USING VISUAL ANALYTICS
TO SUPPORT STUDENT LEARNING
Sujoy Chaudhuri and Shabbi Luthra
“ I tell you and you forget. I show you and you remember. I involve you and you understand.”
Confucius (500 BC)
Since February 2014 we have been examining the vast quantities of data at ASB using interactive
visualizations. Why are we doing this? Because we believed Edward Demming when he said, “If you do
not know how to ask the right question, you discover nothing”. We felt that there was a need to develop
visual interfaces that encouraged questioning and conversations around data rather than making
judgments and reaching conclusions when looking at data. In this chapter we describe one of the many
applications we have developed - an application focused on visualizing NWEA MAP data.
We aggregated 16,000 individual NWEA MAP test results conducted at ASB between Spring 2009 and
Fall 2014, collecting results across the Mathematics, Reading and Language Usage disciplines.
While the NWEA calculates and distributes projected/expected proficiencies for each student based on
his/her RIT score, we chose instead to build our own model of student achievement using a linear mixed
model that accounts for grade, age (in months) of the test taker, his/her gender, whether or not the
individual receives instruction in English as a foreign language, and random student level effects. We did
this because we believe that the student community at ASB (and indeed in other International Schools)
is substantially different than NWEA’s normed groups.
The methods and results of our data wrangling and model selection process are perhaps too academic
to describe here. However, here is a summary of our findings:
we found significant gender effects in achievement on math vs reading and language skills. Boys
are expected to score higher than girls in math, but receive lower test scores in both reading
and language usage.
EAL students are expected to do worse than their peers by as much as 10 RIT points in reading
and language usage tests, but we found no significant effect for EAL status on math