Buildings don't teach kids

Del Sol - a very nice high school, but this $100 million school building doesn't teach kids

Really. I know it is tough to believe, but buildings do not teach kids. $1 million buildings or a $100 million building, the buildings do not teach kids.

That said, Governor Sandoval proposed sweeping hundreds of millions of dollars out of the construction funds from Nevada's school districts (seriously, they don't need to be building new schools at this point and they can hold off renovating for a few years too). That construction money would be moved to the state's general fund to offset further general fund cuts to K-12 education. Ironically, the general fund revenue helps pay teacher salaries (teachers, btw, do in fact teach kids).

Democrats want to ensure that money, about $300 million, remains available to construct and remodel schools. They argue this will produce several hundred jobs in construction and allow the districts to remodel old school buildings.

This is a rather ironic situation in which Democrats have thrown themselves between a rock and a hard place. You see, Democrats want money for their friends in the construction union, but that comes at the expense of funds for their friends in the teacher union.

In other words, at this point in time, Democrats are wanting to fund jobs for construction workers at the expense of jobs for teachers and school employees. I'm surprised Republicans haven't made this into a PR nightmare for the Democrats by railing on them for misplaced " anti-teacher" priorities.

 Oops!

Still, worrying about union needs puts Democrats in yet another hard place. As much research demonstrates, the interest of both the construction and teacher unions do not always line up with what is best for students.

Harvard gives award for bogus journalism

The faculty at Harvard's Joan Shorenstein Center on Press, Politics, and Public Policy awarded the Goldsmith Investigative Journalist prize to the former Las Vegas Sun writer Marshall Allen for his series on health care in Nevada.

I submitted the following letter to the Shorenstein Center,

I'm writing to express my displeasure that your organization has selected Marshall Allen as this years winner of the Goldsmith prize for investigative journalism. Mr. Allen ruined what otherwise would have been an excellent series by concluding (without an ounce of evidence) that profits were being put ahead of patients.

While it is true that Nevada has more for-profit hospitals than non-profit hospitals, the anecdotal evidence raised by Mr. Allen was not significant enough to warrant such a conclusion. In fact, the most damning evidence Mr. Allen found suggested that UMC (the government run hospital) was the worst. When I looked into the issue further I found data (both quantitative and survey data on patient satisfaction) from the U.S. Department of Health and Human services that showed the quality of care in Nevada was not statistically different than the quality of care in any other state in the U.S. Furthermore, the patient satisfaction survey showed that while the non-profit private St. Rose hospital was favored the most by patients, the non-profit government UMC hospital was favored the least.

In the end I could not find enough evidence for Mr. Allen to reach his conclusion leading me to the believe he simply fit the evidence he found into his own personal bias to draw a sensational conclusion. A good investigative journalist should be above inserting his or her own personal bias in substitution of solid evidence. For this error I am very disappointed that your organization awarded him such a prize.

  I ran down Marshall's bogus conclusions in another blog post.

It is official...again...school vouchers work

Dr. Greg Forster at the Foundation for Education Choice released a report titled "A Win-Win" recently. The report reviewed the empirical data on school vouchers and student achievement and school vouchers and their systemic effects on public schools.

So what are the results?

9 out of 10 random assignment studies show that students benefit from being able to use a school voucher to attend a private school. No study shows that students are harmed by vouchers (one study from D.C. shows that winning a voucher - whether they used the voucher or not - does not result in greater student achievement, but who cares, that same study also showed a 23 point gain in graduation rates for students who used the voucher).

As for systemic effects, that is what happens in the public schools when students leave with vouchers, the evidence shows (overwhelmingly) that public schools respond by IMPROVING! Who knew competition would actually work? (Hint: I did)

21 studies show that public schools improve when faced with competition, 1 study (again D.C.) showed no improvement. Here is the kicker with D.C. - public schools were insulated from competitive pressure because they didn't lose a single penny when kids defected to private schools.

Walter Williams and Reason

Above, Dr. Williams talks with Reason.tv about his life experience, academic work and the role of government in our lives.

If you want to learn more, atch Dr. Williams' 1980s documentary "Good Intentions" on how government programs help maintain poverty in America.

What happened to the antiwar movement?

Reason TV tries to answer one of the leading questions of our time....what happened to the Anti War movement?

The next question that deserves an answer....why does Michael Bay get to keep on making movies?

Las Vegas Sun...still the worst paper in Nevada

I believe this is where Brian Greenspun gets his ideas

The Las Vegas Sun is so bad, in fact, they've been relegated to an "insert" in the Las Vegas Review Journal - in other words, they have to pay people to read them).

I have many gripes with the paper (logical failures)(complete misunderstanding of how the state's budget works) and how, for years, they've bought into the doom and gloom budget scenario of every state agency including education. My gripes mostly have to do with how the Sun editorials and some of their writers reach their conclusions through their collective small intestine. Sorry to be crude, but it is true.

Ok, back to the story....

On what must be a slow day, the Las Vegas Sun has an article titled "Anti-Tax zealots say their tax hikes keep pledge intact."

