Over the Long Term, NAEP Scores Are Way, Way Up

By 10/26/2015

4 Comments | Print | NO PDF |

In anticipation of new NAEP scores coming out this week, I thought it would be useful to spend some time reflecting beforehand on what we know on a macro scale. So rather than focus on year-year-changes or commit other sins of misNAEPery, I’m using data from the NAEP Long-Term Trend data series, which goes back to 1971 for reading and 1973 for math. Here are the scale score gains, on a 500-point scale, over the last four decades (* signifies statistically significant):

4th grade math

• All students: +25*
• White students: +27*
• Black students: +36*
• Hispanic students: +32*

8th grade math

• All students: +19*
• White students: +19*
• Black students: +36*
• Hispanic students: +32*

12th grade math

• All students: +2
• White students: +4*
• Black students: +18*
• Hispanic students: +17*

4th grade reading

• All students: +13*
• White students: +15*
• Black students: +36*
• Hispanic students: +25*

8th grade reading

• All students: +8*
• White students: +9*
• Black students: +24*
• Hispanic students: +17*

12th grade reading

• All students: +2
• White students: +4*
• Black students: +30*
• Hispanic students: +21*

There are at least four important things the data are telling us:

1. Although NAEP scores barely budge year-to-year, over the long term, NAEP scores are way, way up. Remember that no matter what happens this week.

2. Math scores are rising faster than reading scores. There’s nothing new to say here, except to note that this is a large-scale reminder that math scores are easier to improve than reading scores.

3. There are clear age trends emerging in the data. Fourth-graders have made greater gains than eighth-graders, and eighth-graders have made larger gains than twelfth-graders. In fact, we see statistically significant gains in both subjects and in all races, except the composite scores in 12th grade.

4. Changing demographics are masking how much NAEP scores have improved. Although all races are rising individually, scores are rising faster for black and Hispanic students than they are for white students or for the overall composite. As I wrote earlier this year, “Because NAEP takes a representative sample, it’s also vulnerable to something called Simpson’s Paradox, a mathematical paradox in which the composition of a group can create a misleading overall trend. As the United States population has become more diverse, a representative sample picks up more and more minority students, who tend to score lower overall than white students. That tends to make our overall scores appear flat, even as all of the groups that make up the overall score improve markedly.” Here are the percentage of NAEP 4th-grade test-takers who were white over the various testing years:

• 1971: 84
• 1975: 80
• 1980: 79
• 1984: 75
• 1988: 75
• 1990: 74
• 1992: 74
• 1994: 76
• 1996: 71
• 1999: 69
• 2004: 59
• 2008: 56
• 2012: 53

These demographic trends are nearly identical at every age level, and they’re wreaking havoc on our ability to neatly understand our national results. As is clear in the data above, in both math and reading and at every age level tested, all races are improving at least as fast as the nation as a whole. Achievement gaps are closing as black and Hispanic students have made even faster progress.

Similar trends are playing out in other subjects like geography, history, and civics.

I note all this in anticipation that this week’s NAEP results aren’t likely to show much change from the last NAEP results in 2013. Scores may appear “flat,” but we should think of the entire American education system as like a glacier; it may be moving at an almost imperceptibly slow pace, but it is moving.

— Chad Aldeman

This first appeared on Ahead of the Heard.

Comment on this article
  • Michael Olneck says:


    Are the gains distributed evenly across the NAEP proficiency levels, or are they concentrated at particular levels, e.g., lower end?

    Are the distributions of individual scores any less unequal than in the past (which they would be if gains were concentrated at the bottom)?

  • Ze'ev Wurman says:

    It might be helpful if you commented on the huge drop in so-called “white” students in late 1990s. It is not that the white population vaporized overnight or that Hispanic armada landed on American shores in 2000, but rather that OMB changed the way ethnic and racial qualifications were recorded putting a (heavy) finger on the scale to record students as Hispanic. The federal guidelines of recording ethnicity in cases when students refuse to state also lead to over-identification as Hispanics. It does not only show the political power of the Latino caucus, but makes longitudinal comparisons even harder, as Hispanics and white prior to about 2000 are not necessarily the same as Hispanic and white after 2000.

  • Peter Meyer says:

    And to what do you attribute these “way up” results?

  • sandy kress says:

    The rise wasn’t uniform, Chad. Take a close look at the LTTT NAEP data. The 90s were flat. The 2000s were up. The 2010s appear to be flattening again. We need to understand better that and perhaps why this is happening in order to inform our work going forward.

    It’s not as simple as we’re up.

  • Comment on this Article

    Name ()


    Sponsored Results

    The Hoover Institution at Stanford University - Ideas Defining a Free Society

    Harvard Kennedy School Program on Educational Policy and Governance

    Thomas Fordham Institute - Advancing Educational Excellence and Education Reform