Science and Mathematics Achievement
- Factors Influencing Achievement
Given the differences in course taking, differences in science and mathematics achievement are not surprising. Trends in science and mathematics achievement since the early 1970s reveal persistent differences by race and sex at ages 9, 13, and 17 despite the narrowing of many gaps. 
Male students score slightly higher than female students on the National Assessment of Educational Progress (NAEP) science and mathematics achievement tests at all ages. (See figures 2-1 and 2-2.) At age 17, the gap between males' and females' mathematics and science scores is smaller than in the 1970s, but the narrowing of the gap is not statistically significant.
See appendix tables 2-3, 2-4, and 2-5.
See appendix tables 2-8, 2-9, and 2-10.
Male and female students have similar mathematics proficiency at ages 9, 13, and 17, although males' average scores are slightly higher. In previous years, female students at age 9 had a slight edge over male students, but in 1992, male scores edged higher than those of females. (See figure 2-1.) Although males showed the most gains at age 9, female students improved most at age 17. The result of these increases is a similar percentage of males and females scoring at or above selected anchor points. In 1992, 82 percent of males and 81 percent of females scored at or above 200 at age 9, 78 percent of both sexes scored at or above 250 at age 13, and 60 percent of males and 58 percent of females scored at or above 300 at age 17. (See appendix table 2-6.)
Female students also score lower than male students on the NAEP science assessment at ages 9, 13, and 17. (See figure 2-2.) Although the differences are small (from 1 to 3 percent lower), they are statistically significant and have been persistent since 1970 (U.S. Department of Education 1994). The gap between males' and females' science achievement is greatest at age 17, although female students' scores have increased significantly since 1982. In 1982, 45 percent of male and 30 percent of female 17-year-olds scored at or above 300 on the NAEP science assessment. In 1992, 51 percent of males and 42 percent of females in that age group scored at or above 300: a 6-percentage-point increase for males and a 12-percentage-point increase for females. (See appendix table 2-11.)
The differences in mathematics and science achievement by race/ethnicity are much more pronounced than differences by sex, although they have narrowed during the past decade. Mathematics scores improved for white, black, and Hispanic students at ages 9, 13, and 17 between 1978 and 1992. (See figure 2-1.) Gains for black and Hispanic students were higher than those for white students. For example, 13 percent more black 17-year-olds and 18 percent more Hispanic 17-year-olds scored at or above 300 compared with 12 percent more white 17-year-olds. (See appendix table 2-6.)
Despite these gains, mathematics scores for black and Hispanic students remain substantially lower than those of white students at all three age levels. (See appendix tables 2-3, 2-4, 2-5, and 2-6.) The median scores for black and Hispanic students at all three age levels are lower than the 25th percentile scores for white students. The gap between white and black mathematics scores at ages 9, 13, and 17 narrowed between 1978 and 1992, although it is still substantial. The gap between white and Hispanic mathematics scores narrowed at ages 13 and 17, but has remained constant at age 9. (See figure 2-3.)
See appendix tables 2-3, 2-4, and 2-5.
As with mathematics scores, differences in science scores persist across racial/ethnic groups. Scores for whites are substantially higher than those for blacks and Hispanics at all age levels, and differences are greatest at age 17. (See figure 2-4.) Science scores increased for students at all three ages between 1982 and 1992, although scores for some groups increased more than others. The gap between black and white and between Hispanic and white science scores narrowed for 9-year-olds between 1982 and 1992. Fifty-one percent of black 9-year-olds scored at or above 200 in 1992, compared with 39 percent in 1982, a 12-percentage-point increase. The percentage of Hispanic 9-year-olds scoring at or above 200 increased from 40 percent in 1982 to 56 percent in 1992, a 15-percentage-point increase. The comparable gain for white 9-year-olds was from 78 percent in 1982 to 86 percent in 1992, a 7-percentage-point increase. (See appendix table 2-11.) No narrowing of the gap was evident for black or Hispanic 13-year-olds or 17-year-olds.
See appendix tables 2-8, 2-9, and 2-10.
Factors Influencing Achievement
- Family Background
- Characteristics of Schools
Some of the differences in mathematics and science achievement by race/ethnicity can be explained by family background characteristics and school characteristics other than the role of course taking already cited. Minority students are more likely than white students to come from families in poverty, to have parents with low education levels, and to attend "disadvantaged" schools (Peng et al. 1995).
Family background characteristics have a considerable influence on minority participation and achievement in science and mathematics education.
