The Washington Post's Steven Perlstein (via Matt Yglesias) takes a look at how economic, technological and business evolutions in recent decades have negatively impacted civic leadership at the regional level, as regionally focused companies that previously viewed their fortunes as tied to particular regions or communities have disappeared, and the larger multinational corporations no longer have a particular stake in specific regions or communities. Though Perlstein doesn't mention it, this is an issue that is particularly relevant for education: Historically, business leaders have had both an interest in a well-educated workforce and the authority and political clout to lead on education issues at the state or local level. Business leaders played a major role in supporting standards-based reform in the 1990s, and business groups still play a key role on education issues at the state or local level in some places. But the same factors that Perlstein mentions can also impact business engagement on education issues. I spoke recently with an education reform advocate in one "rust belt" city who mentioned that business leaders had played a prominent role in supporting quality education in the city in the 1980s and 1990s, but that this was no longer the case, as the corporations that had once been head-quartered in the city either no longer existed or had moved elsewhere, and the previous generation of leaders with strong commitments to the city were aging and passing away.
January 2012 Archives
January 24, 2012
January 19, 2012
In a recent Ed Week interview, Secretary of Education Arne Duncan signals that he'd like to use the bulk of $550 million in new federal RTT funds to fund an RTT competition for school districts--something the Congressional appropriation language gives him the authority to do.
This makes some sense, given that some districts that are leading the way on education reform suffer the bad luck of being located in states that either didn't participate in RTT, or just kinda suck on education reform. On the other hand, the idea raises a whole host of questions and implications, including:
• Who would be eligible? Would a district RTT competition be limited to districts in states that did not win RTT grants? Or to those in, in Duncan's words "states that are less functional"? (And how exactly is that quantified?) Would only districts larger than a certain size be eligible? Would there be a minimum required percentage of low-income or otherwise at-risk students?
•Do D.C. and Hawai'i get another bite at the apple? Hawai'i, one of the original RTT winners, is also the only U.S. state to have only a single school district. Does this mean Hawai'i would be eligible to compete in a district competition as well? The state has performed so poorly in its implementation of the grant that the Department of Education has threatened to revoke the grant. Similarly, the District of Columbia is, in a grave injustice, not a state, but it is treated like one in federal education policy and was permitted to compete in RTT as a state--and also won a grant in 2010. Would D.C., or the District of Columbia Public Schools, be eligible to compete in RTT as well?
•What about charters and CMOs? The original RTT included provisions to promote charter schools--although not as much as many people thought. But would an "RTT for districts" actually shortchange charters? In some states, charter schools are part of school districts, but in others they are their own LEAs--would charters that are their own LEAs be eligible for to compete for RTT for districts? Related, some CMOs now operate networks of charter schools that are in effect a type of non-geographic school district. A few CMOs are larger than many districts. And some high-performing CMOs are at the forefront on key reforms. Should these CMOs be allowed to compete in an RTT for districts (and do the LEA statuses of their schools matter here)?
The idea of an RTT for districts lays bare some underlying issues with the broader concept of how we think about school districts. A "school district" can mean multiple things: It can mean the particular geographic area that falls within the geographic boundaries of a school district, it can mean a local unit of government established in state law, or it can mean a "local educational agency" (LEA) and the schools that LEA operates. "School districts" typically do not appear in federal education law, but rather "local educational agencies" (which in many states include traditional school districts, charter schools, and some other types of local educational authorities, such as those that run special purpose schools). All of this is complicated by the fact, in our federalist system, "school districts" are the pure creatures of state law--and yet they are a key vehicle for most federal education policies and programs. And the evolution towards increased public education choice and portfolio district models further complicate things.
My other concern here is around wasted energy and the difficulties of running an effective competition. Recall that nearly 1,700 applicants (many of them districts) applied for the i3 competition in 2010, despite very long odds (ultimately, there were 49 winners); and 35 states competed for RTT ELC, including many that were clear long shots. You want people to have good judgment, but when there are money and bragging rights on the table, it can be very politically difficult for district leaders who are eligible to say they won't compete. The last thing we need is for thousands of school districts to apply for an RTT grant, diverting district level capacity and talent--already in short supply--into developing applications for a competition few districts are likely to win, and making it very difficult to run an effective competition. Obviously, the parameters of who can compete matter here. It might also make sense for the Department to design some sort of "pre-qualification" process, in which interested districts could submit a short, streamlined application based primarily on empirical data and yes/no questions about their policies, and, based on that, some subset would be invited to apply for the full competition.
