Content Frame
Skip Breadcrumb Navigation
Home  arrow Chapter 8  arrow Qualitative Data Analysis

Qualitative Data Analysis

This module will illustrate how qualitative data is analyzed. A common form of qualitative data analyzed by action researchers is interview data. The following example is an annotated interview between a researcher and a bilingual education teacher and will be used to illustrate step-by-step how the action researcher analyzes and interprets the interview data.

The qualitative process of data analysis is an inductive one, in which the data is examined from a "bottom-up" approach (Creswell, 2005). The specific data is examined to identify more general themes that will be used to understand the meaning of the data.

The example used for this module is the Bilingual Education Teacher Interview found on page 127 of the Mills text.

Step 1: The first step is to transcribe the interview.
Step 2: Preliminary exploratory analysis.
Step 3: Making connections to the research questions.
Step 4: Inter-rater reliability.
Step 5: Interpret findings.

Step 1: The first step is to transcribe the interview.
Most interviews are tape recorded to give the teacher researcher accurate recordings of the interview data. Transcribing is time consuming but serves two purposes in the data analysis process. First, it allows the interview data to be formatted into a usable form. Second, the process of transcribing lets the teacher researcher hear the data repeatedly as it is being transcribed. In this process the researcher becomes more familiar with the data and common themes may begin emerging at this stage.

For tips on transcribing interviews, the following Web site may be useful. This site also has useful suggestions for conducting and tape recording an interview.

To illustrate the steps in this module, the interview of a bilingual teacher has been transcribed. The transcript shows the researcher's questions proceeded by "Q" and the bilingual teacher's responses proceeded by "A."

Bilingual Education Teacher Interview

Step 2: Preliminary exploratory analysis.
During Step 2 of data analysis the teacher researcher is exploring the data in order to become familiar with the interview information. This entails reading the transcript multiple times. From this initial review of the transcript, the action researcher begins to see themes emerging from the data. Sections of the transcripts that reflect a theme are identified. Notations are made to record ideas that the action researcher identifies while reading the data. In the example, you will see notations that are made during this initial stage of analysis.

Step 2 Example

Step 3: Making connections to the research questions.
Step 3 involves describing and further developing the themes from the data to answer the major research questions. The themes identified in Step 2 are revisited with the major research questions as the lens for analysis. For this example a major research question was, "What are the major perceptual barriers to bilingual programs in public schools?" The themes become more refined in order to address this question. The original theme "fear" is now broken into subcategories that better address the question of perceptual barriers to bilingual programs. The new subcategories of what fear entails include: job security (fear of losing one's job), change (fear of changing how things of currently done), and disconnection (fear of losing a common identity, "American").

Create a coding scheme and coding data. The next step is to create a coding scheme that best defines the themes that have been identified and provide a way to break up the data for further analysis. Below is an example of a coding scheme used to analyze this interview data. The codes are then used to identify the specific sections of the interview data that represented the category. For example the code 1,1 was used to identify the category "Fear in terms of job security."

New Page 1




1. Fear

1. Job Security

2. Change - Ethnocentric

3. Lack of knowledge of bilingual programs


The transcript is re-read with these codes in mind and sections of the data are bracketed and coded. Later when these sections of the data are looked at within themes or categories, it is useful to examine the data within "blocks" of text so that the data can be read in context versus only including the sentence(s) that specifically describes the data of interest. Some researchers will use color codes to identify these "blocks" of text to provide a visual cue of the different ideas/themes represented in the data. Colored highlighters are useful for distinguishing the different themes embedded in the interview data.

Step 3 Example

Step 4: Inter-rater reliability.
To ensure reliability of the coding scheme, it is valuable to have another's perspective. You may want to ask others involved at your school to assist you, or if you are working as a research team you will have members of your team review the data. Each person will review the transcript and use the above coding scheme to code the data. Results are then shared and any discrepancies are discussed and resolved. Changes in the coding scheme may include additions, deletions, and clarifications.

Step 5: Interpret findings.
Once all of your interview data have been coded the data is then divided into themes. This can be done by cutting up the interview data "blocks" and sorting them into each of the codes. Pasting the data onto index cards may assist the sorting process. You will have to make multiple copies of the transcripts as data may be placed into more than one category. Important note: Always keep an original in its entirety.

The data is then reviewed within the themes or categories, and an understanding of each theme is reached. Quotes may be selected that best illustrate the meaning of the category; this provides a "voice" to people interviewed when describing the data. Notes and comments can be written on the index cards to denote the researcher's ideas while examining the data. For example, the bilingual teacher interview data that was coded under fear of job security (1,1) will be grouped with other interview data similarly coded to better define this theme or category.

This process of qualitative analysis will be repeated with each type of qualitative data that is being collected. For example, field notes of classroom observations and school documents can be examined and sorted using the same process. Comparison among the multiple data sources will serve to validate the data interpretation through triangulation (see chapter 4 for issues of validity).

Pearson Copyright © 1995 - 2010 Pearson Education . All rights reserved. Pearson Prentice Hall is an imprint of Pearson .
Legal Notice | Privacy Policy | Permissions

Return to the Top of this Page