Struggling reader or emerging biliterate student? Reevaluating the criteria for labeling emerging bilingual students as low achieving

Hopewell, S., & Escamilla, K. (2014). Struggling reader or emerging biliterate student? Reevaluating the criteria for labeling emerging bilingual students as low achieving. Journal of Literacy Research, 46(1), 68-89.


In U.S. education, there are two areas where common practices just don’t seem to make sense. One of those areas is how we work with learners whose first language is not English. Another area is the way we assess basic skills like reading and how we use the results of assessments to make high-stakes decisions that affect students’ opportunities to learn. The study reported in this article addresses both of those problematic areas.

Hopewell and Escamilla hypothesized that the way reading assessment test scores are interpreted for bilingual students can differ greatly, depending on how those doing the interpreting view the nature of the process of learning to read for bilingual students. In Colorado, where Hopewell and Escamilla did their research, bilingual students are evaluated only on their reading ability in English. Those students scoring below a cut score are deemed “at-risk” and put into intervention programs that may include ability grouping. Hopewell and Escamilla propose that this “monolingual” approach may incorrectly label learners who are not at risk for reading difficulties but do need English language interventions. With the well-documented evidence we have about the negative and lasting effects of being consigned to “low” reading groups, the possibility that many bilingual students are being placed inappropriately in remedial programs is troubling.

The researchers here tested their hypothesis by looking at assessments in both English and Spanish, and then also by looking at the relationship between the two, using a model they developed called “Trajectory Toward Biliteracy”. In this model, both Spanish and English assessments are used. Score ranges for both languages were used rather than strict cut scores. Hopewell and Escamilla maintain that for Spanish/English bilingual learners, it can be expected that scores on English assessments will be slightly lower than scores on Spanish assessments. Acceptable score ranges for both languages are proposed, and Hopewell and Escamilla state that as long as the discrepancy is not large, a student may be considered to be making satisfactory progress, even though the student’s scores in English may be below state cut scores.

Hopewell and Escamilla found that when bilingual third graders were assessed in English only, 84% were labeled as “at-risk”. When the same third graders were assessed in Spanish only, that percentage went down to 55%. When the researchers’ proposed ranges from the “Trajectory Toward Biliteracy” were used to interpret these same assessment data, the percentage of children labeled “at-risk” went down to about 40% (accounting for children scoring below desired ranges in English, Spanish, or both).

These findings clearly point to the need for caution when using assessment results for making placement decisions. This is probably true for all kinds of students, though the focus here is on bilingual students. If we rely on single assessments that do not effectively measure what learners can do, but only look for deficits, we are bound to underestimate learners’ abilities and strengths. In the case of bilingual students, abilities in BOTH languages need to be taken into account. It is clear that failing to do so can lead to an appalling amount of underestimation, and to potentially inappropriate placements that can have devastating consequences for learners.

Another caution highlighted here is to make sure we understand learners and the ways they learn better before devising and mandating high-stakes assessments. We need multiple assessments, and we need better assessments. We need to understand better how the process of learning to read works for bilingual learners, both in their home languages and in their second (or maybe even third or fourth or more!) languages. To automatically label learners as “at-risk” on the basis of one English language assessment demonstrates a lack of understanding of bilingual learners, and of literacy assessment, that is, quite frankly, appalling.

Hopewell and Escamilla propose a way of ameliorating the problem. Their model may not totally eliminate the problems they present to us here. They do at least offer a compromise that may help educators look at bilingual learners’ test data in more informed ways. Whether this compromise will be accepted on a large scale in Colorado is an open question. Whenever we look at ways to improve the ways we assess and provide literacy instruction for bilingual students, walls go up. Researchers may propose ideas, but the walls of monolingualism, ethnocentrism, and probably even outright racism will continue to prevent their findings from being accepted, implemented, or maybe even heard at all. I’d like to say that is changing, but this article tells me there is still a long way to go.