Do ability-based streaming groups benefit students' learning?

Full Question: My primary school says that you are saying "There is no evidence that placing students in ability-based streaming groups benefits students' learning. The academic gains from these types of groupings are too small to be significant. In fact, there are minimal benefits to the higher achieving students and negative effects to the lower achieving students." This is only mentioned in regards to math streaming, streaming in other subjects is continuing. My thinking is this, in ability-based groups the teacher will be able to devote more time to each student (teacher spends less energy on adjusting to different skills, study ability etc. for the widely differing ability of students) and therefore enhanced learning will result. Thoughts?

Answer: “Yes, ability grouping has little impact on much (including in math). Yes, teachers prefer ability grouping for the reasons you note, but that does not mean it works! Students learn much from each other; teachers are more attentive to student diagnoses and needs when they have to attend more to the students--but here are the killers: a) Particularly in high school, students who are later blooming and in lower grades usually have NO chance of moving up as they have not been exposed to the material of the upper grades (and  thus can rarely catch up); and, b) Count the color--in too many schools ability grouping works like apartheid to keep the African American and Hispanics in the lower grades--surely indefensible. Like you, I can give many reasons why ability grouping SHOULD work, but the evidence is pretty compelling--and for the equity reasons surely indefensible. The evidence also means that those teachers in more heterogeneous classes have as great an impact on their students without the equity downsides. The fascinating question is how come they are successful when all the SHOULDs say the opposite?” (John Hattie, personal communication, July 17, 2018).