Program Evaluation Toolkit for Harm Reduction Organizations
Qualitative data analysis is different from quantitative analysis, as qualitative analysis is not an objective exercise. The purpose of a qualitative analysis is to assess what the written or narrative data is telling you about the program and organize that into themed categories that will help you draw conclusions.
The qualitative data analysis process typically follows the following four steps:
Become familiar with the data
Review evaluation questions
Look for important concepts
Establish key themes
Step 1: Become familiar with the data
Once all the data has been transcribed, it is helpful to first read through the data a couple of times to become acquainted with it. As you read, jot down basic observations and gut reactions that emerge so that you can reference these down the line.
Step 2: Review evaluation questions
After you read through the text, review the evaluation questions to determine which of them will be able to be answered using the qualitative data you have collected. This is a crucial step, as it determines which sets of data will be used in the evaluation and which sets will not. Once you have identified the data that will be used for your evaluation, it is helpful to pull it into a separate document that organizes the text by the evaluation question it helps to answer.
Step 3: Look for important concepts
Once the qualitative data has been pulled and organized, the next step is to code the data based on the patterns that emerge. By reading through the data, your team can create a list of common terms and themes. The patterns can either be similar concepts, phrases, or beliefs, or they can be similar types of responses to questions (e.g., negative responses vs. positive responses). As with quantitative data, coding the data based on patterns helps narrow down a large set of narrative into helpful, more manageable categories. A helpful tool for identifying patterns within qualitative data sets is Voyant. With Voyant, you can use the platform to identify phrases or words that are relevant for your evaluation and search for them in the text. For example, you can identify key phrases such as “syringe van” or “outreach hours” and Voyant will identify them in the text along with the other vocabulary and phrases used in large qualitative data sets. If you set “hours” as a key term, Voyant will help you view other terms which appear in the same sentences or paragraph, such as “closed”, “open”, and/or “later.” This allows you to explore large volumes of qualitative data and see how many people referred to “hours” in relation to “closed” and “later.”
Step 4. Establish key themes
The two types of qualitative analysis that are commonly used for program evaluation are narrative analysis and thematic analysis.
| NARRATIVE ANALYSIS | THEMATIC ANALYSIS |
|---|---|
| This type of analysis focuses on analyzing respondents' experiences and motivations by looking closely at the individual stories that they share and interpreting their meaning. | This type of analysis focuses on using the patterns identified in step three to determine and compare overcharging themes across the qualitative data sets to tell a larger, overarching story. |
When engaging in your qualitative analysis process, it is helpful to keep the following in mind:
Here are some resources on conducting a qualitative analysis:
