LINGUIST List 15.1243

Sun Apr 18 2004

Review: Software: QDA Miner

Editor for this issue: Naomi Ogasawara <naomilinguistlist.org>

Directory

1. Elmar Henning, review of QDA Miner

QDA Miner, Provalis Research.

	
Elmar Henning North-West University South Africa
	
[This is a review of the software package announced in
http://linguistlist.org/issues/15/15-911.html
which complements the previously posted review of WordStat in
http://linguistlist.org/issues/15/15-1171.html 
					 --Eds.]
	
QDA Miner is an analysis tool for analysing textual data from a
qualitative perspective with a number of useful features for
linguistics research such as tagging or encoding texts, performing
annotation, and performing assorted retrievals according to set
specifications. The QDA Miner incorporates a large selection of tools
to find patterns in texts between so called metadata or inserted codes
or annotations. A wide range of text formats are supported anything
from plain text files to dBase files. When combined with sister
applications WordStat and SimStat it provides integration of
qualitative and quantitative content analysis methods.
	
During my review of the QDA Miner, I was very impressed by the vast
number of possible text project uses for this application. QDA Miner
is one of, if not the best multi-use text application that I have ever
worked with. It is relatively easy to use, but at the same time one
has to know what one is doing to get sensible and above all useful
results out of it. During my review, I worked through the Demo project
included in the evaluation edition just to find my feet and to get an
idea as to what the QDA Miner can do. As I come from a database
background and work alongside linguists and their research data on a
daily basis, I tested the QDA Miner by creating a dummy corpus project
by using the data that had been gathered for use in one of the Corpus
Linguistic Research projects running at the North-West University. I
found that the QDA Miner was most useful indeed. And indeed provides
some unique and very useful functions that prove very useful for
linguistic research. The extreme flexibility of the QDA Miner makes it
a contender for consideration for a virtually limitless range of text
projects.
	
The analysis features of the QDA Miner are very useful for text
analysis. In the Dummy Corpus project for instance, I used the
variables as metadata definitions to run searches through imported
data with, I must add, surprising ease. The coding of the texts is
equally simple, fist you define you code and then drag and drop it in
the appropriate place. There are no forced codes and one is left to
one's own devices. I especially like the double click feature for code
insertion and the colour coding of each defined code, these colours
makes the distinguishing between the various codes easy. Another great
feature of the QDA Miner is its ability to import virtually any type
of file under the import document feature. Although one has extensive
freedom with variables, codes and potential imports, the analysis
features of the QDA Miner makes analyzing texts in various ways very
easy.
	
However there are some less than pleasing, what can be called niggles
present in the QDA Miner. Its date format is weird for one, and the
limit on the variable length can be bothersome. But this doesn't
detract from the usefulness of the QDA Miner. One learns to give short
but meaningful names to variables, and one does eventually get used to
the weird date format. The nice thing about the QDA Miner from my
point of view is the fact that one has virtually unlimited freedom
when defining variables, as there are no fixed or required
variables. QDA Miner Provides great coverage for general text
analysis, but unfortunately is a bit lacking in some areas. For
example: the when coding a text document it would be nice to be able
to annotate individual words in the text itself not only alongside the
text. When one performs word class tagging or annotation the display
with the coding information becomes rather clustered. But then again,
not everyone is going to use the QDA Miner for word class tagging.
	
The QDA Miner is extremely useful due to the fact that it incorporates
some functions and features that are rather unique. It removes the
necessity to use multiple research tools because it includes most of
the best and most frequently used tools such as Metadata searches,
encoding functions etc. When it comes to usability, the QDA Miner is
second to none. There are multiple ways to access the same functions,
multiple access points for each component such as cases, codes,
variables etc. Each case, code or variable can be accessed via a menu
that wraps only the functionality attached to the specific function
under scrutiny. In short the QDA Miner caters for both beginners and
advanced users. The basic interface and work environment is easy to
understand and presents all the basic functions with easy
access. There is also a fairly comprehensive help function and the
Demo project is invaluable to the first time user. This coupled with
the PDF manual and you are ready to go out and use the QDA Miner to
its fullest extent.
	
The QDA Miner provides a good coverage of the most popular and
required functions in a text analysis tool. Provision has been made
for sister applications like WordStat and Simstat plug-ins to increase
coverage of more specialized functions. A feature that is lacking in
my opinion is the ability to physically code the text in the text
window, at the moment the text can only coded in the code window
alongside the text window. But that apart, the tool is of high quality
and is overall highly functional and will become an invaluable text
analysis tool in future. There are unique menus for each component to
modify option unique to that specific function and thus giving the
user the option of customizing each component in the desired
fashion. By giving each component its own menu and shortcuts, the
applications become easy to use and understand.
	
One the whole, I am very impressed with the functionality provided by
the application, it has a wide variety functions to cater for a wide
range of text implementations. To tell the truth, the QDA Miner can be
used for virtually any text driven project to perform a wide range of
analyses.
	
ABOUT THE REVIEWER

Elmar Henning is currently the manager of the Language Technology
Laboratory at the North-West University, South Africa. He is currently
working on his Masters degree in the combination fields of Computer
Science and Corpus Research, where he is currently involved in the
development of custom made research tools for Corpus research material
management, data retrieval and annotation.