50 Future Forwards: Exploring Frontiers in Education
An additional set of bar graphs present the frequency distribution of raw RIT scores and individual
growth across the entire grade since the beginning of the academic year. The graphs may be brushed to
show students scoring/growing within the user selected range in order to answer questions like:
Who grew most//least over last year? Were they in my top or bottom group? Can I change
strategies for next year?
The search box may be used to find and display a single student. Note that all visualizations in each card
respond to selection/filtering actions performed on any one student - such that multiple questions may
be seen at once. Using the search box will highlight the student in each radial plot, showing how the
individual has performed, how far above/below expectations they are, and what their growth
trajectories look like over the course of their tenures at the school.
We also developed an interface to help with differentiation, again with the intention of reducing the
number of clicks necessary to view a report. Using a force layout, we plot all students in a grade by
subject level as blue nodes, sized by their RIT score (stronger students are sized larger than weaker
ones). Grey circles are also drawn around each student to depict the difference between their actual
and expected scores. Links are drawn in a light grey between students and each goal, the length being
determined by the inverse of the RIT score attained on tests of that goal.
A force layout uses Verlet integration to allow simple constraints. Spring-like attractive forces based on
Hooke’s Law are used to attract student-goal pairs of end-points of the graph towards each other, while
simultaneously, repulsive forces like those of electrically charged particles based on Coulomb’s Law are
used to separate all pairs of nodes (see http://en.wikipedia.org/wiki/Force-directed_graph_drawing for
more information). We chose in this case, to fix the goal nodes to positions around a unit circle in the
centre of the drawing area.
As shown in Figure 6 below, students clustering closer to the center of the visualization scored lower on
the MAP test. Focus however, on student CW, located towards the top right of the figure. She is clearly
performing well above expectations (blue circle is larger than the grey one), but the closer proximity to
Data Analysis and Probability tells the same story as Figure 4 above (indeed, the two figures highlight
the same student).
A series of controls on top of the interface allows users to quickly change between looking at different
grades, disciplines, the raw RIT score, or student growth (see Figure 7). An additional set of bar graphs
present the frequency distribution of raw RIT scores and individual growth since the beginning of the