A short introduction

winMAX 97, like Atlas.ti, also has its roots in Berlin, Germany. One is tempted to ask whether it is the famous Berliner air that makes people develop QDA software…. who knows. The developer of winMAX 97 is Udo Kuckartz. He is associated with the Free University of Berlin. Those of you who are interested in some of the Udo Kuckartz's theoretical ideas about qualitative data analysis more generally and about winMAX in particular, click on the next link to see a list of references.

Program features

All features marked with an asterisk (*) are only available in the professional version.

UP

Coding

The smallest unit of text you can code is one word. Text is selected by clicking with the left mouse button on the first word of the text passage you wish to code and with the right mouse button on the last word of that passage. All code words that are created either during the coding process or beforehand, are represented in form of a sideways tree hierarchy. You probably are familiar with this kind of a tree structure from the Windows 95 Explorer.

There is no limit with regard to the number of code words you can use or with regard to the number of nests and overlaps that result from the coding process. The code word tree can be up to 10 levels deep. Modifications of the coding scheme is possible by copying or moving all coded text at a code word to another code word. The code words of individual coded text passages can be deleted directly (see below). The tree hierarchy however is not as flexible. Code words cannot simply be moved to other places in the tree. It can be done though - by using the above explained features of copying and moving coded text segments around.

In the left margin of the displayed text, winMAX offers you three types of information. After you write a memo, you will see a little yellow post-it like note in the very left column. In the middle column of the left margin you see the codes displayed against the text, and the third column is for displaying line numbers. The display of these three columns is optional. You can activate or de-activate them.

The option of seeing the codes displayed against the text offers you a quick overview of where you have done some coding already and allows you to see where coded text segments are nested or overlapped. Thus, it gives you a kind of topographical map of your coding scheme. The coded segments are referenced by a straight line with a little square box in the middle. If you move the mouse to this box, the code word for this segment will be displayed. If you hold this position and click on the left mouse button, you have the option of deleting the code word for this text passage.

UP

Memos*

You can attach an up to 15 page long memo to each line of text. As mentioned above, memos are displayed as little post-it-notes in the left margin next to the text. One way of reviewing and editing them is by clicking on this symbol. Each memo is date and time stamped. In addition, in the memo entry window there are extra fields for adding the name of the author and for attaching as many code words as you like to the memo. This information can be used to filter, search and sort the memos. Other criteria for sorting, searching and filtering are the activated texts, the activated code words and search strings based on the written text of the memo. The memo text can be copied and pasted in any word processing software and vice versa.

UP

Text Search and Auto Coding

winMAX 97 offers you a number of option to search the text. A total of 5 search strings can be used and it is also possible to logically combine them with AND and OR operators. You can search for the exact word or only the stem of a word, within the text or within the code words. The results are either shown directly in the text, in form of a list or are saved to a data file. If you save the results as a data file, you have the option of displaying the keyword within the context. You define how much context will be shown when you formulate the search. The options are to either choose one paragraph or a particular number of lines as context. This is also true for the auto coding function.

For the purpose of auto coding, it is necessary to generate the output as list. The list displays all the hits, including the project and data file name plus the line number of the location where the keyword was found. Now it is possible to auto code all of the hits or to go through the list one by one to check on all the hits before a decision is made to code them or not. If one decides to code all the hits in one go, the next step is to choose a code word. Then the hits plus the previously defined context is coded automatically. If one decides to first check on every hit, the previously define context is of no relevance. Each time one decides to code a hit, one can choose the amount of context that should be coded as well. In addition, every hit can be coded with a different code word.

UP

Search Engine

A single code search can be performed very quickly. You only have to activate (or deactivate) the documents and code words you are interested in. It is also possible to conduct some more elaborate and complex search operations. Besides boolean operators, winMAX 97 offers you 10 different types of logical combinations like intersection, at least, just one, less than, if inside, if outside, near and followed by operators*. In addition, you can use variables as filters in a search.

The results of a search can be saved under a new code word. This allows you to question your data further by performing another search on your results. Thus, winMAX 97 also offers you a "system closure" feature.

UP

Variables, Importing and Exporting Tables

The possibility of importing and exporting tables allows you to combine qualitative and quantitative data. For every document or answer to an open ended survey question, you can enter variables like age, gender, profession, income, location, etc. These variables can be used as filters and in combination with code words in a search. This means, that you can for example look for the co-occurrence of code word A AND B in all the responses of men, age category 5, who live in Munich.

Variables in DBF format can either be imported from or exported to statistical packages like SPSS, SAS and SYSTAT*. This is a feature, which is especially relevant for the analysis of open ended questions in surveys.

UP

Other features

• Merging of two different winMAX archives*.
• When searching for co-occurrences, you can at the same time also evaluate the sub-codes of those code words used in the search*.
• You can review data file information (i.e., name of project, creation date, length of text given as number of line, number of coded segments and number of memos).
• WinMAX 97 offers a number of function that support the analysis of open ended question. This includes the above mentioned data entry window for variables*, a text pre-processor that allows you to create one long data file that contains all of the open ended questions and answers you want to analyze*, and a pre-coding function that can be used for coding the various answers and questions with pre-defined code words*.

UP

References

Kuckartz, U. (1990): New Developments in Qualitative Computing: MAX - A New Program for Linking Quality and Quantity. Bulletin de Methodologie Sociologique published by the International Association of Sociological Methodology, No. 29 1990

Kuckartz, U. (1991): Ideal Types or Empirical Types: The Case of Max Weber's Empirical Research. Bulletin de Methodologie Sociologique published by the International Association of Sociological Methodology, No. 32 1991

Kuckartz, U./Prein, G. (1995): Between Quality and Quantity. In: Kelle, U. (Ed.) : Computer-Aided Qualitative Data Analysis. Theory, Methods and Practice, London: Sage

Kuckartz, U. (1995): Case-oriented quanitification. In: Kelle, U. (Ed.) : Computer-Aided Qualitative Data Analysis. Theory, Methods and Practice, London: Sage

Kuckartz, U. (1997): A Standardized Text Interchange Format for Computer Programs for Qualitative Data Analysis. In: F.Faulbaum (Ed.), SOFTSTAT '97. Advances in Statistical Software, The 9th Conference on the Scientific Use of Statistical Software, Stuttgart, Jena, New York: Gustav Fischer Verlag, Seite 259-266

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