Development of Literacy in Spanish Speakers

A Project Sponsored by the National Institute of Child Health and Human Development and the Institute of Education Sciences of the Department of Education

Overview | Grants Funded | Projects | Products | Resources

Measures Developed for Research in Spanish-English Biliteracy

In 2000, the National Institute of Child Health and Human Development and the Department of Education Office of Educational Research and Improvement (now the Institute of Education Sciences) joined in funding to develop new knowledge relevant to the critical factors that influence the development of English-language literacy (reading and writing) among U.S. children whose first language is Spanish. This initiative is aimed at addressing three overarching questions: (1) How do children whose first language is Spanish learn to read and write in English? (2) Why do some Spanish-speaking children have difficulties acquiring English-language reading and writing skills? (3) For which children whose first language is Spanish are which instructional approaches and strategies most beneficial, at which stages of reading and writing development, and under what conditions, and what teacher knowledge, teaching skills and instructional strategies are required to ensure optimal outcomes?

In order to address these major questions, researchers within the Biliteracy Research Network are adapting and developing measures for use in their longitudinal biliteracy research. These include an adapted measure of English phonology, a newly developed equivalent measure in Spanish phonology, a demographic survey, and a measure of classroom observation. Further, to assess children's higher order linguistic and literacy abilities, measures of cognate awareness in English, equivalent measures of morphological knowledge in Spanish and English, and contrastive measures of spelling, reading, and pronunciation in both languages have been developed. The instruments themselves are in the preliminary stages of development. Upon completion they will be made available to fellow researchers in the community.

Phonology Assessments in English and Spanish
Led by David Francis, University of Houston and Maria Carlo, University of Miami

There is general consensus that the development of phonemic awareness is an important precursor to the development of reading in alphabetic languages such as English and Spanish. Less clear is how development of phonemic awareness in a primary alphabetic language relates to development of phonemic awareness in a second alphabetic language, and how both of these relate to literacy development in the primary and second languages. The lack of comparable assessments of phonology in English and Spanish has hindered research on these questions for Spanish-speaking English-Language-Learners (ELLs). Considerable research has taken place on the assessment of phonological skills in English resulting in a number of useful instruments for this purpose, with the Comprehensive Test of Phonological Processes (CTOPP) (Wagner, Torgesen, & Rashotte, 2000) serving as the consensual "gold standard." Researchers from the Center for Applied Linguistics and the Texas Institute for Measurement, Evaluation, and Statistics at the University of Houston are developing the Test of Phonological Processing in Spanish (TOPPS), an assessment comparable to the CTOPP. They are also evaluating possible adaptations of the CTOPP that are necessary for Spanish-speaking ELLs.

Classroom Observation: How Language of Instruction Relates to Literacy Development
Led by Barbara Foorman, University of Texas - Houston Medical School

Within a four-year (K-3) longitudinal study of the classroom instructional context of English-language Learners (ELLs) in 144 classrooms in urban Texas, border Texas, and urban California, investigators will examine, within and across language programs, what instructional strategies are used in Spanish and in English and how they relate to student literacy development. To address this central question, researchers will use a description of classroom instruction and they will link classroom instructional data to student growth and outcomes in literacy. In order to provide extensive in-depth observations of classrooms, investigators will use a time-sampling procedure of reading/language arts instruction developed in previous research on early interventions with 1600 children in grades K-4 in Houston and Washington, D.C. schools. In the current project, they have adapted this procedure and will use it to quantify the ratio of English to Spanish spoken by teachers and students in order to examine the extent to which the ratio varies within and across language programs of immersion, early transition, late transition, and dual language. They will also code reading/language arts instruction with respect to the amount and quality of time spent on instructional strategies, student engagement, and teaching effectiveness to address such questions as: How do the differences in linguistic structure between Spanish and English inform the design of instructional formats and strategies to teach literacy skills in the two languages? Are instructional strategies that promote literacy development in English (for ELLs in English immersion programs) similar to those that promote literacy development in Spanish (for ELLs in primary language programs)? In addition, researchers will investigate the extent to which ratios of Spanish to English, instructional strategies, student engagement, and teaching effectiveness relate to student oracy and literacy development.

