Three things that will make a school bad: child abuse, homelessness and mothers who dropped out of high school

Conventional wisdom has it that schools with high concentrations of poverty are bad. But when a team of researchers led by Dr. John Fantuzzo from University of Pennsylvania’s Graduate School of Education (Penn GSE) studied every third grader in the Philadelphia public schools, they found strong student achievement in some schools with high concentrations of poverty. The low-achieving schools were ones with high concentrations of homelessness and child abuse. Not only did the performance of the students experiencing abuse and homelessness suffer, so did their classmates. High concentrations of students whose mothers did not complete high school were similarly harmful. The whole grade level had lower achievement.

“It’s not poverty itself that predicts achievement. It’s other risk factors that are associated with poverty,” said Heather Rouse, one of the study’s authors.

“We can’t cure poverty. But we could work on interventions to connect with kids who experience these other risk factors, through department of housing and other agencies,” Rouse added.

Rouse’s study was published in Educational Researcher in Februrary 2014 and presented at the American Educational Research Association (AERA) annual conference in April. She and her colleagues had access to a treasure trove of data on all 10,000 third graders in the Philadelphia public schools during the 2005-06 academic year. In addition to education records, the researchers were able to look at birth, health, public housing and human services records for each student.

About 70 percent of the entire population qualified for free and reduced price lunch, a measure of poverty. Only 26 percent of the population had a mother without a high school degree. The schools that had a 10 percent higher concentration of low maternal education had significantly worse math scores, reading scores and attendance records.

The researchers found that some risk factors, such as high lead exposure or having a teenage mother at birth, didn’t affect student achievement much. But substantiated cases of child maltreatment, which affected 10 percent of the population, and low maternal education were particularly harmful risk factors. Episodes of homelessness, which affected 9 percent of the students, came in third.

Related story citing research by John Fantuzzo: Poverty and education reform — and those caught in the middle

Ranking colleges by salaries of graduates

In March 2014 Payscale came out with its annual rankings of colleges based on how much graduates earn after 20 years. As before, there was a lot of grousing about the data. How can we trust the self-reported salaries of people who happen to choose to participate in the Payscale surveys? The #1 college — Harvey Mudd — used self-reported data from only 112 graduates. How can we judge the economic value of a highly selective college where the majority of the graduates go onto grad school? All MBAs, law degrees, medical degrees, PhDs — all students with graduate degrees — are not included. The methodology ends up elevating schools with big engineering programs because many engineering undergraduates go on to earn good livings without going to grad school. And that helps explain why engineering-heavy Stanford and MIT rise well above Harvard and Yale.

The Obama administration is working on its own college ratings system, “Postsecondary Institution Ratings System,” that’s supposed to be up and running in the fall of 2015. The government held a symposium in February 2014 where data experts offered their ideas on how the government should design it. The New America Foundation pointed out in a comment letter that the government already collects a lot of salary information . If we could tap into the Social Security Administration database, then we could have a pretty good sense of what post-college earnings really are at every institution.

Is it possible to do that without violating individual privacy? And does Social Security even track wages above its salary cap of $117,000?

US students rank better internationally on new problem solving test than they do on conventional math and reading exams

PISA Problem SolvingHere’s a modest test result to bolster the argument of those who say the American educational system isn’t so terrible. On a new creative problem-solving test taken by students in 44 countries and regions, U.S. 15-year-olds scored above the international average and rank at number 18 in the world. That’s much better than the below-average performance of U.S. students on the PISA (Program for International Student Assessment) reading and math tests conducted by the Organization for Economic Cooperation and Development (OECD).

“We think teaching through problem solving is already more developed in the US than in other countries,” said the OECD’s Pablo Zoido, in explaining why US students have higher problem-solving scores than expected.

Still, Asian countries and regions dominate the top 10 spots in creative problem solving, with Singapore, Korea and Japan taking first, second and third place. Canada, Australia and Finland were the only non-Asian nations to make it into the top 10. Shanghai, which topped the PISA charts in math and reading, was relatively weaker in problem solving at number 6.

This new problem-solving test, whose results were released April 1, 2014, is also a PISA exam run by the OECD. For the first time in 2012, a separate 40-minute test was given to 85,000 students on a computer, which allowed for more interactive and unconventional questions than paper and pencil. The questions were more like brainteasers than math word problems. Sample questions were about buying the cheapest train tickets and finding nearby spots for three people to meet.  Here’s a link to sample questions.

