Leonie Haimson, Class Size Matters
Testimony to the New York City Council Education Committee
on the Mayor’s retention proposal
March 3, 2004
My name is Leonie Haimson and I’m head of a parent advocacy group called Class Size Matters. Though there may be some close calls in educational policy, the Mayor’s retention proposal isn’t one of them. As the 107 academics, researchers, and national experts on testing who signed the letter drafted by Class Size Matters and Advocates for Children agreed, deciding to retain a child on the basis of one test score is not just unfair and unreliable, but it will also lead directly to lower achievement and higher drop out rates. I would challenge this administration to find a single, independent expert who has studied the issue of retention who supports this proposal.On the other hand, those opposed to the Mayor’s announced policy include Dr. T. Berry Brazelton, renowned pediatrician and author of numerous works on child care and development, Robert Tobias, former head of Division of Assessment and Accountability for the Board of Education and now Director of the Center for Research on Teaching and Learning at NYU, and Dr. Ernest House, who did the independent evaluation of New York City’s failed “Gates” retention program in the 1980’s. Other opponents include four past presidents of the American Education Research Association, the nation’s premier organization of educational researchers, as well as many members and the chair of the National Academy of Sciences Committee on the Appropriate Use of Educational Testing, and several members of the Board on Testing and Assessment of the National Research Council. Even the two companies that produce the third grade tests are on record that a decision to hold back a child should never be based upon test scores alone. The third grade tests are norm-referenced rather than criterion-referenced, which means by design a certain number of children must fail them. The Mayor’s proposed policy will also have a hugely negative emotional impact on those children who are left behind. As Dr. Brazelton points out, being held back is highly traumatic for children; Professor Shane Jimerson of the University of California notes that for many, only losing a parent and going blind is more upsetting. Jay Heubert of Columbia University and the study director of the National Academy of Sciences report on testing has pointed out that the decision to retain a student is the strongest predictor that he or she will eventually drop out of school. Indeed, the research on the destructive academic and emotional consequences of retention is so overwhelming that its continued use, according to Prof. Jimerson, amounts to educational malpractice. Holding back approximately 15,000 third graders will also be extremely expensive, costing at least $200 million – money that would much better be put towards programs that have been proven to work, rather than those that have been shown over and over to fail. ( According to DOE, approximately 16,000 students will be retained. Providing them with an extra year of third grade will cost an additional $11,000 each, which totals $176 million. This, plus the $25 million that Chancellor Klein says he will fund in extra services for those students held back, equals $201 million.) For this amount, we could hire enough teachers to reduce class size in grades K-3 in every classroom by three, and still have plenty of funds left over for tutoring for those children who are still struggling.Lost in all the controversy over retention, last week the state released figures showing that 322 elementary schools in NYC made the most improved list, by far the largest number of schools in our history. They improved by token of their fourth grade students’ test scores. We have seen steadily rising test scores at the fourth grade level, with an especially big jump last year.Why? The steady improvement in performance began when the state and the federal government began providing funds in 1998 to reduce class size in the early grades, and last year’s cohort were the first group of 4th graders to have had the benefit of significantly smaller classes since 1st grade. In NYC as a whole, the percentages of fourth graders testing at level one in math sharply declined over four years from 19.4% to 8.7%. In reading, the percentage declined from 21.3% to 8.8%. This amounts to approximately 10,000 fewer fourth graders testing at level One in reading, and about 8,500 fewer in math.The gains in the former Chancellor’s district, where class sizes were reduced the most, were especially dramatic: in these schools, the percentage of fourth graders at Level One declined in four years from 35% to 9.5% in math, and 38% to 13.1% in reading Our statistical analysis of data from 32 NYC school districts shows that the decline in the percentage of fourth graders testing at level one over four years in each district was strongly correlated with how much that district had decreased class sizes in grades K-3, in both reading and math. In math, there was also a strong correlation between the decrease in class size with the increased number of students testing at or above grade level. Unfortunately, instead of continuing to make progress in this area, we are going in the entirely wrong direction. Average class size grew in every grade from K-8 this fall, and we had more 3rd graders in classes of 29 or more. This retention policy will most likely cause even more overcrowding at the 3rd grade level, and thus, to greater numbers of failing students.I’d also like to talk a little about the appeals process that the Chancellor has outlined. Any appeals process is better than nothing, but this one is full of flaws. First of all, for a child who is threatened with retention, his or her parents have to appeal to the teacher, who has to collect a whole portfolio of the child’s work if she agrees that the child should not be held back. The teacher must then go to the principal to make the case for the child, and the principal, if he or she agrees, has to go to the LIS, or local instructional superintendent, who also has to agree that the child should be promoted. Not only will it take till August until the child’s fate is determined, causing months of anxiety and uncertainty, but the process is clearly stacked against promotion. Every one of these individuals must agree not to retain the child, and if any one of them doesn’t, there is no further appeal allowed by the parent and the child will be held back. Instead, the benefit of the doubt should be towards promoting the child, especially when so little research evidence can be found to support the whole notion of retention in the first place.Moreover, parents have lost faith in this administration to carry out any appeals process in an evenhanded and efficient manner. Thousands of children who have been referred for special education evaluations have not received them in a timely fashion this year. According to the Public Advocate’s office, Superintendents are being pressured by the DOE to keep their special education referral rates down, and are keeping referral letters in drawers. As reported in Newsday, the city’s local instructional superintendents have been ordered to flunk at least three principals each by giving them “U” ratings. LIS’s have been told, “The chancellor wants it,’” Reportedly, some principals have also been instructed to fail at least three teachers, no matter what their true level of competence. Similar pressures will likely be brought to bear on these administrators to hold back large numbers of third graders, whatever their real level of achievement and whether or not they think they will benefit from being retained. The Chancellor also said he will provide $8 million for tutoring for children whose promotion is in doubt, which amounts to only $250 per child. Given the fact that these tests will be given in April, and tutors haven’t yet been found or hired, it is also difficult to know how much benefit there will be from these programs, and how many students will be helped to pass their tests, with just a month or less of extra help. All in all, this retention proposal would be an expensive disaster for the children of this city, and I, as well as many other parents, advocates, and experts from around the country urge the members of the City Council to do everything within their power to prevent it from being enacted. Thank you for this opportunity._________________________________________________________
Correlations: changes in 4th grade math scores 1999-2003 per NYC school district based on average change in class size K-3 1998-2002, Leonie Haimson of Class Size Matters and Jacqueline Shannon of NYU
4 year class size changemath level 3&4: four year change 1999-2003Pearson Correlation-.456(**)
Pearson Correlation-.456(**)Sig. (1-tailed).004N33math level 1: four year change 1999-2003Pearson Correlation.425(**)
Pearson Correlation.425(**)Sig. (1-tailed).007N33** Correlation is significant at the 0.01 level (1-tailed).
Correlations: change in 4th grade reading scores per NYC district 1999-2003 based on average change in class size K-3 1998-2002 4 year class changeread: mean scale scorePearson Correlation-.256
Pearson Correlation-.256Sig. (1-tailed).075N33read level 3&4: four year change 1999-2003Pearson Correlation-.152
Pearson Correlation-.152Sig. (1-tailed).199N33read level 1: four year change 1999-2003Pearson Correlation.322(*)
Pearson Correlation.322(*)Sig. (1-tailed).034N33* Correlation is significant at the 0.05 level (1-tailed).
