Once you’ve collected data, you need to turn the raw data into a form that is more useful for driving decision making. That means you need to analyze the data in some way. Qualitative and quantitative data are analyzed in different ways.
There are a variety of ways to analyze qualitative data, and the analysis chosen will depend on the type of data you have and the questions you seek to answer with the data. Although it might sound daunting, certain methods of analyzing qualitative data are relatively easy to learn and implement (for example, the method outlined by the National Sexual Assault Coalition Resource Sharing Project [RSP] and National Sexual Violence Resource Center [NSVRC] in 2014). Other methods require considerable time and training and would likely require the service of an outside evaluator.
One of the benefits of qualitative data is that it can essentially be quantified - that is, you can turn the descriptions into numbers. For example, you might identify the number of times people in a focus group mention that their behaviors changed as a result of your intervention. When you use rubrics or scoring tools for observational data collection, you are immediately quantifying your observations as opposed to recording them as descriptive events. This process can streamline data analysis by reducing the amount of time required and also making the process easier for those who will implement it (Curtis & Kukke, 2014).
For example, one preventionist in Texas shared that her evaluation includes collecting qualitative data that they score using a rubric with predetermined themes and buzzwords. This allows them to track for buzzwords and make quick determinations about the data based on domains of interest that are represented in the rubric. Since they are also the ones doing the scoring, they can see the rich data and use the individual comments from participants as context for the decisions they make based on the data.
When you quantify qualitative data, for example by counting the number of times an idea or concept appeared in the data, you lose some of the richness of the original data, but the resulting numbers can also be useful for telling the story of your work. If you quantify the data, consider keeping examples of the richer content (for example, compelling quotes or images) to help keep the numbers in context and support the point you are trying to make with your data.
Like qualitative data, there are a variety of ways to analyze quantitative data, and different methods are used for different kinds of quantitative data and for different kinds of insight into the data.
For many preventionists, the most used types of analyses will fall under the category of descriptive statistics. These analyses, as the name implies, describe the data. Descriptive statistics include
- Frequencies – an indication of how often something occurred
- Percentages – the percentage of an occurrence
- Means, Modes, Medians – these are all measures of central tendency
Additional insight can be gleaned from inferential statistics which can do things like compare sets of data to see if there’s a significant difference between the two. This is useful for comparing pre- and post-test data, for example. These are slightly more complicated to conduct than are descriptive statistics, but with a little training, anyone can run these using Excel or other tools.
Quality data is key and having a data cleaning strategy is an important component of your data analysis process. Strategic Prevention Solutions has put together some tips to help you get started:
Primary Prevention and Evaluation Resource Kit: Analyzing Evaluation Data (PDF, 112 pages) This resource from the Pennsylvania Coalition Against Rape offers a robust exploration of how these various data analysis options apply to primary prevention work and walks through the process for using some of them.
Data Analysis Online Learning Course (Online Course, requires free account to log in) The Data Analysis Series consists of four courses designed to show users how to enter, analyze, and report on evaluation data captured from pre/post surveys. These courses contain sample data for practice, and users can pause, review, and revisit any portion of the courses.
Curtis, M. J., & Kukké, S. (2014). Activity-based assessments: Integrating evaluation into prevention curricula. Retrieved from the Texas Association Against Sexual Assault: http://www.taasa.org/wp-content/uploads/2014/09/Activity-Based-Assessment-Toolkit-Final.pdf
National Sexual Assault Coalition Resource Sharing Project, & National Sexual Violence Resource Center. (2014). Listening to our communities: Guide on data analysis. Retrieved from http://www.nsvrc.org/sites/default/files/publications_nsvrc_guides_listening-to-our-communities_guide-for-data-analysis.pdf