Problem solving is important because more jobs in the future will require complex problem solving thinking skills. The accompanying report highlighted teaching strategies in Singapore, Japan and Alberta, Canada that may be helping to be develop these skills (see pp.119-124).

In the best-performing countries – Singapore and Korea – 15-year-old students, on average, are able to engage with moderately complex situations in a systematic way. For example, they can troubleshoot an unfamiliar device that is malfunctioning: they grasp the links among the elements of the problem situation, they can plan a few steps ahead and adjust their plans in light of feedback, and they can form a hypothesis about why a device is malfunctioning and describe how to test it

Top US students did surprisingly well on the problem-solving test, far better than their math scores would have predicted. Indeed the jump, or gap, from math to problem solving skills was one of the largest that the OECD found. That might be a clue as to why top US students fare so poorly on international tests, but still grow up to be creative engineers and designers.

Students in the United States perform better than expected on interactive items, based on their success on static (non-interactive) tasks. Interactive items require students to uncover useful information by exploring the problem situation and gathering feedback on the effect of their actions. To do so, students need to be open to novelty, tolerate doubt and uncertainty, and dare to use intuitions to initiate a solution.

On the other hand, US students trail the best-peforming countries on knowledge-acquisition and knowledge-utilisation tasks on the problem-solving test.




Newly minted teachers in 2012 on average owe $429 a month

Screen Shot 2014-03-26 at 11.42.42 AM

Source: New America Foundation

Should future teachers be taking out massive loans to get their master’s of education degrees?

A March 26, 2014 report by the New America Foundation points out that as much as 40 percent of the $1 trillion in student debt outstanding was borrowed not for college, but to pay for grad school. And some 80% of of the debt incurred by students who finished their grad school programs in 2012 wasn’t for people going into medicine, law or business, but for less profitable professions, such as teaching. Indeed, the average graduate of a master’s in education degree finished with more than $50,000 in debt — $8,000 more than the debt of a typical MBA graduate. That’s a 66% increase in the debts of newly minted teachers since 2004. Another way to think about it is that the average newly minted teacher in 2012 has to pay $429 a month in student debt payments. Half owed more.

The report’s authors predict that these teachers and other indebted graduates won’t be able to earn enough money to afford to pay back their loans.  That will leave taxpayers holding the bag, effectively subsidizing schools of education.

The recent spike in debt for graduate degrees should also focus policymakers’ attention on the lack of loan limits for students pursuing graduate degrees and income-based repayment programs that include loan forgiveness benefits. The debt statistics in this New America report suggest that graduate and professional students are likely borrowing at levels that will lead to substantial waves of student loan forgiveness in the coming years.

A footnote in the report explains how grad school loans work.

The federal government lets graduate and professional students finance the entire cost of their educations with federal loans, for any degree, for any length of time, including all living expenses, regardless of the total cost. Graduate and professional students may first take out $20,500 per year in Unsubsidized Stafford loans, and after that, they can tap federal Grad PLUS loans for the rest of the costs. A series of programs, Income-Based Repayment, Pay As You Earn, and Public Service Loan Forgiveness, that lawmakers enacted in 2007 and 2010 let borrowers repay those loans based on a small share of their incomes, regardless of their debt loads. After 10 or 20 years, remaining balances are forgiven. Note that undergraduates face relatively low limits in the federal loan program, thereby limiting the benefits of loan forgiveness under these plans. That is because a borrower entering repayment with $30,000 in federal loans could have his debt forgiven under one of the repayment programs only if he earns an unusually low income for an extended period of time. Someone with a master’s degree who has $80,000 in debt, on the other hand, can earn an average income for his peer group for most of his repayment term and his payment will still be too low to fully repay the debt. He will have a balance forgiven. 

The New America Foundation used the the PowerStats tool from the National Center for Education Statistics to analyze data from the U.S. Department of Education’s National Postsecondary Student Aid Study (NPSAS). That study is conducted every four years.

