Cummins claims for policy decision making (2000)

A substantial research and theoretical basis for policy decisions regarding minority students' education does exist. In other words, policy-makers can predict with considerable confidence the probable effects of bilingual programs for majority and minority students implemented in very different sociopolitical contexts.

First, they can be confident that if the program is effective in continuing to develop students' academic skills in both languages, no cognitive confusion or handicap will result; in fact, students may benefit in subtle ways from access to two linguistic systems.

Second, they can predict that bilingual/ELL students will take considerably longer to develop grade-appropriate levels of L2 academic knowledge (e.g. literacy skills) in comparison to how long it takes to acquire peer-appropriate levels of L2 conversational skills, at least in situations where there is access to the L2 in the environment.

Third, they can be confident that for both majority and minority students, spending instructional time partly through the minority language will not result in lower levels of academic performance in the majority language, provided of course the instructional program is effective in developing academic skills in the minority language. This is because at deeper levels of conceptual and academic functioning, there is considerable overlap or interdependence across languages. Conceptual knowledge developed in one language helps to make input in the other language comprehensible.

Fourth, Ttese psycho-linguistic principles by themselves provide a reliable basis for the prediction of program outcomes in situations that are not characterized by unequal power relations between dominant and subordinated groups (e.g. L2 immersion programs for students from dominant language backgrounds). However, they do not explain the considerable variation in academic achievement among culturally and linguistically diverse groups nor do they tell us why some groups have experienced persistent school failure over generations.

Fifth, why students choose to engage academically or, alternatively, withdraw from academic effort is to acknowledge that human relationships are at the heart of schooling.

Sixth, relationships dominate all participant discussions about schooling in the US. No group inside any school surveyed felt adequately respected, connected or affirmed. Students repeatedly referenced a need for care. What they liked best was when teachers cared about them or did special things for them. Their least liked things included being ignored, not being cared for, and negative treatment. (Popllin & Weeres, 1992: 19)