A few words to begin with

The first developmental steps of Atlas.ti can be traced back to the years 1989 - 1992. As part of a larger research project at the Technical University of Berlin (TUB), Germany, a tool was needed that would support the analysis of qualitative data. None of the existing packages seemed exactly "right" for the kind of analysis the researchers had in mind. As a solution to this situation, a prototype of Atlas.ti came into being and from there things evolved. In 1993, the first version of Atlas.ti became commercially available. The "creator" of Atlas.ti is Thomas Muhr, a psychologist and computer scientist (and not to forget - a hobby musician. Just check out track 2 of the most recent Atlas.ti Demo CD J ).

It cannot be denied that the main structure of Atlas.ti shows some German/European influence. The hermeneutic tradition for example left some obvious fingerprints like the difficult to pronounce name that was chosen for the container that holds all the important objects of an Atlas.ti project: The Hermeneutic Unit. But not only the Europeans left their fingerprints. An American born method for the analysis of qualitative data also informed the development of the program to a great extent. This is the Grounded Theory Methodology, originally developed by Glaser and Strauss in the 1960s and further advanced by Strauss and Corbin in more recent years.

The Atlas.ti Tool bench - Program Features

Atlas.ti is one of the more complex QDA programs and therefore also offers a large number of tools for the analysis of qualitative data. In the following, some of the most important tools will be described:

UP

Coding

Coding is an interactive process and the results are immediately visible on the screen. Code words are displayed in the margins and the length of the coded text passage is marked by a serpent-like line. The smallest text unit that can be coded is one character. There are no restrictions with regard to the number of levels that can be coded nor with regard to the length of a code word. It is however advisable to write longer code word descriptions or definitions as a comment. All code words in Atlas.ti are interactive. If you click on a code word once, the text passage coded with this code word will be highlighted. If you double-click on a code word, the code word's comment will be displayed. The modification of already coded text segments is an easy and quick process.

A code word list is automatically created and updated while you code your data. Behind each code word, two numbers are displayed - the first number tells you how often a code word is used, thus says something about the groundedness of a code word; the second number informs about the amount of relations that exist between the code word and other objects in your project, thus says something about the density of a code word.

UP

Text Search and Auto Coding

Auto coding means that you can search for a string of characters in the text and that the program automatically codes this string of characters with a pre-determined code word. You can choose whether you only want to code the matched string, the entire word, sentence or paragraph. If you wish, you can control this process by deciding each time a match was found whether you want to code it or not. In order to facilitate the text search, Atlas.ti offers you a number of GREP functions.

What is the purpose of the text search and auto coding function? Both functions allow you to "dive" into your data and to give you a quick feel for what's there and what to look for. This can be especially important if you have large amounts of text (and little time to analyze it). Another application is to code the responses to entire questions or other clear distinguishable parts in your data (see also Tips & Tricks).

UP

Memos and Comments

For virtually every object in Atlas.ti you can write a memo or a comment. Memos can be attached to various objects and also related to each other. It is also possible to define memo types. The program offers you a few default memo types, however you are also free to create new types that fit the contents of your memos.

UP

Query Tool

Atlas.ti offers you a wide variety of search options. Each query can be saved in form of a supercode. You are most likely to choose this option if you have complex queries. This way you can re-use them at some later point in time without having to remember it or to make a note of it somewhere on a piece of paper. It can also be seen as a kind of container that holds the results of a search and from where you have access to the coded text segments even after you left the query tool. They also can be part of a network.

UP

Visualization

A distinct feature of Atlas.ti is the way the various objects of an Atlas.ti project can be visualized. Codewords, all text, graphic and audio quotations, memos, comments and primary documents can be visualized in a network view in form of symbols. All of these objects can be linked (and unlinked) and the type of relation between them can be defined. The place of the objects in a network is not static, in other words all objects can be moved around freely without having to obey a default structure. If you like you can build a hierarchy of your coding schema, but this is not the only way you can arrange your code words or other objects in your project. This means that you can start building theories within the program itself. Since from a network view you can access your primary data directly, you can always keep in touch with the content of the coded segments and the surrounding context.

An additional option is to save the network view as a graphic file and to export it to a graphic program. There you can freely manipulate the various nodes of a network view, prepare it for presentation purposes (e.g. for a power point presentation) or for insertion into a research report, etc.

UP

Other Media - Graphic and Audio data

In addition to text, Atlas.ti also supports the analysis of graphic and audio material. Atlas.ti recognizes up to 20 different graphic file formats. Graphic material is coded in form of rectangles. The handling of coded rectangles is similar to handling coded text passages. Audio material can be handled in form of WAV data files. At the moment, it is only possible to code an entire WAV file, however development is under way that will allow for a more extensive analysis of audio material.

UP

Other features

• You have the possibility of creating Hypertext links.

• Support of team work: If you are working in a team, all team members can login as different users. The effect of this process is that all the work that is done on a project (e.g. creating code words, memos, search queries, etc.) carries the name of the current user. This way it is possible to trace the analytic steps of the various team members. If you work on different computers, you have the possibility of merging projects.
• There are a number of tools that help you organize your data and the various objects of an Atlas.ti project (e.g. the Object Explorer and the Family tool).
• You can save your project as a HTML data file. This option can be used for example as a form of publishing a project or as a way to exchange or transmit projects via the internet.
• Based on your data, you can create a SPSS syntax file. This file can be used to export your data to SPSS.

UP

References

Glaser, B. G. and Strauss, A. L. (1967). The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine.

Muhr, T. (1991). Atlas.ti - A prototype for the support of text interpretation. Qualitative Sociology, 14(4), 349-71.

Strauss, A. L. and Corbin, J. (1990). Basics of Qualitative Research, Grounded Theory Procedures and Techniques, London: Sage.

UP

Workshop Booklet

Step-by-Step Introduction to ATLAS.ti v4.2

Due to a 'lucky' mistake, I have now decided to make my workshop booklet publicly available. Even though called 'Workshop Booklet', I do not use it as such in my workshops. I distribute it to the participants, in order for them to be able to retrace the exercises of the various workshop session when back at their own desks in the office or at home. Therefore, the workshop booklet cannot replace a workshop. The focus is strictly on the most important functions of ATLAS.ti and their technical aspects; methodological questions are not discussed. It may, however, help some people to teach themselves and to gain their first experience with the software.

 DISCLAIMER: The descriptions reflect my personal experiences with the programs and should not be (mis)understood as an objective report.