Related stories:

Student loans still growing faster than any other debt and now most likely to be 90-days-plus delinquent

U.S. teachers 6th highest paid in the world

US rookie teachers relatively better paid than veteran teachers

Paying good teachers $20K to move to bad elementary schools works and is cheaper than reducing class sizes

Whites and Asians have a disproportionate share of pre-k seats in NYC


Allocation of NYC Pre-K Seats by Ethnicity

Source: U.S. Department of Education civil rights data 2011

These pie charts show that whites make up fewer than 15 percent of the student population in New York City’s public schools, but they took nearly 18 percent of the city’s limited and coveted pre-kindergarten slots in the 2011-12 school year, the most recent year that data was available. That’s according to an interactive civil rights database released by the U.S. Department  of Education on March 21, 2014.

Asians also took a larger share of pre-k seats than their school population would suggest. Meanwhile, both blacks and Hispanics seem to be on the short end of the stick. Blacks account for more than 28 percent of the public school population, but secured fewer than one quarter of the roughly 23,000 public school seats for four-year-olds. Hispanics make up more than 40 percent of the school system, but received fewer than 38 percent of the pre-k slots.

NYC Mayor Bill de Blasio is pushing to expand the school system to offer public pre-k slots to every child in the city. To win a seat in the current allocation system, parents must register online or in person. A lottery then allocates the limited seats available, but the lottery is not random. Younger siblings of current students are slotted first. Geography matters too. Families who live within the school zone of a particular pre-k program are prioritized.

Related stories:

Notable research on pre-kindergarten education

Could New York’s Pre-k plan pit politics and posturing above kids?

Schools struggle to count algebra classes

Politico posted a nice roundup of new data from the U.S. Department of Education on March 21, 2014 that highlights racial inequities in education. They led with the provocative statistic that more than 8,000 three- and four-year olds were suspended from preschool in 2011, but most of the story covers how minorities have less access to important math and science classes and experienced, well-trained teachers.

Deep into the story, there’s a critique of the data that made me pause.

The federal report found that barely half of Georgia’s high schools offered geometry; just 66 percent offered Algebra I.

Those data are just plain wrong, said Matt Cardoza, a spokesman for the Department of Education. The state requires Algebra I, geometry and Algebra II for graduation, so all high schools have to offer the content — but they typically integrate the material into courses titled Math 1, 2, 3 and 4, Cardoza said. He surmised that some districts checked “no” on the survey because their course titles didn’t match the federal labels, even if the content did.

“It’s the name issue,” Cardoza said. “I think schools just didn’t know what to say.”

This is admittedly a digression from a very important point about access to high quality education in the United States, but I wonder if miscounting math courses is rampant. Increasingly popular International Baccalaureate programs call their advanced math classes IB Math HL and IB Math SL. There’s a lot of calculus in those classes, especially the HL. Are they included in the national data that say only half of all high schools offer calculus?

Student loans still growing faster than any other debt and now most likely to be 90-days-plus delinquent

Delinquent Student Loans

Back in 2010, student loan debt outstanding in the United States surpassed outstanding credit card debt. Student loan debt hit the $1 trillion mark in 2013. It’s still the fastest growing consumer debt and a lot of it isn’t getting paid back. That’s according to the most recent consumer debt report by the Federal Reserve Bank of New York released on February 18, 2014.

The Federal Reserve Bank of New York tracks student loans, both those issued by the federal government and private banks, and releases a quarterly report. In the 2013 year-end report (Q4 2013), outstanding student loan balances reported on credit reports increased to $1.08 trillion, a $114 billion increase for 2013.  About 11.5% of student loan balances are 90+ days delinquent or in default. That makes them both the fastest growing and the most troubled type of consumer loan. (Click on the chart above to see a larger version).

I created this time series chart to show how student loan debt has grown over the past decade. (Click on the chart to see a larger version. Student debt is colored in blue.) The data came from this page on the FRBNY website. I used “page 3″ data, but deleted mortgage debt because its overall size dwarfs all the other consumer debt categories and you wouldn’t be able to see the shocking increase in student loan debt. The debt is increasing due to both new loans being taken out by students and interest charges accruing on old loans.

Student Loan Debt Time Series

In a February 2014 FRBNY blog post analysis of  growing student loan debt, Just Released: Who’s Borrowing Now? The Young and the Riskless!, the authors point out that while overall debt fell between 2006 and 2013 (as households retrenched after the 2008 credit crisis), student loans grew.

“First, student loans grew the most of any debt product in both periods (in percentage terms). Second, the growth in educational debt, like that of auto loans, is concentrated among the lower and middle credit score groups.”

Here’s a link to the Federal Reserve Bank of New York Report and to an interactive chart showing how student loan debt is hitting the young and those with low credit scores the hardest.