January 18, 2012
Last year, when the administration announced that it was going to devote $500 million in Race to the Top funds to the Early Learning Challenge Grant competition, a bunch of K-12 reform folks asked me about the "pre-k Race to the Top." And I always had to start out by explaining that, "actually, Early Learning Challenge is not a pre-k program, but is more about building statewide systems and improving child care quality across the range of programs serving 0-5 year olds"--which was about when my K-12 focused friends' eyes started to glaze over.....
So maybe this chart (click to enlarge) will make it easier for people to understand. The above chart maps states' scores on the ELC competition (on the Y axis) against their percentage of 3- and 4-year-olds in pre-k, as reported by the NIEER yearbook (on the X axis). States that were ELC winners are in yellow.
As you can see, having a high percentage of students in state pre-k didn't translate to high ELC scores. That's not really surprising: Like I said, ELC is about systems-building and childcare quality more than pre-k, and there weren't very many places where having a lot of kids in pre-k could translate into additional points.
What is surprising is that having a high percentage of kids in pre-k almost seems almost to have been a liability in the ELC competition. None of the states with the highest rates of state pre-k enrollment were among the winners, and none of the winning states had more than 20% of 3- and 4-year-olds enrolled in pre-k. If anything, when you ignore the states that have both no pre-k and very low ELC scores, the chart almost suggests a negative relationship between a state's pre-k enrollment and ELC performance.
January 16, 2012
One interesting result of the the seemingly ever-increasing vogue for evoking Finland in education reform conversations is that it's become a sort of conventional wisdom that "In Finland, teaching is a high-status profession, comparable to doctors," a statement I most recently saw evoked by omnipresent Finnish educator and Finnish-style reform (whatever that means) evangelist Pasi Sahlberg in this Ed Week forum. Then someone controlling edweek's twitter feed asked if teachers should be on the same pay scale as doctors.
Obviously, it makes intuitive sense that if we value teaching, we should respect teachers (although I think it's possible to get a bit simple-headed about what this actually means, too--more on that later). But what do people actually mean when they say "teaching should be a respected profession on par with doctors" or that "In high-performing countries, teaching is a respected profession, on part with doctors"?
I think that the implication of such statements, in the U.S. context, is highly skewed by the fact that the U.S. is a real outlier in terms of the amount of respect and compensation accorded to doctors.
U.S. doctors make significantly more than their peers in other developed countries, and roughly 2.5 times as much, annually, as their Finnish peers. The degree to which American doctors are highly compensated was also brought home for me this week in this NYT analysis of occupations held by people in the "Top 1%"--a group that includes fully 27% of doctors working in clinics and doctors' offices, and from 13-21% of those working in other settings, depending on the setting. And our relatively high rates of physician compensation, in international context, are both a contributor to our unusually high health care costs and something that will likely need to change in order to make our public health care commitments sustainable over the long term.
Now, obviously, money is not the only reflection of the respect in which a profession is held. But it is one important indicator, and the huge differences in medical compensation between the U.S. and other nations both shape attitudes towards the profession and reflect other differences in how the profession and training for it are organized that also affect degree of public respect.
In any case, the point is that, if one takes comparisons of the regard in which Finnish doctors and teachers are held to mean that U.S. teachers ought to be paid or regarded in the same way U.S. doctors are, that's likely to lead to the wrong conclusions, not in the least because the levels of both compensation and respect awarded to doctors are likely to decline in future years.
In fact, for all we hear about the tremendous degree of professional respect awarded to Finnish teachers, they actually are paid less than their American counterparts. Now, again, that's only part of the picture: Finnish teachers are paid less on average than American teachers, but they are paid more relative to their college educated peers, in part because Finland has a lower level of income inequality.
Now, obviously compensation is not the only reflection of the degree of respect a profession is awarded or how widely coveted it is by workers--just ask the many young college graduates here in D.C. competing for low-paid Hill or journalism jobs. But that just underscores how complicated it is to say that "teaching should be a well respected profession on par with doctors," and that people who say that need to be a lot more clear (and possibly also thoughtful) about what they actually mean by it.
