Reading results for 15-year-olds
PISA 2006 reading literacy results are not reported for the United States because of an error introduced when the test booklets were printing. 21 Thus the reading literacy results described here come from the PISA 2000 and 2003.
In PISA 2003, U.S. 15-year-old students' average reading literacy score of 495 was not measurably different than the OECD average of 494, and placed U.S. 15-year-olds in the middle third of participating OECD nations (table 3). Fifteen-year-old students in 9 of the 29 other participating OECD-member countries outperformed their U.S. peers (as did 15-year-olds in 2 of the 11 non-OECD countries that participated) in terms of average scores. U.S. 15-year-olds in the top 10 percent scored 622 or higher, a cutpoint score below that of the top 10 percent of students in 7 countries (all OECD countries). The bottom 10 percent of U.S. 15-year-olds scored 361 or lower, a cutpoint score below that of the bottom 10 percent of students in 9 OECD countries and 3 non-OECD countries.
PISA has developed five levels of proficiency to help analyze the range of students' performance in reading within each participating country. 22 The highest level of proficiency identifies students who can complete sophisticated reading tasks, such as managing information that is difficult to find in unfamiliar texts; showing detailed understanding of such texts and inferring which information in the text is relevant to the task; and being able to evaluate critically and build hypotheses, draw on specialized knowledge, and accommodate concepts that may be contrary to expectations. For PISA 2003, the highest level of proficiency corresponds with a score at or above 625 score points. 23
In 2003, nine percent of U.S. 15-year-old students performed at this level (figure 3). The same 7 countries whose top 10 percent of students outperformed U.S. students had a higher percentage of 15-year-old students who performed at this level. The percentages of students performing at this level in countries that outperformed the United States ranged from 16 percent in New Zealand to 11 percent in Sweden.
Change over time
Among the 32 countries that participated in both PISA 2000 and PISA 2003, the average reading literacy score increased in 4 countries and decreased in 5 countries (figure 4). 24 For the United States, there was no measurable change in the average reading literacy score between 2000 and 2003 or in its relationship to the OECD average. 25 However, among the countries that participated in both PISA 2000 and PISA 2003, the number that outperformed the United States increased from 7 to 9. 26
The 4 countries whose scores increased (Latvia, Liechtenstein, Luxembourg, and Poland) were all outperformed by the United States in 2000; although, in 2003, Luxembourg scored lower than the United States, Latvia and Poland were not measurably different than the United States, and Liechtenstein outscored the United States. Among the countries whose scores decreased between 2003 and 2000, only Japan saw its score fall relative to the United States: Japan outperformed the United States in reading literacy in 2000 but was not measurably different in 2003. The score for Sweden was not measurably different between 2000 and 2003, but in 2003 its score was measurably higher than the U.S. score whereas in 2000 it was not.
From 2000 to 2006, the average PISA reading literacy score increased in 4 countries that participated in both the 2000 and 2006 PISA (Korea, Poland, Liechtenstein, and Latvia) (OECD 2007). At the same time, 10 countries had lower reading literacy scores in 2006 than in 2000 (9 OECD countries, Spain, Japan, Iceland, Norway, Italy, France, Australia, Greece, and Mexico, and one non-OECD country, the Russian Federation.)
PISA is being repeated in 2009. More detailed information on the PISA 2003 results can be found in Lemke et al. (2004; available at http://nces.ed.gov/pubs2005/2005003.pdf); and in OECD 2004 (available at http://www.oecd.org/dataoecd/1/60/34002216.pdf). More detailed information on the PISA 2006 results can be found in OECD 2007 (available at http://www.oecd.org/dataoecd/30/17/39703267.pdf). For more information on PISA, see http://nces.ed.gov/surveys/pisa/.
21 In various parts of the U.S. PISA 2006 reading literacy assessment test booklet, students were incorrectly instructed to refer to the passage on the "opposite page" when students actually needed to turn back to the previous page to see the necessary passage.
22 See figure 3 for the cut scores for all five levels of proficiency in 2003. For details about all five levels, see OECD 2004, pp.272–79.
23 PISA has defined levels of proficiency based on specific student proficiencies. These specific student proficiencies remain the same across assessments; however, the score point threshold for students who demonstrate these specific student proficiencies may vary slightly from assessment to assessment. For more details on PISA's levels of proficiency and how they differ from the IEA's benchmarks, see appendix A.
24 Note that statistical comparisons between PISA 2000 scores and PISA 2003 scores need to properly account for linking error. See appendix A for details. Because of low response rates, data for The Netherlands were not discussed for PISA 2000 nor were data for the United Kingdom for PISA 2003.
25 Large standard errors for the United States in 2000 may account at least in part for the fact that the U.S. reading literacy score was not measurably different between 2000 and 2003 as well as the fact that the score was not different from the OECD average in 2000.
26 The total number of countries that outperformed the United States in 2000 includes the United Kingdom, which had no official score in 2003 because of low response rates. The total number of countries that outperformed the United States in 2003 includes The Netherlands, which had no official score in 2000 because of technical problems, and Hong Kong, which did not participate in 2000.