Children from poor families have less access to learning materials and educational activities (Oakes 1990a) and are less likely to complete high school. Socioeconomic status (parental occupation, education, and income) accounts for a substantial amount of the differences in mathematics achievement (Ekstrom et al. 1988). Persistence in high school is strongly associated with family income. Students from low-income families are more likely to repeat a grade and to drop out of high school than students from higher income families. One-third of low-income students who repeated a grade were dropouts in 1992. (See appendix table 2-12.)
A larger percentage of minority students than of white students come from families in poverty with less access to learning materials and educational activities (Peng et al. 1995). Black children, in particular, are more likely than other children to live in single-parent families and to live in poverty. Only 34 percent of black children under 18 live with both parents compared with 79 percent of white, non-Hispanic children. (See appendix table 2-13.) Thirty-nine percent of black families with children under 18 are below the poverty level compared with only 12 percent of comparable white, non-Hispanic families.
Parental education is the single most important predictor of participation in mathematics and science (Berryman 1983; Malcom et al. 1985). Those most likely to go to college or to graduate school are those whose parents went to college or to graduate school. The parents serve as role models and mentors in encouraging their children to have high educational aspirations (Oakes 1990a).
Minority students are more likely than white or Asian students to have parents with low educational attainment: 32 percent of Hispanic, 15 percent of black, and 12 percent of American Indian eighth graders, but only 6 percent of white and 8 percent of Asian eighth graders, had parents or guardians who did not finish high school (Pavel et al. 1995, p. 13). Students at all age levels whose parents had less than high school education scored lower in science and mathematics than students whose parents had higher levels of education. Among students ages 9 and 13, however, the science and mathematics scores of students whose parents had less than a high school education improved more since 1978 than those whose parents attended school longer. (See appendix table 2-14.)
Mathematics achievement is also related to parental immigrant status. Asian students, regardless of immigrant status, score higher than white students in mathematics at grades 4, 8, and 12. (See appendix table 2-7.) Asian eighth graders whose parents are immigrants (i.e., the children are first- or second-generation immigrants) have higher grades and higher mathematics scores than those whose parents were born in the United States (Kao and Tienda 1995).
Characteristics of Schools
Many factors contribute to unequal participation of minorities in science and mathematics education, including tracking, judgments about ability, number and quality of science and mathematics courses offered, access to qualified teachers, access to resources, and curricula emphases. Schools, particularly secondary schools, in urban areas with a high proportion of economically disadvantaged or minority students offered less access to science and mathematics education (Oakes 1990b).
Many schools continue to group students according to ability levels. Grouping students by ability level is more prevalent in mathematics than in science and is more prevalent in grades 9-12 than in the lower grades (Weiss 1994). In both science and mathematics, classes with a high proportion of minority students are more likely to be "low-ability" classes than are classes with a low proportion of minority students. For example, in grades 9-12, 29 percent of the classes with a low proportion of minority students are labeled "low-ability" classes, but 42 percent of the classes with at least 40 percent minority students are so labeled. Conversely, 61 percent of the classes with a low proportion of minority students, but only 9 percent of the classes with a high proportion of minority students, are labeled "high-ability" classes. (See figure 2-5.)
See appendix table 2-15.
Being labeled by ability has a profound impact on student achievement because teachers tend to have different expectations of students in the various groups. Teachers in high-ability classes are more likely to emphasize the development of reasoning and inquiry skills than are those in low-ability classes. Students in low-ability classes are more likely to read from a textbook and less likely to participate in hands-on science activities, are more likely to spend time doing worksheet problems, and are less likely to be asked to write reasoning about solving a mathematics problem. (See appendix table 2-16.)
Qualifications of Teachers
Minority students also have less access to qualified teachers. Mathematics classes with a high proportion of minorities are less likely than those with a low proportion of minorities to have mathematics teachers with majors in the field. (See appendix table 2-17.) Schools with a high proportion of minorities, however, do not differ from schools with a lower proportion of minorities in teachers' highest degree earned. (See appendix table 2-18.)
The instructional emphases in largely minority classes are likely to differ as well. The teachers in science and mathematics classes that have a high percentage of minority students are more likely to emphasize preparing students for standardized tests and are less likely than those having fewer minority students to emphasize preparing students for further study in science or mathematics. (See appendix table 2-17.)
 The National Assessment of Educational Progress (NAEP) has been collecting data on student achievement in science and mathematics (and other fields) since 1969. Conducted by the Educational Testing Service under contract with the National Center for Education Statistics, NAEP assesses the academic achievement of a nationwide sample of students at public and private schools to gauge progress in educational attainment.