TOPICS > Education

Report Card

November 13, 2003 at 12:00 AM EDT


MARGARET WARNER: Is all the recent focus on training students in basic math and reading skills paying off? The 2003 National Assessment of Education Progress, or NAEP, test scores for fourth and eighth graders were released today, and the results were mixed. There was no significant change in reading scores, but math scores have improved. In 1996, 37 percent of fourth graders and 39 percent of eighth graders were not meeting even basic grade requirements in math. That dropped to 23 percent of fourth graders and 32 percent of eighth graders this year. Further up the achievement scale, only 19 percent of both grades were proficient in 1996 in math. Now, 28 percent of the fourth graders and 24 percent of the eighth graders meet that standard.

This is the first year that all 50 states and the District of Columbia took part and thus can be compared. The gap is huge. For example, in fourth grade math, New Hampshire was at the top with 43 percent proficient or above, 44 percent at basic level, and just 13 percent below basic. Contrast that with the two bottom scorers. Mississippi had just 17 percent of fourth graders proficient in math, 45 percent at the basic level, and 38 percent below. And in Washington, D.C., only 7 percent were proficient, 29 percent at basic level, and 64 percent below basic.

For more on these results and their significance, we’re joined by Lynn Olson, senior editor at Education Week. And welcome, Lynn.

LYNN OLSON: Thank you.

MARGARET WARNER: On balance, how do you read this? Is this an encouraging report?

LYNN OLSON: Well, certainly in math it is encouraging. I think what you’re seeing is a continuation of a decade-long trend of slow, steady improvements in math scores. In reading, the results are pretty stubborn. Reading scores have been fluctuating by about four points up and down since the test was first given in 1992.

MARGARET WARNER: Help us, before we go on, to explain these… just briefly these basic standards. I mean, there’s below basic and there’s basic. Then, there’s sort of above basic, then there’s proficient and then there’s advanced.

LYNN OLSON: NAEP has three achievement levels: Basic, proficient and advanced. And those are set… the proficient level, which is the one people talk about most, represents solid performance over challenging subject matter. So it’s a high standard, and in fact, some people have complained that it may be too ambitious compared to where states may be setting the bar for proficiency.

MARGARET WARNER: All right, now, go back to this trend that you had noticed, that is apparent in these numbers– that is that math has been improving but reading, not. What do educators say about why that is?

LYNN OLSON: Well, one of the things people have been saying to us is that math is something you really learn in school, and reading is something that children learn not only in school, but also out of school. Are they read books to at home? Are they in a literature-rich environment? And so it’s harder in some respects to see the same sorts of changes in reading. We also know that there are a lot of efforts to improve early reading now at both the state and the national level. But those are fairly recent, and we may not be seeing the results of those efforts in the test scores yet.

MARGARET WARNER: The other trend I noticed was that, where there is improvement, particularly in math, the improvement is great, or has been greater at the lower grade, at fourth grade than it is at eighth grade. Why is that?

LYNN OLSON: That’s also very typical, that what we’ve tended to see is that it’s easier to show improvement in the early grades, much harder in the middle school level, and particularly at the high school level, where there’s some concern that 12th graders who take the national NAEP may not have much motivation to take it seriously and to do well.

MARGARET WARNER: This may be the time for you to explain really what these NAEP tests are and how do they differ from the state tests that we really hear about all the time, the ones mandated under No Child Left Behind that states, I think, are giving every single year to every single child?

LYNN OLSON: NAEP is often called the nation’s report card because it’s really the only national barometer we have of how students perform in the core academic subjects based on a representative sample, nationally and now in every state. About 340,000 students took the test in each subject this year. That’s different than the test that states give, where every state selects a different test, sets a different cut score for what “proficient” means, has a different set of content standards that those tests are based on.

MARGARET WARNER: And is it the case that those states are going to have their state test scores now, and levels, like proficiency, measured against the NAEP results?

LYNN OLSON: Well, that’s an interesting question. There’s nothing in federal law that says the government is going to compare the two, but NAEP is really the only independent measure out there to which you can compare all these state results. Under the No Child Left Behind law, for the first time, every state has to participate in NAEP. Every state also has to bring 100 percent of their students up to the proficient level on state tests using the state’s definition of proficiency, within the decade. So clearly this is an independent measure that people are going to be looking at to try to make sense of state results.

MARGARET WARNER: Now, as we pointed out, this is also the first time that states have been compared. As I recall, members of Congress were very opposed to these state-by-state comparisons.

LYNN OLSON: Right, this isn’t the first time they’ve been compared. It’s the first time that every single state is required to participate. When NAEP started in 1969, it was very explicit on the part of Congress, on the part of the states that they didn’t want to be able to compare results across states. In 1990, they started to do a voluntary state sample. But in past years, no more than 41 states and the District of Columbia have ever participated — at most, and it’s been lower than that in some years. So this is the first time we really have all states on board.

MARGARET WARNER: And as we showed, the gap certainly between the highest and the lowest is tremendous. Now, what ramifications is that going to have?

LYNN OLSON: Well, you know, it’s hard to interpret gaps across states, in part because people wonder whether the samples of students who take the tests with accommodations, the samples who were excluded differ by state.

MARGARET WARNER: I’m sorry. You have to explain that term.

LYNN OLSON: Sure. NAEP selects a sample of students to take the test in every state. Some students need special test materials or procedures to be able to take the test– for example, more time or a large-print edition of the text. There are there some accommodations that NAEP doesn’t allow. You can’t read the reading test out loud to students. Some states let their reading tests be read allowed to students, so those students fall out of the NAEP sample.

MARGARET WARNER: So you’re saying, in other words, though there are national standards here and it’s graded nationally, in some states, the pool is a little different.

LYNN OLSON: The pool may be a little different, and states often talk about that when you try to compare these results across states. I think what’s interesting, as you look across states, is not only how many students score proficient in one state versus another, but also what do the gaps look like in each state between various groups of students and what’s the trend line like over time? Is the state making progress compared to where it started on these tests?

MARGARET WARNER: I’m sure you’ve looked more closely at the data. What does the data say about the success of efforts to close the achievement gap between, say, white students and minority students?

LYNN OLSON: I think that we’re seeing progress in some states. Nationally, there was a lot of gap-closing; then it stopped for a while. What you’re seeing now is, in a couple of states certainly, fewer and fewer African Americans, Hispanic, low-income students scoring in that below basic level, and that’s good news, is that states are starting to bring up the bottom performers in those groups.

MARGARET WARNER: Finally and briefly, there have been a lot of reports in the news lately, I think more about the state tests, raising questions about how credible they are. For instance, reports that some states lowered their standards so more students would be declared proficient, or some states holding back students so they wouldn’t go into the next pool, the next grade, testing grade. Is this… are these results subject to any kind of manipulation like that?

LYNN OLSON: These results really are not, and that’s one reason why people are going to look at the percent of students who score proficient on a state test and the percent who score proficient on NAEP. Now if the percentage who score proficient on a state test is way higher than the percentage score proficient on NAEP, that doesn’t necessarily mean that the state’s definition of proficient is wrong and NAEP’s definition is right. But it does mean that people should ask questions about how the state has set that bar and what it means.

MARGARET WARNER: Lynn Olson, thank you.