Psychological Reports, 2009, 105, 970-974. © Psychological Reports 2009

ORTHOGRAPHIC TRANSPARENCY AND MORPHOLOGY OF SPANISH–ENGLISH COGNATE ADJECTIVES1 JOSÉ A. MONTELONGO, ANITA C. HERNÁNDEZ, ROBERTA J. HERTER California Polytechnic State University San Luis Obispo, California CARISSA HERNÁNDEZ Allan Hancock College Santa Maria, California Summary.—The English and Spanish languages share over 20,000 cognates. Cognates are words that are orthographically, semantically, and syntactically similar in two languages. In 2009, Montelongo, Hernández, and Herter collected orthographic transparency ratings for over 2,000 Spanish–English cognate nouns and cognate adjectives drawn from the Juilland and Chang-Rodríguez’ Frequency Dictionary of Spanish Words. The present analysis of the cognate adjectives in the Montelongo, et al. norms identified orthographic and morphological characteristics which affected ratings of cognate transparency. The analysis identified an initialletter effect: the earlier an English word deviates from its Spanish equivalent, the lower it is rated. Similarly, the more orthographically similar an English suffix is to its Spanish suffix equivalent, the higher its rating.

Cognates are words that are orthographically, semantically, and syntactically similar in two languages. For example, the English adjective “complete” is the equivalent of the Spanish adjective “completo.” There are over 20,000 Spanish–English cognates, distributed throughout all frequencies of usage in the English language (Montelongo, 2002). Lado (1957) and other educators have suggested using cognates to teach English language learners because they hear obvious resemblance to each other. Teachers can help Latino English language learners develop their reading comprehension skills through cognate recognition strategies (Jiménez, 1997). They can also use concept induction exercises to help their students learn morphological regularities (Garrison, 1990). Including cognates in context vocabulary exercises to help students decipher unfamiliar noncognate words can also facilitate vocabulary acquisition (Montelongo, Hernández, Herter, & Hernández, in press). Despite such recommendations, cognates remain underutilized in language classrooms (Fitzgerald & Cummins, 1999). Some educators hesitate to teach cognate-related strategies because there are “false cognates” that are orthographically identical or similar, but not semantically equivalent. For instance, the word “red” signifies a Address correspondence to J. A. Montelongo, Ph.D., College of Education, California Polytechnic State University, San Luis Obispo, CA 93407 or e-mail ([email protected]). 1

DOI 10.2466/PR0.105.3.970-974

MORPHOLOGY OF COGNATE ADJECTIVES

971

color in English. In Spanish, however, the word “red” means “net.” An English language learner thinking that “red” was the same as “net” would probably misunderstand the text. While the vast majority of Spanish cognates are semantically similar to their English equivalents, a small number of false cognates can be troublesome (Grabe, 1991). Another reason why cognates are underutilized is that cognate recognition is not an automatic process (Jiménez, 1997). When a group of students was asked to identify cognates in reading passages, many cognates went uncircled (Nagy, García, Durgunogulu, & Hancin-Bhatt, 1993). Why were some cognates recognized and not others? One answer is that Spanish–English cognate words possess various amounts of orthographic transparency. For example, the English word “natural” is identical to the Spanish word “natural.” English language learners recognize the likeness between these cognates. Other cognates are not as transparent. For instance, Latino English language learners would have trouble recognizing cognate pairs such as loyal–leal as equivalents (Montelongo, 2002). Research has not differentiated between cognates which are easily recognized from those which are not. This has limited the utility of cognates. To provide researchers with calibrated stimulus materials, Montelongo, Hernández, and Herter2 collected ratings of orthographic transparency for Spanish–English adjective cognates drawn from Juilland and Chang-Rodríguez’s Spanish frequency count (1964). Thus, researchers would have cognate transparency ratings along with a measure of word frequency. The purpose of this study was to analyze the Montelongo, et al. norms for those characteristics which affect the orthographic transparency ratings of cognates. Orthographic Analysis Montelongo, et al.2 collected ratings of orthographic transparency for over 800 Spanish–English adjective cognates drawn from Juilland and Chang-Rodríguez’s Spanish frequency count (1964). The ratings were collected from 41 university students who completed the ratings for course credit. Since the participants were instructed to base their ratings solely on the orthographic features of the Spanish–English equivalents, no attempt was made to ascertain their proficiency in Spanish. The mean rating of orthographic transparency of the cognates was 5.1 (SD = 1.2) on a 7-point Likert-type scale. Ninety-four pairs of English cognate words (such as “identical–identical”) were rated as identical to their Spanish equivalents. Many of the words in the 6.00 to 6.999 range differed by only an accent mark, as in the pair habitual–habítual. The majority of pairs fell into the 5.0 to 5.9999 category. These cognate pairs differed in Montelongo, J. A., Hernández, A. C., & Herter, R. J. (2009) Transparency ratings for Spanish–English cognate words. URL: http://digitalcommons.calpoly.edu/lib_fac/26. 2

972

J. A. MONTELONGO, ET AL.

their final letter (native–nativo), consonant word ending (generous–generoso), or in the presence of the double consonant in the English word (different–diferente). The cognate pairs rated lower than 5.0 contained vastly different word endings (mentioned-mencionado). Many of the cognate pairs that fell into the 1.0 to 3.0 range differed mainly in their first letters (uncertain–incierto) or possessed vastly different spelling (feigned–fingido). The distribution of the ratings, with examples, is included in Table 1. TABLE 1 Cognate Ratings For Adjectives on a 7-point Scale Rating 7.0 6.0 to 6.999 5.0 to 5.999 4.0 to 4.999 3.0 to 3.999 2.0 to 2.999 1.0 to 1.999

Example

No. (N = 813)

Percent

natural–natural habitual–habítual generous–generoso indecisive–indeciso attentive–atento faithful–fiel svelte–esbelto

94 42 342 190 99 39 7

11.6 5.2 42.1 23.4 12.2 4.8 0.9

Initial Letter Effect Raters attached more significance to differences occurring at the beginnings of words than those occurring later. The cognate pairs which contained letter changes in the initial positions of the words tended to have lower ratings than those cognates which had letter changes at the ends of words. For instance, the word “able” and its cognate equivalent, “hábil” differ on their first letters. Such pairs received a mean rating of 3.2. The cognate pair, “typical” and “típico” first begin to differ on their second letters, and the pair, “double” and “doble” on their third letters. The mean ratings for such pairs were 3.8 and 4.2, respectively. Moreover, the differences among the means become smaller the later an English word and its Spanish equivalent begin to differ. For instance, the mean difference between the cognate equivalents that differ in their first letter (3.2) and those that differ in their second letter (3.8) is greater than the mean difference between those cognate equivalents that first deviate in their ninth letter (6.1) and those that differ later (6.3). Table 2 presents the mean ratings for the positions at which letter changes between the Spanish and English cognate equivalents begin to occur. The Pearson product-moment correlation between the position at which the first difference occurs and the mean rating of the cognate equivalents was r33 = .69 (p