How we redesigned our rating scale: From a behavioural science based UX research perspective
We’ll elaborate how we used research from behavioural sciences to revamp the rating system in BYJU’S Classes.
BYJU’S Classes offers two-teacher synchronous online tuition for students ranging from grade 4 to 10, primarily based in India.
In our current rating system, where students after attending an online class, rate the class experience and the tutor’s performance based on various parameters. But there were certain problems that we faced here:
- The ratings were not generating proper actionable insights
- There were significant drop offs when it came to rating page
Therefore, we took up this project to improve the rating system and fetch optimal ratings from the students.
Objective of the Project
- To find which rating system is better for each cohort of students to be able to engage them better
- To increase the number of ratings collected, and reduce drop- off rates
We identified few key problems among others in our existing rating system:
- Same parameters all across the rating scale, irrespective of good or bad ratings
- Ambiguity in the usage of emoji based and star rating simultaneously
- Reading through as many as ten parameters may lead to cognitive load for children.
Our research process
Instead of directly jumping into ideation, we, a team of designers and researchers took a step back and tried to understand the rationale behind the rating system.
We undertook a three step approach for this: Primary, secondary research, and competitor analysis among other methods.
We attempted to ask the following questions at the beginning of research:
- How do people rate?
- What are the popular systems of rating?
- Do children rate differently from adults?
- What variables are we trying to measure?
- What biases are involved in the rating?
- Star Rating VS Emoji Rating System
- The optimal spread of the rating scale
Findings from our target audience
- Children can accurately perceive emoticons to assess their own feelings
- A 3 point facial scale is preferred compared to a 5-star rating system
- Kids may use emojis in more unstructured form than adults
- Children in the age group 8–9 can reliably answer questionnaires
- Rather than make individual judgments, people would always tend to conform to a group.
- User’s personality, mood, environment, the urgency of the requirement, eventual gratification, and the influencers, with their proximity to the user, all of these factors weigh in
- A good of people 74% prefer to rate on mobiles over other devices.
Alongside, we also broke down the various popular as per their advantages and disadvantages, and tried to the best system of rating suited for our purpose.
Understanding different systems of rating
Binary System 👍 👎
3 Point Scale 🙂 😐 🙁
Intensity Scales(Like Claps 👏 in Medium)
4 Point Scale
Conclusion
After going through all the insights and keeping in mind the characteristics of our target users, we shortlisted Binary and 3 point system of rating.
We decided to conduct A/B test (evaluative research) with both these systems with our real users across different age groups, and see which works best and provides effective results.
Since every feature is context specific, what may work for one may not work for others.
So, definitely the final take away would be test out and experiment with the recommended systems based on the target users.
As we are aware, even large companies have consistently thrived upon recommendations and testing in order to implement any lasting change.