State says ELA scores are up a little bit. What's your take?

In just a little while, we’ll have some analysis of the newly released state English language arts test scores, which show that children in grades 3 through 8 are, on average, more proficient in reading and writing than they were last year, by 7 percentage points. But first we need your help sorting through the numbers morass! (State Education Department Commissioner Richard Mills wasn’t lying when he said at a press conference today, “We’re going to give you more data than you’ve ever seen before.”)

The broadcast of today’s state press conference, as well as the slide show that Mills presented there, is now online. The state also made available an enormous PDF of every single school’s proficiency breakdowns and scale scores, by grade. And for the really ambitious armchair data analysts, there’s also a 5-megabyte Microsoft Access database of all of the state’s raw data.

I’ve posted the state’s full press release, which touts “steady, moderate growth” across the state, after the jump. Feel free to leave a comment with your insights about the data dump.

STUDENTS STATEWIDE CONTINUE STEADY, MODERATE GROWTH ACROSS GRADES 3-8 ON ENGLISH TESTS; ACHIEVEMENT OVER TIME MEASURED FOR FIRST TIME Achievement in English in grades 3-8 has improved overall this year, according to 2009 test results announced today by Regents Chancellor Merryl H. Tisch and State Education Commissioner Richard Mills. Seventy-seven percent of students across grades 3-8 achieved the standards in English this year, compared with 68 percent last year. And across the various groups of students, fewer showed serious academic problems in English. Despite these improvements, many students are still not meeting the standards. Testing of students in English in each grade, 3 through 8, began in 2006. Prior to that, students were tested only in 4th and 8th grade. In 2006, the Regents also adopted a grade-by-grade curriculum that sets clear expectations for instruction. The grade-by-grade tests, together with the individual student data system, enable the Department to track the academic growth of classes of students over time. It is now possible to see, for example, how students who were third graders in 2006 are performing this year as sixth graders. In the past, the Department was able to look only at specific grade levels and how those grade levels compared from year to year. Both types of data – growth over time as well as year to year comparisons – are being presented today. “The Regents are committed to an open evaluation of data. We’ve therefore asked the Commissioner to examine today’s results in terms of the percentage of students meeting the standards as well as student growth over time,” Regents Chancellor Merryl H. Tisch said. “By focusing on scale scores, a clearer picture of student performance emerges. That picture shows progress, but also points to the areas needing additional attention, particularly for children in the gap. ” State Education Commissioner Richard Mills said, “These results show that the children who began the grade-by-grade curriculum in the early grades are making bigger gains than those who started later. And the children who took the third grade test this year performed better still –showing that state foundation aid, pre-K, and rigorous instruction are paying off. ” Today’s results indicate that students who began the 3-8 grade-by-grade curriculum in the early grades typically make bigger gains as they progress through school than did older students who started the grade-by-grade curriculum in later grades. Middle school performance, which was very low when the grade 3-8 curriculum began four years ago, has improved significantly over time – indicating that these students are better prepared to do high school-level work. The results also show that while the percentage of students meeting the standards has gone up significantly, average scale scores show only moderate improvements as students progress from one grade to the next. The average student scale score across all grades, 3-8, increased by only four points this year, as compared with three points last year and two the year before that. Over time, many students have improved their performance enough to move over the line from Level 2 to Level 3, indicating proficiency. But the increase in the average scale score for those students over time was often smaller. Among the reasons credited for the overall progress of students on the English exams are the following:

  • The State has invested significantly more resources in education over time.
  • Universal pre-kindergarten has been expanding and reaching more students each year, so children are better prepared for school.
  • The Regents adopted a grade-by-grade curriculum in 2006, helping to better guide instruction. Schools are aligning their instruction with the grade-by-grade curriculum.
  • Schools have created literacy teams of teachers and are increasing professional development opportunities to improve instruction.

The achievement gap in English continues to slowly close, and similar trends are emerging for different groups of students over time. Following are some examples:

  • Students with Disabilities: More students with disabilities met the standards as they progressed from one grade to the next (except among this year’s eighth graders). As well, fewer students with disabilities are scoring at Level 1 (i.e., showing serious academic difficulties) as they progress through the grades. For example, 37.3 percent of students with disabilities who took the English exam in 2006 as third graders scored at Level 1; only 0.7 percent of those same students scored at Level 1 when they took the exam this year as sixth graders. While the results for students with disabilities have improved overall, they remain low.
  • English Language Learners (ELLs): More of this year’s fourth, fifth, and sixth grade ELLs have advanced to Levels 3 and 4 as they have moved through the grades. The percentage of ELLs scoring at Level 1 has decreased significantly. For example, 31.5 percent of ELLs who took the English exam in 2006 as third graders scored at Level 1; only 0.8 percent of those same students scored at Level 1 when they took the exam this year as sixth graders. Overall performance, however, remains low and the improvements are small.
  • Black and Hispanic Students: Across all grades, black and Hispanic students have experienced significant growth in meeting the standards. Last year, 52.9 percent of black students, across grades 3-8, met the standards; this year 64.3 percent did. Similarly, last year, 52.6 percent of Hispanic students, across grades 3-8, met the standards; this year, 64.8 percent did. These gains exceed the gains achieved by white and Asian students statewide.
  • Need/Resource Categories (NRCs): The various NRCs include: Low Need schools, Average Need schools, Rural schools, Urban-Suburban schools, Large City schools, and New York City schools. Across all grades, 3-8, districts in all NRCs have made significant gains in the percentage of students scoring at Levels 3 and 4. Their scale score gains, however, were generally much smaller.
  • Big 5 School Districts: In each of the Big 5 City School Districts (New York, Buffalo, Rochester, Syracuse, and Yonkers), the percentage of students scoring at Levels 3 and 4 has significantly increased over the past four years. And in each of the cities, students’ average scores have also increased over those four years.

New York’s testing system, including the grade 3 through 8 tests, passed a rigorous peer review conducted by the U.S. Department of Education. In addition, distinguished national experts on the state’s Technical Advisory Group meet regularly to review all aspects of New York’s assessment system; that group said that the State used commonly accepted psychometric practices in designing and administering these tests. Students receive a specific scale score on the tests which falls into one of four levels:

  • Level 4 – Exceeds the learning standards
  • Level 3 – Meets the learning standards
  • Level 2 – Partially meets the learning standards or meets part of the learning standards
  • Level 1 – Shows serious academic problems.

A more detailed breakdown of results as well as school by school results is available on the web at www.nysed.go