Those who have learned to read in the United States are probably familiar with the direction to “just sound it out” when faced with an unknown word. When young readers use this strategy to help them decode written words, they rely not only on knowledge of letters and sounds but also upon the understanding that each word can be broken down into component sounds, or phonemes.
But is the underlying process of learning to read, say, “cat” in English and “gato” in Spanish the same? “No” suggests a new study, “How important is teaching phonemic awareness to children learning to read in Spanish?” to be published in American Educational Research Journal by Stanford Professor Claude Goldenberg and colleagues.
Learning about how individual sounds make up spoken words — phonemic awareness — is a cornerstone of beginning reading instruction in the United States. When teaching kids to read in English, it is standard practice to begin in kindergarten and early first grade by teaching them the sounds of the words independent of the letters: for example, “cat” consists of three sounds (k-aa-t).
The study shows that Spanish-speaking students achieved reading comprehension in Spanish without such instruction, raising questions about the approach being used to teach a growing number of Spanish-speaking children in the United States to read in Spanish.
Rather than a one-size-fits-all approach to reading pedagogy, U.S. educators should reconsider their focus on phonemic awareness for students receiving early literacy instruction in Spanish, says Goldenberg, who is on the faculty at the Stanford Graduate School of Education.
While there has been much research about the benefits of phonemic awareness for beginning reading instruction in English, Goldenberg said that the study cautions against uncritically generalizing findings beyond the particular population being studied.
“Here's a specific instance of where reading policy and practice has been very markedly influenced based on a context — English-speaking kids learning to read in English — that is not all that general,” he said. “We need to be aware of the assumptions we're making when we take findings from one linguistic context and apply them to another.”
For Goldenberg, who was born in Argentina and grew up speaking Spanish, the topic is personal. He taught junior high and first grade for several years in predominantly Latino neighborhoods of Texas and Southern California and has long been concerned with challenges Spanish-speaking learners face in the U.S. education system.
In this study, Goldenberg and his colleagues examined the Spanish oral language and reading skills among 571 Spanish-speaking children in first and second grades from three groups: children learning to read in Spanish in Mexico, children learning to read in Spanish in the United States, and children learning to read in English in the United States.
The researchers note that reading instruction looks quite different in the two countries. In the United States, beginning reading pedagogy has been strongly influenced since the 1980s by phonemic awareness research. The Reading First pedagogical framework, adopted as part of No Child Left Behind in 2001, recommended an increased focus on phonemic awareness beginning in kindergarten. As reforms spread through the nation, these recommendations affected instruction even in districts where students were learning to read in Spanish.
In contrast, early reading pedagogy in Mexico does not place such an emphasis on phonemic awareness — the understanding that spoken words can be broken down into their constituent sounds.
Although reading instruction in both countries includes “phonics” — an approach that helps beginning readers learn the relationship between letters and the sounds they represent —only in the United States does phonics instruction follow or accompany the phonemic awareness instruction, which calls attention to the individual sounds in spoken words, not to the connection between letters and their sounds.
Teachers of beginning reading in Mexico also focus more on syllables, once individual letters are taught, and the meaning of what students are reading, as compared with teachers of beginning reading in the United States, according to the study.
In light of these differences, the researchers wanted to compare the children’s achievement in reading over time in Mexico and the United States.
At the beginning of first grade, the children in Mexico scored much lower on reading and phonemic awareness assessments than their Spanish-instructed peers in the United States, who received considerable phonemic awareness instruction in kindergarten and on into first and sometimes second grade. The Mexican children scored even lower in Spanish phonemic awareness than the students in the United States who were instructed in English (whose phonemic awareness was also measured in Spanish).
But by the end of second grade, a striking shift had occurred. Although the Mexican students continued to score lower than U.S. students — again, even those instructed in English — on phonemic awareness, the Mexican students had caught up to or surpassed the U.S. students in what truly mattered: reading achievement.
What might explain the reversal? Previous researchers, mostly in Mexico and other parts of Latin America, had hypothesized that phonemic awareness may be less important for students learning to read in Spanish than in English. For one thing, in Spanish, the relationships between letters and their corresponding sounds tend to be more consistent, whereas a single letter in English can represent a range of different sounds that must be learned.
In addition, these researchers, whose work has been largely ethnographic and observational, found that early readers in Spanish seemed to be focused at the level of the syllable. (A syllable can have one or more phonemes; in Spanish, for example, while the word “pipa” (pipe) has 2 syllables--pi and pa--it has four phonemes, two in each syllable: p - i and p - a.)
“They don’t see kids really dealing with phonemes, and they see phonemic understanding as an outgrowth of actually learning to read,” Goldenberg said.
More definitive answers about Spanish literacy will depend on future researchers taking up these questions with studies that randomly assign participants to experimental and control groups. Goldenberg hopes his work will propel the field to take that next step.
Meantime, these insights are influencing Goldenberg’s current project: a collaboration with Stanford GSE doctoral student Elliott Friedlander MA ’08. They are conducting a randomized control trial with a professional development program aimed at improving literacy instruction in Rwanda.
“The language there is Kinyarwanda, and our Spanish phonemic awareness study has really informed our discussions, because their professional development model is heavily based on research here in the United States, mostly with English-speaking kids,” said Goldenberg. “Phonemic awareness instruction is one of the five pillars of their professional development.”
Although Kinyarwanda shares some features with Spanish in that it has a similar transparent and syllabically-based spelling system, it also has its own complexities and unique tonal features.
Any instructional approach, Goldenberg explained, will probably need to take these distinctions into account.
Goldenberg added that this study’s findings do not address whether kids who learn to read in Spanish and are taught phonemic awareness have an easier time learning to read in English. “There are arguments for both sides,” he said. “It’s another great question for a study.”
The study was co-authored by Leslie Reese from California State University, Long Beach; Tammy D. Tolar and David J. Francis from the University of Houston; and Antonio Ray and Rebeca Mejía-Arauz from ITESO University, Guadalajara, Mexico. Funding was provided by the National Institute of Child Health and Human Development (NICHD).
Tanner Vea, a PhD student in the Learning Sciences and Technology Design program at the Graduate School of Education, is an intern in the GSE communications office.