respective speakers.)
The option to either keep things very simple or to make
them really complex is characteristic of TAMSAnalyzer, and
also true for its analysis capabilities. You can choose to
just conduct a basic search for a combination of codes (the
equivalent to, for example, "show me all instances in all
texts, where I coded a text 'personal lifestory' and
'industrial changes' at the same time") or to compile a
proper data summary report based on the search's findings,
in which you can let TAMSAanalyzer group the results in
categories (for example "Speaker"). Other reports available
are: co-coding frequency, code count, word count,
interrater-reliablity, data comparison table, co-occurence
table, graph data and graph code sets. Each report offers a
multitude of options, which can be as useful as it can be
confusing. At the end the reports can be saved in XML
format and edited in other applications, for example a word
processor.
It would be impossible to describe all of TAMSAnalyzer's
features and some of them have yet remained unknown or
unused even to myself, for example the possibility to
publish a TAMSAnalyzer database as a SQL-database, thus
allowing multiple researcher to work on it simultaneously.
Partly this is also due to TAMSAnalyzer's homegrown user
interface, which does let you do simple things quickly but
at the same time stands in the way of some of the
software's more powerful features. It needs to be said,
however, that TAMSAnalyzer is free and the baby of a very
dedicated social researcher, Matthew Weinstein, whose
expertise in qualitative research clearly and very
understandably just outdoes his skills at user interface
programming. Nevertheless the software is extremely stable
and with the help of the comprehensive manual does the job
very efficiently. Great work!