The Education Gadfly Weekly

Janet Hyde, Sara Lindberg, Marcia Linn, Amy Ellis, Caroline Williams
National Science Foundation
July 2008

That boys are the math whizzes and girls suffer acute equation phobia (hence Barbie: "Math class is tough!") is a common stereotype. It's fueled by evidence that shows that high-school girls take fewer math and science courses than boys, a difference that also serves to explain why boys have historically performed better on standardized math tests in high school. But new findings from a report out of the Universities of Wisconsin and California (Berkeley) are doing their part (along with the other, similar findings that preceded them) to undermine the gender typecasting. Researchers studied achievement data from over 7 million students and found that, for grades 2 to 11, there's no longer a difference between the math performance of boys and girls. Females are still underrepresented in the highest levels of STEM (science, technology, engineering, math) careers, though. Researchers aren't clear exactly why, but they point to the higher variability of boys' scores. For instance, among white eleventh graders in one state studied, the 99th percentile (the very top of the academic curve) contained twice as many boys as girls. So even though average scores are similar between the genders, boys appear to outnumber girls in the ranks of the highest scorers, which may leave them better prepared for high-level math courses in college. Find the short report here (and read it for a fee).

Emily Cohen, Kate Walsh, and RiShawn Biddle
National Center on Teacher Quality
July 2008

This is an indispensable companion to a Fordham study released earlier this year, which found that many big-district teacher contracts are ambiguous and that their school leaders may have more decision-making flexibility than they (or others) think. The National Council on Teacher Quality (NCTQ) gathered and codified the bargaining agreements on which Fordham's study was based; now it has done the same for state laws as well and produced this fine report. Its authors found, after analyzing the data, that state laws, and sometimes courts and labor relations boards, often plug the holes that are left open in teacher contracts. In other words, ambiguities in local bargaining agreements don't always yield autonomy because other forces cramp it. Every state, for instance, has a law that sets the number of years a teacher must serve to gain tenure. In addition, state laws frequently set requirements for policies regarding evaluation, dismissal, class size, and salary. California, for example, stipulates "10 different steps that must be taken before the dismissal is finalized, perhaps explaining why just 100 dismissal hearings were heard in the state between 1996 and 2005." Of course, teacher unions play a central role in pushing such laws through legislatures. NCTQ cites the case of New York, where unions pressured lawmakers to embed "a provision in the 2008-2009 budget that made it illegal to consider a teacher's job performance as a factor in the tenure process." This clever ruse tossed a monkey wrench into New York City's efforts to increase teacher accountability, even though the district's teacher contract was silent on the issue. Read much more about it in the report.

Andrew C. Zau and Julian R. Betts
Public Policy Institute of California
2008

What if you could predict the likelihood that a fourth-grade student would, years later, pass the California High School Exit Exam (CAHSEE)? Two researchers at the Public Policy Institute of California claim to have done just that. By using a longitudinal and comprehensive data-set from San Diego, authors Zau and Betts have identified academic, language, and behavioral markers that help predict the probability that a given pupil will pass the test. For example, a good "behavior GPA," based on a student's classroom conduct, is strongly correlated with passing the CAHSEE. Zau and Betts contend that developing an early warning and intervention system, based on such data, would be smarter than employing rushed tutoring programs during eleventh and twelfth grade. We'd add that states might also "vertically align" their assessment systems so that passing the fourth-grade test means you're on track to pass the high school test. (According to our Proficiency Illusion report, such alignment is missing in a majority of the states we studied.) There's much more to dig out of this interesting study, which you can find here.

Mayoral control of schools is surely no silver bullet, but in the case of Baltimore, where Mayor Sheila Dixon is, according to the Baltimore Sun, "floating the idea" of taking over the schools, it would be a leaden musket ball. The city's relatively new education CEO, Andres Alonso, is quickly making big changes; he's shifted authority away from the district's central office, for example, and has given greater responsibility to principals. Baltimore's entrenched bureaucrats don't like Alonso's style, and it seems they've communicated their distaste to Dixon. She recently told a Sun columnist about the CEO, "You can't come in and change everything." Imagine wanting to take charge of the schools in one's city in order to retard the pace of change! In June, Dixon brazenly criticized Alonso on a radio show: "I cautioned him not to move so quickly in some areas." But quick movement to reform broken classrooms is, in fact, exactly what Baltimore needs--far more than it needs mayoral control.

