What Happened to Spanish?

How high-stakes tests doomed biliteracy at my school

By Grace Cornell Gonzales

Illustrator: Ricardo Levins Morales

Ricardo Levins Morales

Andrés, normally high-energy and jittery, was shaking so hard he could barely hold his pencil. Oscar, who used to come into the classroom smiling and who wrote me love notes in his student-teacher journal, was rhythmically banging his head on his desk. Geovanny would barely sit down; he walked in circles around his desk, slamming into people or furniture that got too close, muttering angrily to himself.

Looking around, I couldn’t believe these were the same 3rd graders I had started the year with. On the first day of school they had been so eager to learn.

All my students at our Oakland, California, bilingual K-5 school were English language learners. Most came from families that had emigrated from Mexico relatively recently, and I had two newcomers, one from Mexico and one from Peru. Until 3rd grade, these kids’ school days had been primarily in Spanish, with English language development for two blocks in the afternoon. Our program was designed to value Spanish and promote literacy in students’ home language first, with the ultimate goal of biliteracy. In 3rd grade, we switched to English instruction for language arts and math. It was a somewhat abrupt change, but my class had been enthusiastic about tackling more English time. And the year had been full of exploration and questions about our world. We went to Alcatraz Island to learn about the native birds. We mucked about in tide pools and explored local sea life at the aquarium. We had a pet crayfish that hatched a ton of tiny translucent crayfish babies. We studied the solar system and tracked the phases of the moon. We made dioramas to show the structure of an Ohlone village. Yet now, in April, we found ourselves silent, everyone staring at a test in English that would take up a month of our instructional time.

I was at least as anxious as the kids. I walked around behind them, peering over their shoulders, terrified by every wrong answer I saw. I wanted so badly to remind them about the time we learned that one trick in math class, or how we reviewed what that particular word meant last week and compared it to its Spanish cognate. Of course, I knew that I couldn’t.

I also knew that my job was on the line. It was my second year in the classroom. I was still two months away from finishing my teaching credential, still two years away from tenure. I began teaching with no meaningful training or experience, right out of college, through a teaching fellows programs. My second year was better than my first, but I was still a long way from being a confident, experienced teacher. My principal didn’t hesitate to let me know just how aware she was of my failings.

There was no fair way to talk about this test to the kids. We wouldn’t get the results back until the fall, so it wouldn’t affect moving on to the next grade. But it would affect our school’s Academic Performance Index (API score). In year 5+ of Program Improvement Status under No Child Left Behind (NCLB) rules, that was a big deal. It could also affect my job. My principal looked carefully at how our students had improved from the previous year. Every fall, she prepared certificates for each teacher, stating how many of their students had moved from “far below basic” to “basic,” from “basic” to “proficient.” She presented them in front of the whole staff, so it was clear who had “moved” 10 students and who had moved only four. Everyone was expected to clap and look happy to receive the certificates and the accompanying dose of public praise or shaming.

Looking at Andrés, I wasn’t surprised that he was shaking. His dad was an intimidating man, and I could almost imagine the conversation at his house the night before: “Andrés, you better do your best on that test or they won’t let you go to 4th grade!”

We had tried to explain to our parents that this test wouldn’t affect their children’s report cards or whether they were promoted. But it wasn’t that surprising that they didn’t believe us. It didn’t make sense that the test their children would cry over and struggle through for a full month was irrelevant to the kids. And, to be fair, the STAR test results would be looked at in aggregate, over several years, to determine whether our students could be reclassified as Fully English Proficient by middle school. In middle school, if they were still classified as Limited English Proficient, they would enter a tracked system that might limit their opportunities in high school and beyond.

Testing English learners, some of them newcomers, in English always leads to a host of problems. But in our bilingual program, which was designed to value and make space for Spanish, it seemed like a particular betrayal of trust. Over the four years that I worked at my school before moving to another district, I saw just how confusing and damaging high-stakes standardized testing can be to both bilingual kids and to bilingual programs themselves.

Biliteracy Succumbs to Testing Panic

From kindergarten through 2nd grade, we spent a lot of time telling our kids how important Spanish was. We worked hard to make a separation between the languages—students spoke only Spanish with their homeroom teacher and only English with the teacher who taught the two English blocks in the afternoon. During Spanish time, we provided high-quality literacy and math instruction: We highlighted academic vocabulary in Spanish, focused on reading comprehension and writing skills, and taught phonics and conventions. This was all in accordance with our program’s philosophy. Our school was dedicated to helping students develop their Spanish for the long term. We started with a 90/10 ratio of Spanish to English and worked to about 50/50 by 3rd grade, which was the ratio the program was meant to maintain through 5th grade. Students were supposed to graduate bilingual and biliterate, proud of their identities and connected to their communities.

But in California at that point, annual high-stakes standardized testing started in 2nd grade. By April of 2nd grade, our students had received exactly zero language arts or mathematics instruction in English. Our transition to English instruction in these subjects began in 3rd grade. How is it fair to evaluate kids, teachers, and programs based on a test taken in the wrong language?

The solution, at least according to our administrators, came in the form of a test prep boot camp. Second-grade teachers were expected to carve daily time out of their Spanish instructional blocks for a month or so before the test. During this time, they would switch to English and teach transferability and test prep strategies. The idea was that, with enough willpower, they could make up for the fact that these kids’ entire educational lives so far (as far as language arts and math were concerned) had been in their home language. Instead of gently transitioning into English language arts instruction at the beginning of 3rd grade, the students were thrust into it through test-prep boot camp.

This sent the message loud and clear, starting at the end of 2nd grade: When the tests roll around, English is the only important language. Your teacher, who had spoken to you only in Spanish before, suddenly switched to English and started talking to you in that language about how to compose a friendly letter or the rules for quotation marks. This new information in a new language was so important that it interrupted and supplanted all the work that you had been doing so far in Spanish.

