« Innovation: Teaching Shakespeare to English-Learners | Main | Guatemalans, Hondurans More Likely to Have Limited-English Proficiency »

ELLs in 8th Grade Lose Ground on Math NAEP


English-language learners in 8th grade performed a tad worse in math on the National Assessment of Educational Progress in 2009 than they did in 2007, according to the scale scores on the test.

The scale score for 8th graders who are ELLs dropped 2 points, from 245 to 243, in two years, which is statistically significant, Jonathan Beard, an associate research scientist for the National Center for Education Statistics, said in an e-mail message. The scores for 4th graders who are ELLs stayed statistically the same over that time. The scores increased by 1 point, from 217 to 218, but that isn't considered significant.

The proportion of English-language learners both in 4th grade and 8th grade scoring "at or above proficient" dropped 1 percentage point from 2007 to 2009, according to a quick comparison of the scores released this week with those from two years ago. (See Tables A-16 and A-24 in the 2009 report for information on ELLs. See Tables A-13 and A-20 in the 2007 report.) This year's NAEP report on math says that 12 percent of ELLs in 4th grade scored proficient or above, while 13 percent did in 2007. Five percent of ELLs in 8th grade scored proficient or above in 2009, down from 6 percent two years before.

But Beard said those particular drops are not statistically significant.

My colleague Sean Cavanagh has written an article this week at edweek.org about how overall math scores on NAEP have stagnated for 4th graders but risen for 8th graders.

Experts on ELLs have characterized the low scores of English-learners in math in 2007 as disconcerting, so I'm sure they will be concerned that the scale scores have dropped for such students in 8th grade. In the past, some experts have delved into NAEP scores for ELLs to gain insights on whether such students in states that require bilingual education fare better than those in states that have greatly curtailed the education method and where English-only instruction is the norm.

My quick examination of the 2009 state-by-state data for ELLs in math doesn't show a clear pattern about the diminished use of bilingual education.

Fewer students in Arizona and California, states that limit bilingual education, scored proficient or above than did ELLs on average nationwide in 4th and 8th grades. But at the same time, a larger proportion of ELLs in Massachusetts, which also restricts bilingual education, scored proficient or above than was true of their counterparts nationwide at both grade levels.

A larger share of ELLs in Texas, which requires bilingual education, scored proficient or above than the national average for ELLs in both grades. Notably, 20 percent of 8th graders who are ELLs in Texas scored proficient or above on the 2009 NAEP, compared with 12 percent nationwide.

But ELLs in New York and Illinois, which also require bilingual education, didn't fare as well as such students in Texas. ELLs who were 4th graders in New York beat the national average for their counterparts scoring proficient or above while those in 8th grade there did not. The reverse was true in Illinois. Fourth graders didn't beat the national average for their counterparts while 8th graders did.

I didn't, by the way, run this quick analysis about ELLs in particular states past Beard, of the Education Department, so some of the differences in proportions that I point out may not be significant.

A news story in Colorado broke out some of the NAEP data on ELLs for that state. Since 2003, ELLs in 4th grade there have increased their scale score in math to 216 from 206. ELLs in 8th grade in Colorado scored behind their counterparts in only three jurisdictions: South Carolina, Virginia, and Department of Defense schools.

I look forward to seeing how researchers interpret these data.


Hi, Mary Ann,
Just a caution: The NAEP statistics on ELLs should be taken with a huge grain of salt. Our 2007 study of urban districts participating in the NAEP (Rivera, C., Shafer Willner, L., and Acosta, B., 2007) showed wide variation in the rates of participation of ELLs in this test. Some districts reported including all of their ELLs, while others said no ELLs participated. Even among districts with higher participation rates, there may be large differences across districts and states in the demographic characteristics of the ELL sample (e.g., home languages spoken, level of English language proficiency, amount of time in U.S. schools, literacy level, etc.). Recently, NAEP has been working to improve its guidance to districts regarding which students should participate, and we hope this will increase the reliability of the sampling. However, that also means that comparisons from year to year continue to be problematic. If all goes well, this will improve in the future, but for now, it is not possible to draw many conclusions from comparisons drawn from such variable populations.

Not speaking from the Department, but given the possible variations listed in the last comment, I would venture to guess that 1-2% is not statistically significant and is actually, quite close. It looks like a plateau. But has the overall population of ELL's gotten statistically significantly bigger in the past year or two? If so, then the school districts might be doing a statistically significant better job even if they appear to be plateauing. Then the question is why have the school systems been able to keep up as well as they did? Better teacher training? Better methodologies? More use of high tech?

The NAEP is given in English. They do allow use of bilingual dictionaries and extra time, separate location but those are accommodations with no proven efficacy. Also, ESOL students have a range of English language proficiency. NAEP could have merely tested more beginning or intermediate ESOL students. Testing in English is illegitimate and shows how this country is being bankrupted by poor education beliefs and policies.

Comments are now closed for this post.

Follow This Blog


Most Viewed on Education Week



Recent Comments

  • Charles: ELLs in our state ARE required to take State standardized read more
  • Melissa: Maybe I'm just becoming jaded, but this feels to me read more
  • Anonymous: Are you kidding me....UNO is an organizaion that literally destroys read more
  • Meg Baker: Are any schools using ACCESS scores for purposes other than read more
  • Dr. Mendoza: This is great news i must say. Hopefully this DREAM read more