"Dixon Eyes Bid to Run Schools," by Liz Bowie, Baltimore Sun, July 27, 2008

"Shift Control of Schools? Why Now?," by Jean Marbella, Baltimore Sun, July 29, 2008

All hail ProComp!, we once were impelled, for it hath shown that teachers' unions and reformers can work together for good. Not so fast. Now we learn from Education Week that Denver's teachers' union, "in a recent newsletter, called on teachers to prepare for a strike if negotiations [to reform ProComp] fall through." ProComp, you may recall, is Denver's well-meaning but rather weak-kneed merit pay plan, enacted in 2004 through a much-ballyhooed union-district partnership. It was understood that the program's specifics would be renegotiated every three years. Thus, the city's school district (which just posted big test score gains), led by Superintendent Michael Bennet, recently proposed changes to ProComp that would raise the starting teacher salary from $35,000 to $44,000 and would bump from $1,067 to $2,925 the bonuses that teachers who work in hard-to-staff schools, or who teach subjects such as math and science, could earn. These changes are specifically targeted to attract more high-quality teachers and increase retention. The union, however, doesn't get it--it's stonewalling Bennet's proposals and wants a 3.5 percent increase in the salaries of all educators. How a 3.5 percent boost will attract new, talented teachers to the district and then retain them is unclear; it's also unclear how the union's plan is different from the imprecise, across the board salary schedules to which ProComp was intended to be an alternative. Work with the unions, yes we can?

"Model Plan of Merit Pay in Ferment," by Vaishali Honawar, Education Week, July 28, 2008

Andrew C. Zau and Julian R. Betts
Public Policy Institute of California
2008

What if you could predict the likelihood that a fourth-grade student would, years later, pass the California High School Exit Exam (CAHSEE)? Two researchers at the Public Policy Institute of California claim to have done just that. By using a longitudinal and comprehensive data-set from San Diego, authors Zau and Betts have identified academic, language, and behavioral markers that help predict the probability that a given pupil will pass the test. For example, a good "behavior GPA," based on a student's classroom conduct, is strongly correlated with passing the CAHSEE. Zau and Betts contend that developing an early warning and intervention system, based on such data, would be smarter than employing rushed tutoring programs during eleventh and twelfth grade. We'd add that states might also "vertically align" their assessment systems so that passing the fourth-grade test means you're on track to pass the high school test. (According to our Proficiency Illusion report, such alignment is missing in a majority of the states we studied.) There's much more to dig out of this interesting study, which you can find here.

Emily Cohen, Kate Walsh, and RiShawn Biddle
National Center on Teacher Quality
July 2008

This is an indispensable companion to a Fordham study released earlier this year, which found that many big-district teacher contracts are ambiguous and that their school leaders may have more decision-making flexibility than they (or others) think. The National Council on Teacher Quality (NCTQ) gathered and codified the bargaining agreements on which Fordham's study was based; now it has done the same for state laws as well and produced this fine report. Its authors found, after analyzing the data, that state laws, and sometimes courts and labor relations boards, often plug the holes that are left open in teacher contracts. In other words, ambiguities in local bargaining agreements don't always yield autonomy because other forces cramp it. Every state, for instance, has a law that sets the number of years a teacher must serve to gain tenure. In addition, state laws frequently set requirements for policies regarding evaluation, dismissal, class size, and salary. California, for example, stipulates "10 different steps that must be taken before the dismissal is finalized, perhaps explaining why just 100 dismissal hearings were heard in the state between 1996 and 2005." Of course, teacher unions play a central role in pushing such laws through legislatures. NCTQ cites the case of New York, where unions pressured lawmakers to embed "a provision in the 2008-2009 budget that made it illegal to consider a teacher's job performance as a factor in the tenure process." This clever ruse tossed a monkey wrench into New York City's efforts to increase teacher accountability, even though the district's teacher contract was silent on the issue. Read much more about it in the report.

Janet Hyde, Sara Lindberg, Marcia Linn, Amy Ellis, Caroline Williams
National Science Foundation
July 2008

That boys are the math whizzes and girls suffer acute equation phobia (hence Barbie: "Math class is tough!") is a common stereotype. It's fueled by evidence that shows that high-school girls take fewer math and science courses than boys, a difference that also serves to explain why boys have historically performed better on standardized math tests in high school. But new findings from a report out of the Universities of Wisconsin and California (Berkeley) are doing their part (along with the other, similar findings that preceded them) to undermine the gender typecasting. Researchers studied achievement data from over 7 million students and found that, for grades 2 to 11, there's no longer a difference between the math performance of boys and girls. Females are still underrepresented in the highest levels of STEM (science, technology, engineering, math) careers, though. Researchers aren't clear exactly why, but they point to the higher variability of boys' scores. For instance, among white eleventh graders in one state studied, the 99th percentile (the very top of the academic curve) contained twice as many boys as girls. So even though average scores are similar between the genders, boys appear to outnumber girls in the ranks of the highest scorers, which may leave them better prepared for high-level math courses in college. Find the short report here (and read it for a fee).