Universities with 10 largest endowments raise tuition for low income students more than for high income students

This week of March 10, 2014, I was playing around with a new data tool, Tuition Tracker, produced by Holly Hacker of the Dallas Morning News and my Hechinger Report colleague Jon Marcus. The tool allows anyone to see the actual tuition charged to students of different income groups at every college and university in the United States. It lays bare how college tuition sticker prices are like hotel room rates or used car prices. Not many students pay them. The Tuition Tracker uses the four years of available data from the federal Integrated Postsecondary Education Data System (IPEDS) produced by the National Center of Education Statistics. The authors’ big takeaway was that universities are actually lowering tuition for high income students and raising it for low income students.

I wanted to see what was going on at the 10 richest, mostly highly endowed universities . So I created this chart by combining data from the Tuition Tracker with NACUBO’s endowment data.

Tuition increases by family income at the 10 most highly endowed U.S. Universities (2008-11)

Top Endowments University (endowment) 2008-09 Net Price for families below $30k* 2011-12 Net Price for families below $30k 2008-09 Net Price for financial-aid families above $110k 2011-12 Net Price for financial-aid families above $110k low income tuition change high income tuition change
1 Harvard ($30 billion) 423 2880 32145 34980 581% 9%
2 Yale ($19 billion) 5187 7852 28564 36442 51% 28%
3 University of Texas ($18 billion) 8184 8967 20965 23594 10% 13%
4 Stanford ($17 billion) 3120 4501 37286 37357 44% 0%
5 Princeton ($17 billion) 3110 8322 24608 34153 168% 39%
6 MIT ($10 billion) 3400 4995 37187 38469 47% 3%
7 University of Michigan ($8 billion) 6144 5431 20071 23098 -12% 15%
8 Columbia ($8 billion) 4870 12081 39860 36551 148% -8%
9 Texas A&M ($8 billion) 4902 5089 18773 19146 4% 2%
10 Northwestern ($7 billion) 12823 15633 40237 41986 22% 4%

* Harvard data was missing for 2008-09. I used 2009-10 instead. For University of Texas, only Austin campus tuition figures were used. For Texas A&M, only College Station campus tuition figures were used. For University of Michigan, only Ann Arbor tuition figures were used.

Source: Tuition Tracker for tuition data and NACUBO for endowment data.

What I found is that, yes, these highly endowed universities are also hiking tuition bills for the poorest students whose families make less than $30,000. While they’re not lowering tuition for upper income students (except for Columbia University!), the tuition increases for wealthier students have been lower, on average, than for the poor students. I’ve highlighted in red the universities that are hiking tuition bills for low income students more than for high income students.

A few caveats and explanations of the data:

1) This federal IPEDs data is tracking only freshmen students who qualify for federal grants, loans and work study programs. There are many high-income students at these universities who are not included in the high-income tuition price*. A good way of reading the chart is to say, “In 2009 Harvard charged a net-price tuition of $32,145 to freshman students from families who made over $100,000 and who qualified for some sort of federal aid.” This net price is the sticker price minus Harvard’s own financial aid and discounts. It also subtracts grants and scholarships from federal, state and local governments. It does not factor in loans or work-study aid.

Even if we expanded beyond this subgroup of high income students to the entire high income population at these schools, I believe the conclusion still holds true that low income students have suffered higher tuition hikes than high income students. Since the sticker price tuition at Harvard did not increase by 581% during these years, low income students still experienced more tuition inflation than high income students.

2) High income families. Many families making more than $110,000 live in expensive cities. It can be a real struggle to send multiple children to college especially if they are on the low end of those six figures.

3) Low income families. Not all families making under $30,000 are as low income as you might think. Many are divorced and the federal government only looks at the income of the custodial parent. The other parent could be a highly paid banker. The financial aid office of many universities requires that the non-custodial parent submit income information and factors that into the net tuition price charged.

Related stories:

College, federal financial aid increasingly benefits the rich

Correction: The definition of of net sticker price was corrected in this blog post. An earlier version incorrectly said that “(Net sticker price) does not factor in the federal financial aid these students are receiving.” The net price is the sticker price minus grants and scholarships from the institution, as well as grants from federal, state or local governments. So it includes federal gift aid, but not loans or work study.

