When clients come to Corona and ask us to help them find answers to their most difficult questions, we typically take a quantitative or qualitative approach to our research. Sometimes, however, we use a combination of both methods. As you might imagine, there can be lots of value in bringing the two types of data together, and a mixed methods approach can offer a powerful resource to inform and illuminate the answers that our clients seek.
But how, exactly, do we determine when it is appropriate to use a combination approach? Although it is essential to understand a client’s goals and the research questions that need to be answered as a first step to any project, it is also important to consider the purpose of potentially combining qualitative and quantitative data. There are several reasons to choose a mixed methods approach, four of which are described below:
- Enriching data – A client could choose to enrich their understanding of a topic or issue by using qualitative work to collect information not obtained by quantitative methods, or vice versa. Let’s say, for example, we have a client who puts on a special event each year. Our client wants to have more information about the kinds of people who attend that event, in addition to understanding how attendees’ experiences might be improved. To achieve these goals, we could conduct intercept surveys during the event to obtain basic demographic information about attendees. Then, Corona could conduct follow-up interviews with select attendees to understand their experiences at the event and how they can be improved.
- Examining findings – We might generate hypotheses from quantitative work that will also be tested using a qualitative approach, or vice versa. Recently, in an effort to evaluate the ways in which school meals are served, Corona conducted in-depth interviews with school administrators. From these interviews, we generated hypotheses that we later tested in a survey with a larger pool of the same audience.
- Explaining findings – In the event that there are unanticipated findings from quantitative work, we might recommend a qualitative approach to understand these results. For instance, Corona conducted one survey with college students to understand the type of beverages they drank. After analyzing the survey we had some surprises regarding the frequency that some beverages were consumed. In follow-up focus groups we were able to understand the reasons around consuming certain beverages, including situations and activities.
- Triangulation – This approach uses qualitative data to confirm or refute results found from quantitative data, or vice versa. The main idea behind this approach is that we can be more confident with a result if different methods produce the same result. Take, for example, a client who wants to choose a new logo. We might test potential logos through a survey with the client’s target audience to see which on they prefer. Then, Corona could conduct an online bulletin board with different participants to see if people pick the same logo as their favorite. If both audiences pick the same logo, our client can find more assurance that the chosen logo will resonate with a broader audience.
Overall, the goal of designing research is to use a combination of any tools to best answer our clients’ questions, and it’s important to remember that we’re not only limited to surveys, focus groups and interviews. In a world where we are constantly gathering information from countless resources (e.g., Census data, sales data, website analytics), the potential combinations are endless and might seem a little overwhelming. That’s why we’re here to help guide our clients so that they can find the most comprehensive, defensible data to drive their decision making.