January 13, 2012
Now we look at ELC Focus Area E, which was about systems for measuring outcomes and progress. Specifically, Focus Area E asked states to address one or both of two criteria:
• Kindergarten Entry Assessment [E(1)]
• Early Learning Data Systems [E(2)]
This criterion was also worth 40 points, divided evenly among the number of criteria the state chose to address. But there's an important twist!
ELC also included a competitive preference priority linked to Kindergarten Entry Assessment: Specifically, states that answered criterion E(1) and earned 70% of the total possible points could receive an additional 10 "competitive preference priority" points, which both created an incentive for states to address the kindergarten entry criterion and made this section worth more points for states that performed well on it.
Because of this incentive, all but 2 states (Maine and Missouri) chose to address the kindergarten entry criterion, which required states to establish a statewide kindergarten entry assessment to measure children's learning and development at kindergarten entry, across all domains of school readiness and using appropriate measures. 30 states, including Puerto Rico, addressed the data systems criterion.
Among the 35 states (+ D.C. and Puerto Rico) addressing criterion E(1), the median state earned 64% of possible points on this section. State scores ranged from 32-90% of total points possible. Maryland, whose statewide Kindergarten entry system many observers believed the ELC requirements were modeled off of, earned near perfect points on this section. Ohio, Minnesota, California, Florida, Washington, and Connecticut also performed well.
All told, 18 states scored well enough on criterion E(1) to earn the 10 competitive preference points for kindergarten entry assessment--a fact that should perplex mathematically astute readers, since this just over half of those addressing the criterion, and the median score of 64% of total points fell below the 70% cut-off to earn the competitive preference points. The explanation is due to how the ELC application and competitive preference priority were scored. Each ELC application was scored by 5 individual peer reviewers, each of whom assigned their own score to each criterion (with scores often varying significantly among reviewers). The total number of points a state received in each section was based on the average of the 5 peer reviewers' scores. But, when it came to the competitive preference priority, each peer reviewer assigned the state a "yes" or "no" on the priority, based on how the individual reviewer scored the state on criterion E(1). If at least 3 peer reviewers gave the state a "yes", then the state won the competitive preference priority points, whether or not its collective score for the section was above 70% of total points. Thus, several states earned the competitive preference points even though their final score on section E(1) was lower than 70% of total points. These states include: Hawaii, New Jersey, New Mexico, Vermont, West Virginia, and Wisconsin. Since none of these states scored enough points on the other criteria to win the competition, it probably doesn't matter much, but it's still interesting. All of the winning states addressed section E(1) and the competitive preference priority, and all but one (Delaware) scored high enough on this section to receive the 10 competitive preference priority points.
Across the 30 states addressing criterion E(2), the median state earned 73% of total possible points, with a range from 33-96%. Minnesota and Maryland earned the highest marks on this section.
Had the ELC competition been scored solely based on states' performance on Focus Area E, the winners would have been: Maryland, Minnesota, Ohio, California, Florida, Washington, Maine, Rhode Island, Colorado. Minnesota's strong performance on this section helps account for how it was able to win an ELC grant despite relatively weak performance in the core areas of the application.
January 12, 2012
My friend Kevin Carey, writing in The New Republic, offers a bold and novel strategy to create new affordable higher education options for students and reign in the growth of college tuition--which has grown much more rapidly than inflation in recent decades.
This idea is in many ways a large scale version of the approach Kevin and I laid out last year to create new "charter colleges of early childhood education"--but while that proposal was focused primarily on developing cost-effective ways to improve the skills and knowledge of early childhood educators, Kevin's is about overhauling the entire higher education system for a world where it's increasingly necessary for all adults to acquire some sort of postsecondary training and credentials.
This gets to an area of confusion that we noticed a lot last year when talking to folks about our paper: Many people criticized the paper (and us) because they thought we were saying that early childhood educators don't need higher education credentials. But that's not really our argument at all. Instead, we're arguing that early childhood educators absolutely need robust postsecondary credentials--but that the existing higher education system is both too expensive and not-really well designed to provide those credentials. Rather than trying to push early childhood educators into a system that is itself becoming increasing unsustainable, we should capitalize on the public interest in promoting higher educational attainment for early childhood workers to develop new, more affordable, and better new types of higher education options that, because they're designed to meet the needs of early childhood workers, can also help model higher education options for a much more diverse pool of people who need to acquire postsecondary credentials.
