Synthesis of mathematics results
While U.S. 4th-graders' average scores increased between 1995 and 2007, 4th-graders in Hong Kong, Japan, and Singapore consistently outperformed their U.S. peers in mathematics, as did 4th-graders in the Russian Federation and Chinese Taipei, both of whom only participated in TIMSS 2003 and 2007 (see supplemental table A-3). U.S. 4th-graders lost ground relative to their peers in England and Latvia who improved at a faster rate between 1995 and 2007, but 4th-graders gained ground relative to their peers in Hungary, The Netherlands, Austria, and the Czech Republic (figure 7).
The same three Asian countries that consistently outperformed the United States in mathematics at grade 4 (Hong Kong, Japan, and Singapore) also consistently outperformed the United States in mathematics at grade 8 between 1995 and 2007 (see supplemental table A-4). In addition, Korea and Chinese Taipei also outperformed the United States in mathematics at grade 8 each time they participated in TIMSS. No other participating countries, however, outperformed the United States in 2007, including 6 of the countries that outperformed the United States at grade 8 in 1995: Australia, Bulgaria, the Czech Republic, Hungary, the Russian Federation, and Sweden. Thus while U.S. 8th-graders have not caught up with their Asian peers in terms of their average mathematics score, they have improved their standing relative to their peers in these 6 countries.
Still, when the ability of 15-year-old students to apply mathematics knowledge and skills to real-world tasks was assessed, 23 countries outperformed the United States in both 2003 and 2006 in terms of average scores (see supplemental table A-5). In addition, 15-year-olds in Hungary, Latvia, and Poland—three countries that did not outperform the United States in 2003—outperformed their U.S. peers in 2006. 35
35 The 2003 average scores of Latvia, Hungary, and Poland were not statistically different than the 2003 U.S. average score; and there was no measurable increase in the scores of the United States, Latvia, Hungary, and Poland in 2006 (i.e., they were not statistically different than their own scores in 2003). Nevertheless, the latter three countries' scores were statistically different than the United States in 2006 on account of differences in the amount of their respective standard errors in 2003 compared with 2006.