The Effect of Second-Language Instruction on the Reading Proficiency and General School Achievement of Primary-Grade Children Marion Potts American Educational Research Journal, Vol. 4, No. 4. (Nov., 1967), pp. 367-373. Stable URL: http://links.jstor.org/sici?sici=0002-8312%28196711%294%3A4%3C367%3ATEOSIO%3E2.0.CO%3B2-T American Educational Research Journal is currently published by American Educational Research Association.
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/aera.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.
JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact
[email protected].
http://www.jstor.org Fri Jun 1 19:51:48 2007
THE EFFECT OF SECOND-LANGUAGE INSTRUCTION
ON THE READING PROFICIENCY AND GENERAL
SCHOOL ACHIEVEMENT OF PRIMARY-GRADE
CHILDREN
MARION POTTS
Cornell University The possibility of interference effects between language systems has militated against the inclusion of foreign-language programs in the primary grades. Yet foreign-language instruction is operating under major federal subsidy, and evidence coming from several professions suggests t h a t such learning would be more effective if begun in early childhood. Authorities in the fields of child development (Gesell & Ilg, 1961), neurology (Penfield, 1964; Glees, 1962), linguistics (Haugen, 1961), and education (Andersson, 1960) recommend that second-language learning be initiated no later than the beginning of formal education. Other educators disagree (Dickinson, 1961; New York State Education Department, 1962), contending the earliest feasible period for foreign-language instruction to be during the eight-to-nine age range. I n the schools of this country, the latter point of view has prevailed. A major concern of curriculum specialists is second-language interference with beginning reading in the native language, and hence, with general school achievement. Underlying this concern are the results of research on bilingualism, some of which have indicated possible language deficit from the coexistence of two languages in a single nervous system (Mitchell, 1937; Smith, 1939, 1949; Darcy, 1952). Generalization from the bilingualism research to the present issue is of questionable validity, however, due to dissimilarity of both the sources and the conditions of interference. Studies of bilingualism deal with interference between two parallel (two oral, or two written) codes, where there has been long exposure to both languages. With primary foreign language programs, concern is with interference from the oral code of a second-language system to the written code of the first, under comparatively limited exposure conditions. This investigation was designed to test for such interference effects. It is somewhat related to the research of Geigle (1957), Johnson and others, (1960), and Leino and H a a k (1963), which dealt with the effects of
368
AERJ
VOLUME 4
NUMBER 4
NOVEMBER 1967
foreign-language instruction on basic learning in the intermediate grades. The present study was concerned with the primary level because of evidence that this is the optimal time for foreign language introduction; i t focused on beginning reading, since possible negative effects in that area are of critical concern; it eliminated the factor of deletion-of-time from the basic curriculum, in order to determine the effects of secondlanguage learning itself.
The subjects were 43 first graders and 37 second graders attending a New York State campus school. Random assignment to one of the two sections a t each grade level was made for both children and teachers. Each of the four classes involved was then divided into random halves: half of each class formed the experimental group; the other halves, the control. All randomization was done separately for the sexes, to control for girls' earlier language maturation. At the beginning of the school year, the California Test of Mental Maturity, a measure of mental age, was administered to subjects in groups of five to seven. To control for achievement press in the home, Hollingshead's two-factor Index of Social Position was given. The experimental group then received French instruction by the audio-lingual method for 15 minutes daily for one school year. During the same treatment period the control group was given dance instruction, both to control amount of time spent on the basic curriculum, and to minimize occurrence of a Hawthorne effect. In ,June, again in groups of five to seven, all subjects took the California Achievement Test, a measure of general school achievement, and the California Reading Test, as a major subsection.
Results of preliminary correlational and multiple regression analyses suggested elimination of the Index of Social Position as a control variable. The covariance analyses then, used two adjusting variables: language mental age, and nonlanguage mental age, as found on the CTMM. Multiple R of the covariates with the California Reading Test was .71, explaining 50.5 percent of the variance of the reading measures; the R with the California Achievement Test was .74, explaining 55.5 percent of the variance of the total achievement criterion. The assumptions underlying use of the analysis of covariance were tested for both the reading and achievement data. Bartlett's test was applied for homogeneity of variance. Procedures given in Winer (1962, p. 586) were used to test homogeneity of within class regression. Dixon and Massey's procedure (1957, p. 197) for testing linearity of regression was applied. In no case did the F-ratio a t the .05 level reach a value making the use of covariance inappropriate.
369
EFFECT OF SECOND-LANGUAGE INSTRUCTION
Two hypotheses were tested. The first stated that there is no difference in English reading proficiency between primary-grade children who have had French instruction for a school year, and primary children who have had an equal amount of instruction in a noncognitive activity. The second hypothesized no difference in general school achievement under the same conditions. Total number of items correct on the California Reading Test was used as the criterion measure to test the first hypothesis. I n testing the second, total number of items correct on the California Achievement Test was the criterion. TABLE 1 Means and Standard Deviations of Control Variables, and Adjusted Means and Standard Deviations of Criterion Variables, for Experimental and Control Groups Experimental Group
Control Variables Criterion Variables
Lang. MA Nonlang. MA ~ e a d i n gRaw Score Achievement Raw Score
S.D.