Whether or not you agree with our ideas, people who care about improving early childhood education need to be deeply concerned about taming the growth in college tuitions, for at least two reasons. First, skyrocketing tuition makes it more difficult and costly to raise the higher educational credentials of the early childhood workforce. Second, unless we reign in college costs, there's a strong risk that public funding to support higher education affordability will wind up cannibalizing or squeezing out early childhood spending. That's because most policy efforts to date to improve college affordability have focused on providing increasing public funds to help students pay for college. But, with ever-rising college costs that outstrip inflation and government revenues, this strategy can sustain and expand access only if it consumes increasing shares of government revenue--witness the explosion in Pell grants over the past few years. These increasing costs for higher education programs make it more difficult to increase investments in early childhood--even though there's reason to think that spending on early childhood might ultimately be a better long-term investment to improve psotsecondary outcomes.
January 10, 2012
Today we take a look at ELC Focus Area D, which asks states to describe their plans for building a great early childhood workforce. This Focus Area, which was worth 40 points, had 2 components:
• D(1)Developing a Workforce Knowledge and Competency Framework and progression of credentials
• D(2)Supporting early childhood educators in improving their knowledge, skills, and abilities
States had to address at least one of these two components. 12 States chose to address only one component, and 35 (including D.C. and Puerto Rico) addressed both. Of the 12 states that chose to address only 1 component, 8 chose to address workforce knowledge and competencies and credentials, and only 4 chose to address supporting early childhood educators in improving their knowledge, skills and abilities.
• For D(1): The median state addressing this earned 73% of possible points, with a range from 28-100%. Delaware earned perfect marks in this section, and New Mexico earned near-perfect marks.
• For D(2): The median state addressing this earned 70% of possible points, with a range from 29-95%. Oregon, Delaware, and Massachusetts all performed well on this section.
• For D as a whole, the average score was 28.6 out of 40 possible points. Delaware, Oregon, Massachusetts, Washington, California, Maryland, North Carolina, New Mexico, and Wisconsin received the highest scores for this focus area.
As I looked through some of the scoring here, it appears that some states may have had difficulty understanding what the reviewers were looking for in each of the two sections here. D(1) appears to be more about policies, systems, and infrastructure, and D(2) to be more about programs, supports and resources for teacher professional development and higher education, but in practice it can be difficult to segregate the two. it's also somewhat unclear why a chart that required states to state baseline numbers and set targets for the number of early childhood educators holding each credential in the state was included in the scoring for D(2), which dealt with professional development opportunities, rather than D(1), which dealt with a system of credentials.
January 10, 2012
This week, I'm going to be taking a look at ELC Focus Areas--3 sections of the ELC application in which states were allowed to make some choices about what elements to address, as opposed to the 2 Core Areas in which states had to address all components. The Focus Areas and Core Areas each counted for 140 points of the application's total 300 possible, with an additional 20 points available for addressing two "Competitive Preference Priorities" worth 10 points each.
As I mentioned last week, California's unexpected win in the Early Learning Challenge competition owed a great deal to its savvy choices and strong performance in the ELC Focus Areas, so it's worth taking a closer look at these areas, starting today with section C.
Section C of ELC included 4 different components:
- Early Learning and Development Standards
- Comprehensive Assessment Systems
- Identifying and addressing Health, Behavior, and Developmental Needs, and
- Engaging and Supporting Families.
States applying for ELC had to address at least 2 of these, but could address all 4. A total of 60 points available for the section were divided among the number of categories the state chose to address. (So if a state chose to address 4 sections, each would be worth 15 points; if a state chose to address 2 sections, each would be worth 30. Because this complicates comparing state scores, I've chosen to look at them based on the percentage of total possible points earned by a state in each section, given its choices).
Of the 35 states (+ D.C. and Puerto Rico) that chose to apply for ELC:
- 17 addressed 2 parts of Focus Area C
- 13 addressed 3 parts
- 7 addressed all 4 parts
Of the nine states that won grants:
- 2 (Maryland and North Carolina) addressed all 4 parts
- Massachusetts addressed 3 parts
- The remaining 6 states addressed 2 parts
Which parts did states choose to address?
- C(1): Early Learning and Development Standards, was definitely the most popular section to address, with only two states (Arkansas and Iowa) not choosing to address it. All the winning states addressed this section. States generally performed better on this section than others, with the average percentage of points earned (76%), median (77%), and range (37-98%) about 10 percentage points higher than those for other sections. California, Delaware, and Maryland did particularly well on this section, with California earning a near-perfect score.
