Engagement Strategy using Motivational Profiles
Deliverables
Abnormal Data Trend
Upon reviewing the data for a client’s annual report, we saw unusual data trends. We wanted to find out what caused these unexpected behaviors. I designed and conducted a qualitative research that then triangulated the qualitative data with the quantitative data.
Implications
With a bimodal distribution in the dataset, it became clear there were two different groups participating in our workshops. The first typically completed the program. The second started and struggled to continue early on. All of the participants, however, had been motivated in making healthy lifestyle changes upon joining. We wondered: Were the circumstances, challenges or underlying characteristics of the first group different from the second? Were the groups identical other than strategies they used to overcome challenges? How could we tailor our support to those who abandoned early on?
Mixed Methods Research
In order to better understand the phenomenon, I recruited 6 participants for an in-depth qualitative interview. Following the deep listening method, I let the participants be my guide, and did not intervene. Based on the triangulation of the qualitative and quantitative data, I was able to identify three hypothesis, which we invalidated two thirds of. I used non-parametrics statistics to account for the small sample size. From this hypothesis, I generated a questionnaire, to help test generalization. I recruited two participants from outside our organization to help review the wording and clarify instructions, and then sent it to 84 alumni to collect data from a bigger sample. Using their results, and their outcomes within our program, we were able to identify two distinct motivational profiles. Our current approach only supported one of the two. We understood we could increase the rates of completion for participants whose preference was the other motivational type by adapting our processes to include their preference.
Since I did not know which hypothesis to test at first, I combined a narrative analysis and a themes analysis for each of the interviews. The themes analysis used grounded theory to identify key topics discussed by the participants. In the case of the narrative analysis, I used syntax within the narration to determine the presence or absence of certain conditions before and after critical events.
Testing my three hypothesis revealed one hypothesis worth investigating further. Transforming the qualitative interviews into binary data at the document level, I was then able to reconcile this with the quantitative data we had in regards to program participation.
Creating a Motivational Profile Measurement
From the various statistical works, I was able to extract three key questions which could act as an abridged questionnaire to determine the highest likelihood of a participant’s motivational style. This, in turn, allowed us to customize various aspects of the support offered to participants throughout the workshops.
Designing the Experience
Below are the mockups I made for the pilot test. These mockups show the message displayed when participants view their profiles after taking the quiz. Two colleagues crafted and reviewed the text based on the research findings I shared with them, and the development team helped create the different pathways for the pages within this pilot micro-site.
GitHub Project (Public)
Repository (Private)
My Role
Researcher
UX Designer
Project Manager