Litmus Research Planning

Cambridge University Press & Assessment Internship | Behavioural Research Support

Overview

Litmus is a digital platform for English language practice tests, with over 3.5 million tests taken. Despite its reach, moderated research into user experience was limited. During my internship, I worked with the Litmus UX team to plan behavioural research aimed at uncovering usability issues and informing future design improvements.

The Challenge

Litmus lacked behavioural data and contextual insight into user influences. The team needed to clarify behavioural goals, identify pain points, and establish a research foundation for future design. My focus was on clarifying target behaviours, applying behavioural frameworks, and help prepare for research that could drive effective design improvements, especially addressing unfinished tests and underused feedback.

(Image displaying Litmus ongoing agile design cycle, with the stages of the testing process for applying behavioural insights in EdTech (Nesta, 2020) mapped onto it. My contribution is shown at the ‘Target’ stage, where the process begins.).

My Role

As an experience design intern with a background in design for behavioural change, I was invited to supported the Litmus team in planning there research. My role involved introducing behavioural frameworks like COM-B, facilitating work to define target behaviours and mind-sets for students and teachers, and collaborating closely with the lead UX Designer of the Litmus team, to help guide research planning. Ensuring the team was equipped to continue behavioural analysis beyond my internship.

The Process

1. Defining Target Behaviours and Mind-sets

We began by reviewing research gaps and assumptions, focusing on the team’s use of the TESTS process for behavioural change in educational design, I entered at the “Target” stag.

The first step was to clearly define the behaviours we wanted to change. I recommended identifying both problematic and target behaviours early to sharpen research focus amid limited data. Additionally, I suggested separating behaviours from mind-sets to enhance clarity and improve future research direction.

We defined the key behaviour of:

Key Student Problematic Behaviou

Users starting a test but not completing it.
Not accessing feedback via review modes/full test cycle.

Key Student Target Behaviours:

Students start and complete test.
Access and engaging with feedback as part of full test cycle.

(Screenshot of board showing collaborative brainstorm of target and problematic behaviours and mind-sets of students with highlighted examples)

2. Applying Behavioural Frameworks

As the TEST process did not include its own behavioural change framework I introduced the Com-B framework (Michie, van Stralen, and West, 2011) at the heart of enquiry, as a simple but holistic framework for behavioural change.

We worked with the COM-B model to generate behavioural assumptions, in line with the categories of Capability, Opportunity, and Motivation. To start considering what would influence or block the target behaviour.

(Screenshot of COM-B model brainstorming of attributes surrounding changing student behaviour, with images of models and enlarged examples)

Having trouble considering motivation, I used secondary research to introduce a model combining theories around testing motivation to deepen understanding of low-stakes test engagement.

3. Planning for Research Execution

In our final session, I worked to ensure the Litmus team could engage with behavioural change in their research productively:

Using behavioural thinking in considering interview questions for user interviews.
Discussing structuring affinity diagramming to reflect COM-B components, in answer to the assumption driven mapping of behaviours to refine behavioural goals and mind-sets.
Ensuring the research plan was robust, and aligned with behavioural change principles.
Presented the Behavioural Design Wheel for future intervention design.

(Screen shot of ‘Action model and key concepts of central motivation learning theories’ (Helm and Warwas, 2018), with my annotations for presentation)

The Outcome

By the end of my internship, the Litmus team had:

A clearly defined set of target behaviours and mind-sets.
A behavioural framework (COM-B) embedded into their research planning.
A structured approach to generating and analysing user insights.
Tools and models to guide future design decisions and interventions.
Greater confidence in applying behavioural change theory to UX research.

Reflection

This project demonstrated how behavioural change design can strengthen UX research, especially in educational contexts. I helped the team move from broad assumptions to structured inquiry, to enable more targeted and impactful research. Though not present for the data collection, my contribution ensured the team had the tools and clarity to continue confidently.