In general, the states that scored well on Focus Area C are the same states that scored well on the ELC as a whole. Had the application been scored solely on states' performance on Focus Area C, the 9 highest scoring states would have been: Delaware
California, Washington, Ohio, Massachusetts, Minnesota, Pennsylvania, North Carolina, New Mexico. Maryland would have been one of the 9 highest ranking states on this Core Area if it had addressed only the 2 focus areas in which it received the highest scores (Standards and Health/Behavior/Development) rather than all 4. It does not appear that any states that did not win the competition would have won had they chosen to address fewer components of this focus area.
January 05, 2012
When ELC grant winners were announced last month, California's win was one thing that had a lot of folks scratching their heads. How, folks asked, can a state with California's budget woes, that lacks a fully operational statewide QRIS, that just got ruled out for Round 3 of Race to the Top and had to return a federal grant to build out its data systems be a winner in this competition?
I took a look at the scoring results to find out. The primary answer is that California made good choices in the 3 "focus areas" of ELC (where states could choose among a menu of items to address), maximizing its scores in these areas. California scored more total points in the 3 focus areas than any other state: 125 out of 140 total. That made up for its relatively weak showing in the "core" areas--those that all states had to address, including state's past track record and commitment to early childhood and QRIS systems. Had the grant been awarded based on core area performance only, California would have come in 17th, ruling it out for funding, and it ranked 20th among states on the QRIS component.
California's performance on the focus areas was due in part to savvy choices about what components to address: It selected only 2 out of 4 items in "Section C" of the application--and scored very well on both the Standards and health promotion items that it addressed. California's early learning standards received near perfect marks, higher than any other state, and only three states garnered a higher percentage of possible points on the health promotion item. For the other two focus areas, dealing with workforce (Section D) and data (section E), California addressed only 1 of 2 possible items, again playing to its advantage to maximize points. Critically, for Section E, it chose to focus on kindergarten entry assessment and did not cover data systems--which would certainly have been a weakness for the state.
The state also did a good job of presenting its track record in the strongest possible light, enabling it to get more points in those areas than many observers might have predicted.
Now, California's win raises two big questions:
First, because of the way the scoring fell out relative to the amount of funding available, California's award will actually be significantly smaller than what the state requested. But the scores its application received were based on what it said it would do with the full amount of funding. Which components of the state's application will go by the wayside now, and how much does that undermine the plans for which reviewers rewarded them with points?
Second, is this grant still a good deal for California, given the state's ongoing budget woes and the reduced amount of funding it receives from the feds? These grants asked states to do a lot of things, and I'm not at all sure that Tennessee and Louisiana, which would likely have been strong contenders particularly in light of what we saw in the scoring, didn't make the right choice to sit this one out, due to sustainability concerns. What's the chance that the state ends up having to give up this grant as it did the data grant?
(Disclosure: I advised other states (but not California) on their ELC applications).
January 04, 2012
This week I look at interesting takeaways from the state ELC applications and scoring.
States competing for the Early Learning Challenge grant earned points for how well their application addressed the grant's selection criteria, as well as how they addressed two "competitive [reference priorities." But in addition to the selection criteria and competitive priorities, ELC also included an "absolute priority"--a priority that states have to address in order to be considered for the grant.
Three states that applied for the Early Learning Challenge--Iowa, Kansas, and West Virginia--were judged by reviewers not to meet the absolute priority (as was Puerto Rico). It doesn't really matter, since none of these states scored well enough on the other criteria to qualify for funding. But it's still striking that these states went to all the work of completing the grant application but failed to address the absolute priority. None of 5 reviewers thought that Iowa met the absolute priority, only one reviewer though Kansas did, and two reviewers thought West Virginia did (other states had at least one reviewer that did not think they met the absolute priority, but a majority of reviewers thought they did). How did these three states fail to meet the absolute priority?
Reviewers thought that Iowa did not provide enough information to show that its plan would result in improvements in early learning and development outcomes for high-need children. Reviewers of Kansas' application were concerned that the state did not clearly define or identify its population of high-needs children or how its plans would address the needs of this population, and that many implementation activities were not planned before 2014. Reviewers of West Virginia's application also felt the state did not adequately define or identify its population of children with high needs to show how activities would specifically benefit this population.