With its full-featured and text coding capability, The Concept System Core Program can be used as an analytic tool for qualitative analyses. In fact, The Concept System Core Program is the only current software product that enables group-based or team-based qualitative analysis. This means you can conduct a truly ‘participative’ qualitative study, with interviewees or focus group participants becoming an active part of the coding and analysis team. And, The Concept System Core Program is the only qualitative analysis tool that enables you to map your results in the form of a concept map and produce pattern matches that show the degree of consensus among coders.

Qualitative analysis is a large methodological field, and no single tool can possibly accomplish all possible qualitative analyses. The Concept System Core Program is especially useful for conducting thematic analyses of text data. For instance, you can use The Concept System Core Program to analyze interview transcripts, focus group transcripts, or memos and reports.

The basic steps involved in doing a qualitative analysis using The Concept System Core Program are:

1. Set up a Concept System project.

You will need to set up a separate project for each unique set of statements you want to map. In some qualitative analyses, you will do all of the analyses on a single set of statements and therefore only need one project. In others, you will want to do a separate analysis for each major research question, generating a different set of statements each time.

2. Transcribe your data into text form.

Most qualitative research is based on text. Even in studies that are observational in nature, the observer usually eventually records or transcribes their observations into text form.

You can use the Concept System to enter or transcribe your text, or you can use virtually any other Windows-based . If you use The Concept System Core Program , you will have to first login to your project. You may have multiple team members working on coding for the same project. As in all concept mapping projects, you will have to create new users each having a UserName and Password, and they will each need to be given Administrator status in order to use the . multiple coders cannot be working on the same text file at the same time. But, different coders could work on different files at the same time or on the same file at different times.

If you use another to transcribe your data, you must save the file either as ASCII text (with a .txt suffix) or in Rich Text Format (with a .rtf suffix).

You may put all of your text into a single document or, if your project is a large or complex one, you may enter your text into multiple documents. For instance, you might have a separate document for each person interviewed, for each focus group conducted on a single topic, or for each organization observed.

3. Code the text data into statements for analysis.

For this step you must use the Concept System . In the , you use the Code Menu to mark statements in your text document(s). If you have multiple source documents, you can code each one separately. Just remember that you are limited to a total of 200 statements in a single project (i.e., analysis).

The process of coding statements for export into a mapping project is sometimes called ‘text abstraction’ because you are abstracting small text entities from a larger text document.

4. Export the Coded Statements into your Project.

This is simply accomplished with the Export Command. If you have multiple coded documents, you can open them in succession and export the coded statements from each, selecting the ‘Append’ option each time you export.

5. Use The Concept System Core Program to Analyze the Results.

At this point, you have a set of statements that you have abstracted from one or more qualitative source documents. The statements have been exported into a concept mapping project. They can be treated like any set of statements that we wish to concept map. You might wish to review the general steps in a concept mapping project to see what you should do next. Here’s a standard set of steps you might follow:

a. Create Users. If you are going to be the only one classifying the statements into themes, you will have to have a do at least five complete sorts of the statements. Since each user is limited to only a single sort, you will have to define multiple ‘dummy’ users for each sort you wish to do and assign each of those users to do a sort. If you will be working with multiple coders or analysts, you will have to define a user for each person each time they wish to sort the statements.

b. Create Demographic Variables. If you think you will want to do separate subanalyses for different subgroups of coders or analysts, you should define one or more demographic variables to classify the users. This will enable you to easily compute separate maps for different user subgroups, compare user groups on their ratings, and conduct pattern matches of different groups. In a stakeholder-based qualitative analysis (where you might have interviewees or focus group participants help conduct the analysis, for instance), you will almost certainly want to have demographic variables defined.

c. Define Rating Variables. The Concept System Core Program allows you to rate each abstracted statement on one or more variables of interest. For instance, you might want each statement rated for its relative importance. Or, in a stakeholder-based analysis, you might have each rater state how relevant the statement is to their experience. Please note: you will not be able to do pattern matching unless you have one or more ratings.

d. Assign Sorts and Ratings. Once you have set up your data classifying plan, you will need to assign the defined users to do a sort and/or ratings as desired.

e. Edit and Lock the Statement Set. Before you perform the analyses, you should look over the statement set one more time to make sure all of the ideas you want are included and to proof-read the statements. Once you have the statement set edited, don’t forget to lock the statements.

f. Sort and Rate the Statements. The heart of the analysis is the sorting or classifying of the statements into categories of similar ones. This is directly analogous to a thematic classification in qualitative analysis. You may sort the statements as many times as you wish by creating a new user for each sort. You must have a minimum of five completed sorts before you can compute a map. In many analyses, you will also want to complete one or more ratings of the statement set. Or, you may want various stakeholder groups to rate the statements. Each user can be assigned to do a single sort and any or all of the defined ratings.

g. Compute the Maps. This is probably the easiest step! You need to have at least five completed sorts before you can compute a map.

h. Interpret the Maps. This is the process of selecting the number of clusters you want in the maps, assigning cluster names (or thematic labels), and exploring the meaning of the various maps you can produce.

I. Pattern Matching. If you have any ratings, you will be able to do a pattern matching analysis. You may want to do a consensus pattern match to examine the degree of agreement between two coders or two stakeholder groups on a rating variable. Or, you might want to do a consistency pattern match for one or more individual coders as a way of examining the repeated measures reliability of their codings done at two different times. Or, you might want to do an outcome pattern match that examines the relationship between an expectation rating and an observed outcome rating or measure (collected at a later date).