Orig. Mean
Adj. Mean
17.27 17.07
92.71 86.15
-
16.36
68.83
Control Group S.D.
Orig. Mean
Adj. Mean
-
16.87 18.79
93.67 92.46
-
70.66
17.83
70.35
68.52
43.88 192.03 197.51
-
47.90 195.79 190.30
Analyses of covariance were computed using the formulae for unweighted means analysis described in Winer (1962, p. 594). Table 1 reports the original and adjusted test means for reading and achievement data, together with mean mental ages for comparison. Because the treatments TABLE 2 Analysis of Covariance of Reading Scores (Adjusting for the Effects of both Language MA and Nonlanguage MA). Sum of Squares
df
Mean Square
F
P
Treatment Subjects (within)
86.90 11,146.00
1 76
86.90 146.66
<1
>.05
Total
11,232.90
77
Source of Variation
+
where y = 0.212 0.422 Total regression coefficients y = 0.212 $0.442 Pooled-within regression coefficients y = total items correct on the California Reading Test z = language mental age as measured by the CTMM z = nonlanguage mental age as measured by the CTMM
-
AERJ
370
VOLUME 4
. NUMBER 4 . NOVEMBER 1967
TABLE 3 Analysis of Covariance of Achievement Scores (Adjusting for the Effects of both Language MA and Nonlanguage MA). Sum of Squares
df
Mean Square
F
P
Treatment Subjects (within)
979.51 71,721.66
1 76
979.51 943.71
1.04
>.05
Total
72,701.17
77
Source of Variation
-
+
where y = 0 . 5 3 ~ 1.282 Total regression coefficients y = 0 . 5 2 ~$1.342 Pooled-within regression coefficients y = total items correct on the California Achievement Test
z = language mental age as measured by the CTMM z = nonlanguage mental age as measured by the CTMhl
might have resulted in either greater or less progress, two-tailed tests of significance were applied. When t,he final scores were adjusted to take into account the mental maturity of the children, no significant differences were found between the groups, either in reading proficiency (F < 1.00), or in general achievement (F = 1.04, p > .05). Tables 2 and 3 summarize the analyses of covariance for all data.
The results of this study revealed no non-chance difference, in reading or in general school achievement, between the group which had been instructed in a second language and the group exposed to a noncognitive activity. If interference effects did occur, they did not detract from overall functioning as measured by the criterion tests. Other data from interference theory literature can be related to this finding. Interference is known to vary with the degree of differentiation between two lists or two response systems. With high differentiation, there will occur more intralist (within language) associations and fewer extra-list (between language) associations-hence, less interference. This would probably be the case when the organism has already learned one system well under one set of circumstances: home environment, close affectional relationships, gratification of primary needs; and later learns a new one under a very different set of stimulus situations (classroom environment-related), as were conditions in this study. Another possible explanation for the absence of evident interference is that of "set." Underwood and Schulz's (1960) selector mechanism and Postman's (1961) background conditioning include subject and situational variables. Subjects' responses are influenced by associations with these variables: these change the stimulus situation. In A-B, A-C, where an English response is called for by the first, and a French response by the
EFFECT OF SECOND-LAXGUAGE INSTRUCTION
371
second, some aspect of the situation is always different (interlocutor, environment, etc). The A-B, A-C paradigm of interference might be said to no longer exist when the child perceives the differential elements in the two stimulus situations. Important to the present study is the fact that the reading-writing system of a language is a different code from its oral system. While interference often occurs between two parallel systems, one would not necessarily predict its occurrence between an oral system of one language and a written system of the other. We no longer have A-B, A-C, but A-B, D-C. I n the first situation, English graphemes are responded to with English phonemes. I n the second, the real world (situationally including a stimulus cue associated with French), is responded to with the French oral system. Moreover, underlying the separate learning of the English writing system and the French oral system is evidence that the neural processes for reading and writing are located in a different cortical area from those of speech (Penfield, 1964). Of additional importance is the fact that the subjects were instructed in the second language for only 15 minutes daily, making it unlikely that this exposure would create evident interference with skills practiced throughout the day. Yet this is the situation which we wish to generalize. Under typical classroom conditions, there seems to be little chance for significant interference with learning the written system of the first language, from the oral system of a second. This study found none, either in reading or in general achievement as related to reading. Review of the literature further indicates that the combination of first-language overlearning (Carroll, 1961) and unfixed neural connections (Penfield, 1964) which exist during the six-seven age range, make this a highly favorable period for the initiation of second-language learning.
