The teachers of Y7 and Y8 chose two different approaches to collaborative reading.

The curriculum level grade distributions for Hub X (91 students) and Hub Y (63 students) both showed highly significant improvements (p ≤ 0.001) i.e. the students’ reading really did improve in both groups. But which approach would you choose? Is either approach actually better than the other? To improve on the visual comparison, you need to calculate the strength of association or effect size.