单项选择题

2
The longitudinal study demonstrates that students who receive ESL (English as a sec ond language) instruction are far better than those taught primarily in their native lan guage. A comparison of the three-year exit rates for students in ESL and bilingual programs shows that those who receive ESL instruction test out faster and in higher percentages than those who receive instruction in their native language, regardless of the grade in which they entered school. For example, 79.3 percent of the children who entered ESL programs in kindergarten tested out, while only 51.5 percent of those who received their education in their native languages did. Likewise, 72.9 percent of the LEP (limited English profi ciency) students who entered programs in the first grade tested out. while only 38.5 of those in bilingual programs did. For students who entered LEP programs in the ninth grade, 91.6 percent of those in native-language instruction classes still hadn’t tested out after three years, as compared with 78.1 percent of those in ESL classes.
Furthermore, children who had been in ESL classes tested 4 higher in English and math once they exited LEP programs than those who had received native-language instruc tion. Of the LEP students who entered in kindergarten or the first grade, 49 percent of those who had been in ESL classes eventually read at grade level, while only 32 percent of those who had been in bilingual classes performed that well. In math. the statistics are even more impressive. Of the children who entered in kindergarten or the first grade, more than 69 percent of those who had been in ESL classes eventually performed at grade level or above, as opposed to 54 percent of those who had been in bilingual classes.
Naturally, the study provoked a barrage of criticism from the highly political and vo cal bilingual lobby, which prompted the New York City Board of Education to issue a pa per in November 1994 mitigating the findings of the study and ignoring the distinction be tween students in ESL and bilingual education programs. Rather than exit rates, this paper focuses on the achievement of LEP students during the period in which they are in bilingual or ESL classes. The authors show that although the scores of LEP students were below av erage on the English-language test, their scores in all areas showed improvement; they point out that in math, there were insufficient data on the progress of LEP students to draw valid conclusions.
However, a report on citywide mathematics test results in New York in the spring of 1995 deals more fully with the math scores of the 26,248 students who were examined the previous school year in Chinese, Spanish, or Haitian Creole. According to this document, only 16.6 percent of these children performing at or above grade level in mathemat ics. Although this represents an improvement of 1.1 percent over the scores of the previous year, it discredits the argument that native-language instruction keeps performing at grade level in subject areas. Although LEP students are improving faster than the national norm, they continue to perform far below the norm.
The author believes (Par