*Addendum: After I posted, Holly Hacker, the creator of the Tuition Tracker, emailed me that you can get some sense of how large of a share of students we’re capturing through this IPEDS data by looking at this separate NCES data. About 70 percent of dependent undergrads with incomes below $40,000 received federal financial aid in 2007-08, compared with 33 percent of those with incomes above $100,000. Too bad it’s not broken down by college. But safe to say that the high-income tuition figures we’re using apply to fewer than half the high-income students at Harvard.

US rookie teachers relatively better paid than veteran teachers

I had some critical feedback over my March 7, 2014 post about US teachers being the 6th highest paid in the world, according to a UNESCO analysis. One reader questioned the absurdity of a chart that says US teachers make $105 a day when their average annual salary is $54,704. I could imagine many intelligent members of my family asking the very same question so I wanted to step back and explain these purchasing power parity income figures better.

If you were to multiply $105 x 365 days per year, you would get an annual salary of $38,325. That would be equivalent to a pre-tax annual salary of $54,704 with an effective tax rate of 30%  (that’s federal, state and local taxes combined). That sounds about right to me.

If you simply compared this $105 against teacher wages in other countries, you’d end up thinking that the countries with the strongest currencies had the highest paid teachers. For example, the Swiss franc is very strong now. $100 US Dollars buys you only 88 CHF (Swiss Francs). It wouldn’t be fair to say the average Swiss teacher salary of 75,000 CHF*  translates to more than $85,000 or 57% more than a US teacher’s salary of $54,000.  That’s because a bus ride will run you almost $4  and a one bedroom apartment in the city center will  cost $1600/month. It’s a lot more expensive to live in Switzerland if you’re paying in dollars. What you want to know is what is a teacher’s standard of living? Can they afford to buy rib eye steaks and a car?

So economists have figured out a purchasing parity index to compare wages across countries. They start with a basket of goods (food, shelter, transportation, clothes, utilities, etc.) and calculate how much each currency can buy of these things. In the example above, you would calculate how much of that American basket of goods a Swiss teacher can buy on his or her salary. According to UNESCO, Swiss teachers are pretty well off. On their salaries Swiss teachers can buy 24% percent more stuff than American teachers can, on average. Not 57% more.

My critic also pointed to this fantastic OECD chart showing where teachers salaries rank within their own countries.

Teacher SalariesIt shows that in most other countries, teachers earn less than other college graduates. (Spain is the only country where teachers make more than the average college graduate.) But in the US, the differential is huge. American teachers earn a lot less than other college graduates. Italy and Israel have similar salary gaps.

I wonder if this gap has something to do with the inequality of income distribution in the United States. The high end of the income distribution (I’m thinking CEOs, fund managers and orthopedic surgeons) creates an unusually high average salary in the US. Unless we start taking away stock options or imposing higher taxes on high income Americans, we not going to get rid of that. Even if we paid teachers $100k a year.

I also wanted to point out that in the top blue chart, US teachers have the 12th highest salaries in the world, not the 6th highest. It shows that many countries with top ranking students on the PISA test are paying their teachers more. But note the fine print. This blue chart is a ranking of veteran teachers’ salaries after 15 years of experience. The chart I originally showed averaged the salaries of all teachers, including rookies.

When you drill down deeper into the blue chart, you’ll notice that the starting, minimum salaries for American teachers are relatively high. But your salary doesn’t grow much over time. Whereas in Korea, rookie salaries are quite low, but veteran teachers are paid handsomely.

Maybe what matters is paying excellent teachers a lot to stay in the profession.

* It varies by canton. This is a rough approximation.

Related story:

U.S. teachers 6th highest paid in the world







U.S. teachers 6th highest paid in the world

Teacher Salary ComparisonI tend to think of US teachers being relatively poorly compensated in our society, especially when compared to say, private equity fund managers. And I always thought that many developed nations paid their teachers far more than we do in the United States. So I was surprised to see that U.S. public school teachers are the sixth highest paid teachers in the world, according this January 29, 2014 UNESCO analysis (p. 254) that adjusts wages by domestic purchasing power so you can compare different currencies and countries more fairly. (Click on the chart to see a larger version where you can read the country names and $$ more easily).

The other interesting thing is that Asian countries, whose students top the international charts on PISA tests, don’t rank highly on this teacher salary chart. The highest paid teachers in Asia are in South Korea, the 21st county on the list.

Follow up post:

US rookie teachers relatively better paid than veteran teachers


Older Posts »