Weak and Strong Novice Readers of English as a Foreign Language: Effects of First Language and Socioeconomic Status

This study examined individual differences among beginning readers of English as a foreign language (EFL). The study concentrated on the effects of underlying first language (L1) knowledge as well as EFL letter and vocabulary knowledge. Phonological and morphological awareness, spelling, vocabulary knowledge, and word reading in Hebrew L1, in addition to knowledge of EFL letters and EFL vocabulary, were measured. The study also investigated the effect of socioeconomic background (SES) on beginning EFL readers. Participants included 145 fourth graders from three schools representing two socioeconomic backgrounds in the north of Israel. The results indicate that knowledge of English letters played a more prominent role than knowledge of Hebrew L1 components in differentiating between strong and weak EFL readers. The Linguistic Coding Differences Hypothesis was supported by L1 phonological awareness, word reading, and vocabulary knowledge appearing as part of discriminating functions. The presence of English vocabulary knowledge as part of the discriminant functions provides support for English word reading being more than just a decoding task for EFL beginner readers. Socioeconomic status differentiated the groups for EFL word recognition but not for EFL reading comprehension.

Key Words: Foreign language, reading, socioeconomic status, weak and strong beginners

Automatic decoding, semantic knowledge, and metacognitive abilities are essential for first (Ll) as well as foreign language (FL) reading comprehension (Biemiller, 2003; Durgunoglu, 2002; Lundberg, 2002; Nation & Snowling, 2004; Stanovich, 2000). Poor readers are often characterized by their inefficient decoding skills (Swanson & Alexander, 1997; Vellutino & Scanlon, 1986) as well as less developed vocabulary knowledge (Nation & Snowling, 2004; Snow, Burns, & Griffin, 1998). The present research focuses on individual differences in reading that explain weak English foreign language (EFL) reading acquisition.

The Linguistic Coding Differences Hypothesis (LCDH), which claims that difficulties in Ll linguistic codes (specifically phonological/orthographic, syntactic and semantic) will transfer to FL learning (including reading), provides a framework for understanding students' individual differences in language learning. The authors of the LCDH have found that middlehigh SES, older students with significant differences in native language skills in high school and college also exhibited significant differences in their FL achievement and proficiency (Ganschow, Sparks, Javorsky, Pohlman, & Bishop-Marbury, 1991; Sparks et al., 1998; Sparks, Ganschow, Javorsky, Pohlman, & Patton, 1992). In the present research, the LCDH was tested with Hebrew-speaking elementary school students from two different socioeconomic status backgrounds.

Both English and Hebrew are alphabetic orthographies. English is considered to be an opaque orthography due to its complex orthographic as well as syllabic structure (Seymour, Aro, & Erskine, 2003; Spencer, 2000). Vowelized Hebrew is considered a transparent orthography in that direct grapheme-phoneme translation results in correct word recognition (Geva, Wade-Woolley, & Shany, 1993). Unvowelized Hebrew is considered opaque due to the fact that many unvowelized Hebrew words presented in isolation can be read in numerous ways. In the case of unvowelized Hebrew, the reader is dependent on context in order to correctly decode many words.

DECODING HEBREW WORDS BY ROOT AND WORD PATTERN

Hebrew belongs to the Semitic family of languages and is a root-based morphology. Unlike the concatenative or linear morphological structure of English words, most Hebrew words or word stems are made up of consonantal roots that provide semantic information and word patterns that are affixed or intertwined with the roots in a nonconcatenative manner. Word patterns integrate three root letters (Shimron, 2003). For example, in the word talmid "pupil," the root "lmd" loosely refers to the semantic meaning "studied" and the nominal word pattern is denoted by "ta- -k" Word patterns are mainly represented by vowels and affixes, and they denote syntactical category (Ravid, 2003). The root and word pattern form a bound morpheme with specific phonological, morphological, and semantic characteristics. Young Hebrew speakers develop a sensitivity toward the root-word pattern structure of the language as a result of acquiring speech and reading in a root-based morphology (Berman, 2003; Levin, Ravid, & Rapaport, 2001; Ravid, 2003), particularly when it is taught explicitly in school as a word recognition technique (Bentin & Frost, 1995). In acquiring English reading, native Hebrew speakers may face difficulties resulting from the possible intrinsic differences in word recognition processes that stem from the different morphological constructs of Hebrew and English. It may be that different strategies are used in performing word recognition in the two languages (Geva, 1995; Koda, 1995; Wydell & Butterworth, 1999). When students are studying a FL that is based on the same linguistic properties as their native language, they may use the same skills as they do in their first language for new word recognition. This hypothesis is compatible with the LCDH. When there are considerable differences between the two orthographies, then word recognition strategies may need to be adapted.