The problems started in 2nd grade, but they didn’t end there. In the 4th and 5th grades, students were supposed to have social studies and science in Spanish. The idea was that they would continue expanding their academic vocabularies and their literacy skills in their home language during these blocks. However, during my second year at the school, 4th grade switched its science blocks into English because of concerns about student scores on a standardized science test given at the end of that grade. This rippled up to 5th grade. Suddenly, our bilingual program wasn’t looking so bilingual anymore. By the time I left the school, 4th and 5th grade were de facto English Only, and our maintenance bilingual program had essentially become a transitional program ending in 3rd grade.

Double Testing and Double Standards

Another thing that hit us hard as a bilingual school was double testing. At that point, bilingual elementary school teachers in California were expected to test in both languages. However, our principal hinted to us that the Spanish test results wouldn’t really count. Since the kids would be exhausted after the first round of testing, we were asked to test them first in English, and then in Spanish. We never looked at or analyzed the Spanish scores, even though they were probably a better indicator of how much students had learned.

To emphasize the fact that the Spanish test was a throwaway, nonbilingual homeroom teachers were not required to give it to their classes. Our school had two such teachers. In both cases, they were paired with a partner teacher who was bilingual and taught their students during the Spanish blocks. For example, in 3rd grade, my partner teacher and I switched classes daily—she taught my students math in English and I taught her students science in Spanish. Her students were in the same program as mine and had received the same amount of Spanish instruction from kindergarten to 3rd grade. However, because she didn’t have a bilingual credential, her students didn’t take the Spanish standardized test.

Of course, language isn’t the only problem with these tests. Some states allow students to choose the language in which they are tested for language arts, and others provide side-by-side translations of the tests. However, testing students in their home language does nothing about issues of cultural bias in standardized tests. It does little to help newcomers and other students who may not have had access to formal schooling or literacy instruction in their home language. It doesn’t take into consideration the generally poor quality of translations, and the fact that regional variations in vocabulary in a language like Spanish can be extremely confusing for students of diverse backgrounds.

It also doesn’t recognize that bilingual students can’t be treated like two monolingual students rolled up into one: Their oral language and literacy development in both languages necessarily proceed at a different pace and along different axes than those of their monolingual peers. Most importantly, it doesn’t factor in the reality that the messy, joyful work of learning to think, create, and reason—in one language or two—can never be captured by clicking a multiple choice response on a computer screen or by bubbling in an A, B, C, D, or E with a No. 2 pencil.

As for my students, they were required to take the full standardized test, language arts and math sections, in both English and Spanish. We tested Tuesday, Wednesday, and Thursday, for two hours every morning, for about a month. As can be imagined, not a lot of high-quality instruction happened afterward on those days. Our administration wanted us to “make it fun.” We had pep rally assemblies and special snacks each morning of testing. But there’s no quantity of granola bars that can make 8 and 9-year-olds tolerate that much testing and that much tension for that long without starting to crack.

What Are the Consequences?

Our district said it supported bilingual programs. So did our administrators. It’s easy to say that you want kids to be bilingual and biliterate. But when you encourage schools and teachers to do something and then punish them for doing that very thing, there’s a problem with your system.

We know bilingual education leads to academic achievement in the long term; kids in bilingual programs will, by middle school, often perform at higher levels than their peers in monolingual programs (Thomas and Collier, 2002). But that doesn’t mean that 2nd and 3rd graders, who have barely begun to practice reading and writing in English, will do as well on English standardized tests as students in an English Only program. How could they?

Expecting the impossible sets bilingual programs up for failure. At the end of the day, you don’t need to pass laws banning bilingual education in order to eradicate programs. A system of standardized testing that punishes schools for teaching native language literacy is enough of a deterrent by itself.

I don’t think Andrés, Oscar, or Geovanny ever forgave me for that month that I made them sit quietly in their seats and bubble in answers with No. 2 pencils. Why should they? Third grade ended on a sour note, for them and for me. We never recovered the joy that we had felt at the beginning of the year when we created crayfish habitats or measured our shadows in chalk on the playground.

The systems in place during my second year of teaching—NCLB, program improvement status, California’s API scores and STAR tests—are giving way to new standards, new tests, and new ways of ranking and punishing schools. Under the surface, nothing has changed. The English tests are still the tests that count, and kids still know that. Bilingual schools still pay for trying to implement bilingual programs.

But these days, more families, students, and teachers are speaking out and opting out. For many of these advocates, the needs of English language learners are an important consideration. During the MAP boycott in 2013 in Seattle, teachers specifically cited the negative effects on emergent bilingual students as one of many reasons they believed the tests were doing harm. At International High School in Brooklyn, teachers banded together in 2014 to boycott the New York State English Language Assessment exam, citing how inappropriate the test was for their student population—majority English language learners. These are just two of many instances of teachers standing up for their emergent bilingual students.

As advocates for bilingual education and as teachers in bilingual programs, it’s important that we join with these educators and make our voices heard. We need to talk about how these tests don’t and can’t show what our kids know or what we are capable of teaching. We need to call districts out when they say they support bilingual schools and then punish them for “underperforming.” Otherwise we risk seeing more and more of our bilingual programs go down without a fight.


Thomas, Wayne, and Virginia Collier. 2002. A National Study of School Effectiveness for Language Minority Students’ Long-Term Academic Achievement. Center for Research on Education, Diversity and Excellence, UC Berkeley.

Grace Cornell Gonzales is an editor at Rethinking Schools and is currently working on Rethinking Bilingual Education due out in 2017. All student names are changed.