REFERENCES ANDERSSON, THEODORE. "The Optimum Age for Beginning the Study of Modern Languages." International Review of Education 6: 298-308; 1960. CARROLL, JOHN B. "Language Development in Children." Psycholinguistics. (Edited by Sol Saporta.) New York: Holt, Rinehart, and Winston, 1961. pp. 331-345. DARCY,NATALIET . "The Performance of Bilingual Puerto-Rican Children on Verbal and on Nonlanguage Tests of Intelligence." Journal of Educational Research 45: 499-506; March 1952. DICKINSON, ELSA."Foreign Languages in the Elementary Schools?" New York State Education 49: 25-27; December 1961. J. Introducfion to Statistical DIXON WILFRIDJ. and MASSEY,FRANK Analysis. New York: McGraw-Hill Book Co., 1957. 488 pp.
372
AERJ
VOLUME 4
NUMBER 4
. NOVEMBER 1967
GEIGLE,R ALPHC. "Foreign Language and Basic Learnings." Elementary School Journal 57: 418-419; May 1957. L. Statement on Learning of Foreign GESELL,ARNOLD and ILG,FRANCES Languages in Early Childhood, as given to hfodern Language Association. Cited in William R. Parker, The National Interest and Foreign Languages. Washington, D . C.: Dept. of State Publication 7324, 1961. pp. 9-10. GLEES,PAUL.[ L N e ~ r ~ l o g i cAspects al of Foreign Language Learing in Young Children." Bulletin of the Texas Foreign Language Association 4: 4-5 ;April 1962. HAUGEN,EINAR.''The Bilingual Individual." Psycholingusitics. (Edited by Sol Saporta.) New York: Holt, Rinehart, and Winston, 1961. pp. 395406. HOLLINGSHEAD, AUGUSTB. Two-Factor Index of Social Position. New Haven: Yale University, 1956. 8 pp. (Mimeo.) JOHNSON, CHARLESE .; FLORES, JOSEPH S.; and ELLISON,FREDP. The Eflect of Foreign Language Instruction on Basic Learning in Elementary Schools. Champaign: University of Illinois Foreign Language Project, 1961. 96 pp. (Mimeo.) LEINO,WALTERB. and HAAK,LOUISA. The Teaching of Spanish in the Elementary Schools and the Effects on Achievement i n Other Selected Subject Areas. St Paul, Minn.: St. Paul Public Schools, 1963 78 pp. (Mimeo.) MITCHELL,A. J . ''The Effect of Bilingualism in the Measurement of Intelligence." Elementary School Journal 38: 29-37; September, 1937. Introducing Children to NEW YORK STATEEDUCATION DEPARTMENT. Languages. Albany: the Department, 1 9 6 2 . 4 6 ~ ~ . PENFIELD,WILDER."The Uncommitted Cortex." The Atlantic 214: 7781; July 1964. POSTMAN, LEO. "The Present Status of Interference Theory." Verbal Learning and Behavior. (Edited by Charles N. Cofer.) New York: McGraw-Hill Book Co., 1961. pp. 152-179. SMITH, MADORAH E. "Some Light on the Problem of Bilingualism as Found from a Study of the Progress in Mastery of English among Preschool Children of non-American Ancestry in Hawaii." Genetic Psychology Monographs, Vol. 21, No. 2,1939. pp. 119-284. SMITH,MADORAH E. L'Meas~rement of the Vocabulary of Young Bilingual Children in Both of the Languages Used." Pedagogical Seminary 74: 305-310; June 1949. UNDERWOOD, BENTONJ . and SCHULZ,RUDOLPHW. Meaningfulness and Verbal Learning. Chicago: Lippincott, 1960.430pp. WINER,B. J. Statistical Principles i n Ezperimental Design. New York: McGraw-Hill Book Co., 1962.672 pp. (Received December, 1966)
EFFECT OF SECOND-LANGUAGE INSTRUCTION
373
AUTHOR POTTS,MARIONH.Address: Cornell University, Ithaca, N. Y. Title: Asst. Professor of Child Development Age: 35 Degrees: B.A., St. Joseph's College; M.A., Univ. of Minnesota; Ph.D., Penn. State University Specialization: Language acquisition; cognitive development; instructional strategies.
http://www.jstor.org
LINKED CITATIONS - Page 1 of 1 -
You have printed the following article: The Effect of Second-Language Instruction on the Reading Proficiency and General School Achievement of Primary-Grade Children Marion Potts American Educational Research Journal, Vol. 4, No. 4. (Nov., 1967), pp. 367-373. Stable URL: http://links.jstor.org/sici?sici=0002-8312%28196711%294%3A4%3C367%3ATEOSIO%3E2.0.CO%3B2-T
This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.
References The Effect of Bilingualism in the Measurement of Intelligence A. J. Mitchell The Elementary School Journal, Vol. 38, No. 1. (Sep., 1937), pp. 29-37. Stable URL: http://links.jstor.org/sici?sici=0013-5984%28193709%2938%3A1%3C29%3ATEOBIT%3E2.0.CO%3B2-O
