Saturday, November 17, 2012

GATES FUNDED HARVARD STUDY EVALUATES TEACHER EFFECTIVENESS IN LAUSD AS MEASURED BY STUDENTS’ STANDARDIZED TEST SCORES IN MATH

 

►NEW L.A. TEACHERS TOO OFTEN PLACED WITH NEEDIEST STUDENTS, STUDY FINDS

By Teresa Watanabe, LA Times/LA Now | http://lat.ms/TWfwc3

November 16, 2012 | 5:38 pm :: A new study has found that inexperienced teachers in the Los Angeles Unified School District are disproportionately more likely to be assigned to lower-performing math students, perpetuating the achievement gap.

The study also found that L.A. Unified teachers “vary substantially” in their effectiveness, with top teachers able to give students the equivalent of eight additional months of learning in a year compared with weaker instructors.

Such findings raise “deep concerns,” said Drew Furedi, the district’s executive director of talent management who oversees teacher training. “For us, it’s a call to action.”

The study by the Strategic Data Project, which is affiliated with Harvard University’s Center for Education Policy Research, analyzed the performance of about 30% of L.A. Unified teachers based primarily on their students’ standardized math test scores from 2005 through 2011 in grades three through eight. The study stressed that test scores were only one measure of teacher effectiveness.

The study also found that teacher performance after two years is a fairly good predictor of future effectiveness. That finding could be used to challenge growing moves to overturn laws that give tenure to teachers after just a few years by those who argue administrators need more time to make that decision.

“Two years gives you a substantial amount of information,” said Jon Fullerton, the center’s executive director. [••smf: Before coming to Harvard, Jon served as the Board of Education’s director of budget and financial policy for the Los Angeles Unified School District.]

Fullerton said that L.A. Unified teachers varied more than those in three other school districts studied in North Carolina and Georgia. More so than in the other districts, Los Angeles schools also disproportionately placed newer teachers with less-proficient students –- who are an average six months behind peers assigned to more experienced instructors.

The study did not explore the reason for those findings but was aimed at providing “information and insight” to the district to craft responses, Fullerton said.

In other findings, the performance of math teachers improved quickly in the first five years, then leveled off. Those with advanced degrees are no more effective than those without, although L.A. Unified pays more to teachers pursuing them. And long-term substitute teachers, who have been employed more frequently to fill in amid widespread layoffs, have positive effects in teaching middle-school math but not in other subjects.

No single finding can produce a strategy to erase the district’s substantial achievement gap between white students and their black and Latino classmates, the study said, noting that the difference in performance on fifth-grade math tests is roughly equivalent to more than one and half years of learning. Multiple strategies will be needed, the study said.

Furedi said one key area of action would be the placement of effective teachers with lower-performing students. L.A. Supt. John Deasy has made it clear that principals should strive to “understand where teachers are and place those with success in front of kids who need them most," Furedi said.

 

►ANALYSIS SHOWS DIFFERENCES IN TEACHER EFFECTIVENESS IN LAUSD

By John Fensterwald, EdSource Today | http://bit.ly/QOwKe2

November 15th, 2012 :: In Los Angeles Unified, novice teachers tend to be assigned students who are academically farther behind those assigned to experienced teachers. Before they depart, usually after only two years, Teach for America teachers have a bigger impact on students than that of other new teachers. And National Board Certified teachers significantly outperform other teachers in LAUSD.

These are among the findings of an extensive six-year study of about a third of teachers in LAUSD by the Strategic Data Project, which is affiliated with the Center for Education Policy Research at Harvard University. Researchers have conducted similar analyses of teacher recruitment, development and retention patterns in three dozen school districts and charter organizations nationwide, under work funded by the Gates Foundation. LAUSD’s report, which was released Wednesday, could become a key resource as the district and United Teachers Los Angeles negotiate changes to teacher evaluations and other parts of the teachers’ contract.

The finding with perhaps the biggest implication quantified significant disparities in effectiveness among the district’s elementary and middle school teachers, as measured by students’ standardized test scores. Researchers found that the difference between a math teacher in the 75th percentile – those whose students performed better than three quarters of other students – and a teacher in the 25th percentile was the roughly equivalent benefit to a student of having eight additional months of instruction in a calendar year (technically one quarter of a standard deviation). The differences were greater than the average of the other districts studied nationwide, although similar to the differences found in San Diego Unified. The study covered only 30 percent of teachers in the district – those whose students take the California Standards Tests, primarily elementary and middle school teachers. The differences between teachers whose students take the English language arts tests were less pronounced than with math.

Researchers used a method similar to the district’s controversial Academic Growth over Time, which factors in students’ past test scores and socioeconomic background to determine their teachers’ impact. LAUSD has used the method to rate individual teachers. Because ratings fluctuate significantly year to year, the method has been criticized as a tool for evaluating teachers. But Jon Fullerton, director of the Center for Education Policy Research, said that the method is useful for aggregate trends using group averages of teacher effectiveness – comparing novice and experienced teachers, or measuring the effectiveness of teachers with advanced academic degrees.

Drew Furedi, LAUSD’s executive director for talent management, acknowledged in an interview that test scores are just one of the multiple measures the district will use to measure effectiveness. But the data in the report, he said, most immediately “will help our understanding of placement strategies and distribution challenges in the system.”