Demographic Questionnaire: Biliteracy Research Common Core
Led by Claude Goldenberg, UCLA, and Diane August, Center for Applied Linguistics

The common core (questionnaire) of demographic data is designed to collect information on select clusters of demographic variables across the Biliteracy Research Network projects. Most of the data will be collected at the child and family level and include information on child and male and female heads of household birthplace, length of residence in the U.S., and education; male and female heads of household occupation and income; child and household language and literacy practices; and household size. School level demographic information will be collected on variables such as ethnic and socioeconomic composition and percent of English learners, particularly those from Spanish-speaking backgrounds. Classroom level information will include teacher's credential status and child's reading group size, composition, and language of instruction. Community level information will be collected from the U.S. Census Bureau Web site and include data on income, occupation, ethnic composition, population density, and language use in the census tract where children go to school. The common demographic data core will be used to describe the samples participating in the network studies and their schools, classrooms, and communities. The data will also be used analytically to explore relationships among these variables and the children's literacy development across a range of communities and local contexts.

Assessing Morphological Awareness among Spanish Speaking ELL's
Led by Maria Carlo, University of Miami, and Diane August, Center for Applied Linguistics

In investigating the acquisition of English literacy in Spanish-speaking children, investigators have developed assessment instruments that tap into children's understanding of derivational morphology in Spanish and English. Morphological knowledge is believed to be one determinant of vocabulary growth among monolingual speakers. Additionally, Spanish-English bilinguals may be able to exploit cross-language morphological relationships for the purpose of expanding their lexical-conceptual knowledge base in each language. Investigators studied the development of morphological knowledge with a Cognate Awareness Test and an Extract-the-Base Test.

The Cognate Awareness Test is multiple choice; examinees chose the word most related to the meaning of the target word. Half of the target words are cognates (high Spanish frequency and low English frequency). The rest, non-cognates, are matched in frequency and word class. It was expected that Spanish-speaking ELLs would outperform English monolinguals on the cognate items and perform below the English monolinguals on the non-cognate items. Additionally it was expected that performance on the cognate items would vary as a direct function of Spanish proficiency among the ELLs. Preliminary results are consistent with this prediction.

The Spanish and English Extract-the-Base Test consists of cloze response items, and requires that examinees extract the base-word from a derived form (e.g., discuss from discussion). The examinees are presented with a target word (e.g., discussion) followed by a fill-in-the-blank sentence (e.g., There is something we need to _____). The difficulty of the items is believed to be a function of the type of transformation performed on the base word. The items are of four types, in predicted order of difficulty: neither phonological nor orthographic transformation (danger/dangerous); phonological (discuss/discussion); orthographic (empty/emptiness); orthographic and phonological (width/wide).

Contrastive Measures of English and Spanish Spelling Ability and English Decoding Ability
Led by Liz Howard, Center for Applied Linguistics

Spelling in English poses many challenges, even to native speakers of the language. The possibility for multiple mappings of letters and sound, the retention of historical spellings, and the multi-level system of English spelling (grapho-phonemic, morphemic, lexical) all contribute to the difficulty of spelling accurately in English. This difficulty is compounded for native Spanish speakers, who must also learn to represent sounds that do not exist in their native language (such as th) and sounds that do exist in Spanish but are represented in different ways (such as e/i). As a result, there is a lot of potential for cross-linguistic transfer (both negative and positive) in the English spelling of Spanish-speaking children. We are systematically investigating the spelling development of these students by: (1) determining the specific features of spelling in one language prone to influence from the other; (2)mapping a developmental sequence of English and Spanish spelling for bilingual students; (3) investigating the extent to which English spelling ability relates to Spanish spelling ability; and (4) exploring the potential relationships between spelling and reading.

Developmental contrastive spelling measures were developed in both English and Spanish. Each group-administered assessment consists of two segments containing series of words chosen for specific features and levels of difficulty. The first segment consists of real words administered via a traditional spelling test format, where children are asked to spell the entire word from an oral prompt. The second segment consists of pseudowords administered using a cloze procedure; here, target features (such as long vowel sounds) are deleted, and children are asked to fill in the missing letters after hearing an oral prompt.

