The 60-second card
The finding
Practice sets that mix different problem types beat sets of one type, because students have to decide which method to use.
Do this
After students learn each problem type, mix types in practice and review. Keep blocked practice for the first lessons on a new skill.
Best evidence
A preregistered randomized trial in 54 seventh-grade classes: students with mostly mixed practice scored 61% on a surprise test a month later, versus 38% for students with mostly blocked practice (IES project page(opens in a new tab)). The study meets federal What Works Clearinghouse standards without reservations (WWC(opens in a new tab)).
61% vs 38%
mixed vs. blocked practice on a surprise test a month later
The 5-minute brief
Source: Rohrer et al., IES efficacy study; meets What Works Clearinghouse standards without reservations (WWC review)
Why it works
A page of 20 problems of the same type tells students which method to use before they read the first one. Real tests don't. Mixed practice makes students choose the method, which is the skill they need later. It also spaces practice out naturally, so it combines two memory effects at once.
Most textbooks don't do this. Rohrer's team examined 13,505 practice problems in six common math texts and found under 10% were mixed (Rohrer et al. 2020(opens in a new tab)). That is why this principle is a quick win: teachers can fix it by reordering problems they already have.
What it looks like
| Grade band | In practice |
|---|---|
| K–2 | Once both are taught, addition and subtraction facts are mixed on the same practice sheet. |
| 3–5 | Review sets mix multiplication, division, and fraction comparison, with no labels saying which is which. |
| 6–8 | Weekly mixed sets drawing on every unit taught so far (the trial's design). |
| 9–12 | Algebra review sets mix linear, quadratic, and exponential problems so students must identify the type. |
Common mistakes
- Mixing before students have learned each type. Teach it blocked first, then mix.
- Mixing problem types students can tell apart at a glance. The benefit comes when the problems look similar but need different methods.
- Expecting it to feel easier. Teachers in the trial reported mixed assignments took students longer (Rohrer et al. 2020(opens in a new tab)).
What a leader looks for in 5 minutes
- Do practice and homework sets include problems from earlier units, mixed in, not grouped?
What we don't know yet
The trial tested grade 7 math in Florida public schools. A federally funded replication is underway, and evidence outside math is thinner.
Go deeper
- Researcher
- Doug Rohrer
- Translator
- Craig Barton; Kris Boulton (Chalk & Talk episode 82)
Also called: interleaving, interleaved practice, spiral review
Grades and subjects
Sources
- Rohrer et al. 2020, textbooks (opens in a new tab)
link.springer.com
- Rohrer, Dedrick, Hartwig & Cheung 2020, randomized trial of interleaved math practice (opens in a new tab)
doi.org
- Rohrer et al. 2020 (IES) (opens in a new tab)
ies.ed.gov
- What Works Clearinghouse review of the interleaved math practice trial (opens in a new tab)
ies.ed.gov