Among other findings:

Only about one out of six Teach for America teachers return to teach in Los Angeles Unified after three years. Source: SDP Human Capital Diagnostic in the Los Angeles Unified. (Click to enlarge.)

Only about one out of six Teach for America teachers return to teach in Los Angeles Unified after three years. Source: SDP Human Capital Diagnostic in the Los Angeles Unified. (Click to enlarge.)

New teachers hired through Teach for America and the district’s own Career Ladder program, encouraging paraprofessionals to become teachers, have a positive effect ­– equal to two months and one month extra of instruction respectively in math – compared with other novice teachers. However, about two-thirds of Teach for America teachers, recruited from top colleges nationwide, leave the district after the program’s required two years, while more Career Ladder teachers continue on – a factor LAUSD will want to consider. Over the past several year, about one out of eight teachers were hired through the two programs. Furedi said that the district may do a similar analysis of other teacher preparation programs.

Novice and early career elementary teachers disproportionately are assigned struggling students – those entering the year an average of six months behind their peers. Novices not only are disproportionately assigned to schools with lower achieving students, but also are given more struggling students than experienced teachers within schools. Since experienced teachers on average are more effective, “this is a call to action, for better strategic placement of teachers for student outcomes,” Fullerton said Wednesday in a webinar announcing the results.

Math teachers in LAUSD improve substantially over time in the classroom – roughly equal to an additional three months of instruction by their fifth year. Improvement in English language arts is less dramatic.

Of the teachers who were laid off, 45 percent were in the top two quartiles of effective teachers in Los Angeles Unified. Source: SDP Human Capital Diagnostic in the Los Angeles Unified. (Click to enlarge.)

Of the teachers who were laid off, 45 percent were in the top two quartiles of effective teachers in Los Angeles Unified. Source: SDP Human Capital Diagnostic in the Los Angeles Unified. (Click to enlarge.)

Even though they are automatically paid more under the current salary system, LAUSD teachers with advanced academic degrees are no more effective than those who lack them. However, teachers with a National Board Certification outperform other teachers, by roughly two months of additional math instruction and one month of additional ELA instruction over a year. This could reflect the benefits of the program or the quality of the teachers who pursue certification, the study said. Only about 4 percent of LAUSD teachers have board certification, and most of those teach in high-performing schools, indicating the district may want to encourage placement in schools with greater needs.

In a finding with implications for a state law and district policy requiring layoffs by seniority, the study found that teachers who were laid off in LAUSD as a result of budget cuts were about as effective as teachers who kept their jobs. Since most of the laid-off teachers were less experienced, a slightly higher proportion – 55 percent – were in the bottom two quartiles of performance. However, that also meant that 45 percent of those let go were in the top two quartiles of performers, who, under a more rigorous evaluation system deemphasizing seniority, might retain their jobs.

 

►From the Study: SUMMARY FINDINGS:

1. Teacher effects vary substantially in LAUSD, more than in many other districts. The difference between a 25th and 75th percentile elementary math teacher is over one-quarter of a standard deviation, which is roughly equivalent to a student having eight additional months of instruction in a calendar year.

2. Teach for America and Career Ladder teachers have higher math effects on average than other novices in their first year by 0.05 and 0.03 standard deviations respectively, which is roughly equivalent to one to two months of additional learning. These differences persist over time.

3. LAUSD has increased its reliance on extended substitutes in the last several years. Relative to other new hires in middle school, extended substitutes have large positive effects in math, though not in other subjects.

4. First-year teachers are assigned to students who begin the year academically behind students assigned to more experienced teachers.

5. Early-career elementary teachers change grade assignments at higher rates than more experienced teachers. Teachers have slightly lower effects after a change in grade assignment than those who do not change grades.

6. LAUSD math teachers show substantial growth in effectiveness during their first five years in the classroom with a 0.12 standard deviation gain in their average teacher effect, which is roughly equivalent to three additional months of instruction in a calendar year.

7. LAUSD teachers with advanced degrees do not have higher effects, on average, than their colleagues without such degrees.

8. On average, National Board Certified teachers outperform other teachers with the same levels of experience by 0.07 and 0.03 standard deviations in elementary math and English/language arts (ELA) respectively, which is roughly equivalent to two months of additional math instruction and one month of additional ELA instruction.

9. Performance in the first few years of teaching, as measured by teacher effects, is predictive of later performance. In fact, in the third year of teaching, teachers who previously had high teacher effects for two years (the top 25 percent) outperform teachers who previously demonstrated low teacher effects (the bottom 25 percent) by almost seven months of instruction.

10. Teachers who were laid off in 2008-09 and 2009-10 had similar average teacher effects as their colleagues who were not laid off.

2cents smf’s 2¢:

1. One must take into account the funder of the report. That said, just because the Gates Foundation paid for it doesn’t mean it’s entirely bogus!

2. There are no authors attached to this report, it is an anonymous work.

3. It is not peer reviewed

4. Compounding and publishing data for hire is not the same as scholarly research.

5. The report relies almost exclusively on Math standardized test score results, The report itself says “other research suggest(s) that other factors outside of the classroom have a larger influence on children’s ELA performance (and) current research also suggests that ELA state tests may also be less sensitive to instruction”. One must conclude that “other factors outside the classroom” means poverty, the language spoken in the home and parent’s education. In Los Angeles this means where you live.

 

The Study: SDP HUMAN CAPITAL DIAGNOSTIC, Los Angeles Unified School District November 2012

No comments:

Post a Comment