The English contrastive decoding test is a group-administered, multiple-choice assessment that was developed to examine the connection between reading (decoding) and spelling (encoding), with particular attention to sound/spelling correspondences that are likely to be problematic for Spanish/English bilinguals. The test requires that the student mentally map phonemes onto graphemes by matching a set of sounds (a pseudoword read aloud by the administrator) with a set of graphemes (a choice of four pseudowords written on the student's test paper). Four possible choices exist per question: the correct choice, a Spanish-influenced distracter, and two additional distracters.

Home Literacy and Language Scales
Led by Carol Hammer and Adele Miccio, Pennsylvania State University

As part of the "Bilingual Preschoolers: Precursors to Literacy" project at Penn State University, instruments are being developed that are designed to measure the home literacy environment, capture information about the patterns of language usage in the home, and elicit information about the parents' beliefs about language and literacy development. The Home Activities Scale includes items that elicit information about parent and parent-child literacy activities as well as gather data about the types of literacy materials available in the home and the languages in which the materials are available. The Language Usage Scale is used to gather data about children's and parents' age of acquisition of Spanish and English and the languages the parents and children use in the home when talking to family members and significant individuals in their lives. Questions are also asked that are designed to capture parents' attitudes about bilingualism and their children's language development in Spanish and English. The Parent Belief Measure is designed to elicit information about parents' beliefs about how children learn to talk and read and about what actions parents are taking to support their children's language and literacy development. We have collected data using these surveys/scales. The research team has collected pilot data, and is conducting analyses to determine which items are most valid and reliable. From that information, they will develop a shortened version of the scales/surveys. The Parent Belief Measure about language and literacy development is intended to be an open ended questionnaire designed to capture the range of beliefs that exist within the Puerto Rican culture, as there is minimal information about parent beliefs on these topics. This information will be used in designing culturally appropriate programs and interventions for preschool children and their families.

Measures of Syntactic Awareness in English and Spanish
Led by Alexandra Gottardo, Grand Valley State University

Knowledge of syntax is one measure of oral language proficiency. There is a strong belief among educators that oral language proficiency drives the development of word recognition skills (Cummins, 1994) but recent evidence (Durgunoglu, Nagy, & Hancin-Bhatt, 1993; Geva, 2000; Geva & Clifton, 1994; Geva,Yaghoub-Zadeh, & Schuster, 2000) suggests that this may only be the case for reading comprehension. Some researchers believe that metalinguistic awareness is a global metacognitive skill, with syntactic and phonological awareness being related to each other (Tunmer & Hoover, 1992), while other researchers believe that phonological awareness is uniquely related to reading and separate from syntactic awareness (Gottardo et al., 1996; Shankweiler et al., 1992; Stanovich & Siegel, 1994).

Parallel measures of syntactic awareness were developed in English and in Spanish in order to examine awareness of similar syntactic structures across languages. The items were orally presented by the experimenter and required an oral one-word response by the participants. Items tapped different levels of morphological and syntactic knowledge, including plural morphemes and tense markers (e.g., Sally has a party dress and a school dress. She has two______.; María tiene un vestido para fiesta y otro para la escuela. Ella tiene dos______. ; Every day we play. Today we play. Yesterday we______. Tomorrow we______. ; Cada día jugamos. Hoy nosotros jugamos. Ayer nosotros______. Mañana nosotros______.). This was made to include the same types of structures in both the English and the Spanish versions of the task. When possible, direct translations of the English sentences were used for the Spanish stimuli. Both the English and Spanish versions of the task contained 14 experimental items and two training items that were administered to the children. All items in the oral cloze task were scored according to a strict criterion of correctness. A correct response using the strict criterion required that the child give a single word response that fulfills both the syntactic and semantic constraints of the sentence.

Preliminary data was collected with first grade Spanish-speaking children. The English oral cloze task was related to performance on the English vocabulary measure, but only moderately to one measure of English word reading performance. Spanish syntactic and morphological knowledge was related to performance on the Spanish vocabulary measure but not to Spanish word reading. These findings suggest that syntactic awareness is not strongly related to word reading skill and is a separate skill from phonological awareness.

Overview | Grants Funded | Projects | Products | Resources

Untitled Document
CAL Home | CAL Store | Press Room | Jobs | Contact Us | Site Map | Privacy | Links
Copyright © 2014 CAL