Democrats in Nevada are looking to raise an additional $1.5-$2 billion to keep Nevada's spending bender going (to maintain the same levels of spending during the boom that ended in 2008). Republicans under Sandoval are looking a little less (mostly through gimmicks and fund sweeps) and want to avoid raising taxes (what would be the third time in ten years).

Several Republicans signed a "No Tax" pledge, but it has been long established that a revenue-neutral tax (that is you raise taxes in one area and lower them in another to offset the increase) is OK!  Some Republicans have proposed reducing exemptions on mining while lowering other taxes. Others think including services in the sales tax is a good idea (and it is) so long as they can lower the overall sales tax rate (our sales tax is awfully darn high).

My problem isn't the fact that the Las Vegas Sun wrote an article about a long settled issue, but that the title of the article refers to people who disagree with tax hikes as "zealots."

A zealot in his own right

Brian Greenspun, the editor of the paper, would probably never go so far as to call people who always demand tax increases as "Pro Tax Nuts" ...perhaps because he is one himself.

Look, I can recognize we disagree on how much the government should spend, but Brian Greenspun doesn't even bother trying to understand his ideological opponents.  Few people on the political right are advocating "no to tax increases" because they want to destroy the future, but because government often incompetently manages our money and our affairs. People who argue against tax hikes believe that the way we run government results in waste, excessive spending, and failed policy goals (see public education). Instead of just throwing more money at the problem lets reform how we address the problem - ie, use resources more effectively.

The way of the future

According to Andrew Coulson of the Cato Institute it cost $50,000 in 1970 (in constant 2010 dollar values) to educate a child from kindergarten through high school graduation. By 2009 that figure jumped to $151,000 per pupil.

A good portion of this growing expense is labor cost. In 1969-70, the average U.S. public school district employed one person for every 22.6 students. By 2006-07 the ratio had fallen to one employee per 15.5 students.

Student achievement, as measured by SAT scores and NAEP scores on math and reading, have remained flat over that same time. Graduation rates and drop out rates appear have not improved much either (in Nevada graduation rates have fallen 23 points in the last decade alone). Math and reading (or even sciences) are not the only subjects taught in school but they are subjects that are easily measured. If these scores are not improving how can we be sure students are learning other skills – especially those skills (creativity, teamwork, problem solving) which they will find useful in their future day-to-day lives as adults in the marketplace?

Today we can clearly see that the additional resources spent on education have not purchased goods or services that improve student learning (at least the learning we can easily measure).

Unfortunately, the system is broken – it is political and bureaucratic and many times unresponsive to the needs of the student. Parents have no say in how or where their child will be educated. Teachers have no control over their classroom and little ability to innovate ways to reach children. Principals have no way to remove the deadwood from their schools (teachers who can’t teach and never will be good at it are unlikely to be terminated).

For example, the left-of-center think tank Center for American Progress found that Nevada kept 99.4 percent of its teachers because tenure rules make it very difficult to get rid of bad teachers (yes, bad teachers do exist). Few, if any, industries terminate just 0.6 percent of their staff a year for poor performance.

In the end, bureaucrats and politicians decide how teachers are hired and allocated, what is to be taught and how, and decide how dollars and resources are spent in the classroom.

The result is an unresponsive, unaccountable public school monopoly that only spends 50 cents out of every dollar is on teachers, textbooks and school supplies. No amount of good intentions is going to fix this problem.

 The Edublob!

Some reforms, like virtual education, have the potential to fundamentally reform the way education is delivered. Computers can assist in teaching students – students watch video lessons, take quizzes and are instructed on how to find the right answer if they struggle. The computer can grade assignments for the teacher and print reports on student achievement. All of this frees up the teachers time so they can focus on individual students and where they struggle.

Think about this for a moment. Instead of a teacher lecturing to a classroom of 30 students on the same material to each students regardless of their understanding, students can learn on the computer at their own pace and receive individualized instruction from the teacher when they need it. Fast learners can learn as fast as they want. Slow learners can keep trying until they master the subject – thus, they can build a solid foundation that will continue to grow as they achieve more and more!

Clayton Christensen, a Harvard business professor, predicted in his book “Disrupting Class” that online learning will grow substantially over this next decade. By 2019, he asserts, 50 percent of high school students in the United States will be learning online.

Globally, online learning will become a behemoth of an industry. Based on Organization for Economic Development and Cooperation (OECD) estimates the 34 members of the OECD spend 6.2 percent of their combined GDP on education. That is $2.5 trillion dollars! How much of that spending will be devoted to online and computer based education? I’m not sure.

In 2007, David Kopf of Global Industry Analysts, projected the E-Learning market would grow to $52.6 billion by 2010. According to Kopf, 60 percent of the market is already in the United States. $52 billion global industry means it is already pretty large and since it is projected to grow by leaps and bounds over the next decade, the future of the online learning market appears bright indeed.

Online education has some early service providers which are doing some pretty cool stuff. Others, like Salman Khan, are providing online education for free.  Innovative adopters of education technology are also seeing some pretty impressive results. Carpe Diem charter school in Yuma Arizona has 1 math teachers for 240 students yet scores 90th percentile on the state's math exam. While Khan, and every other online education provider, will continue to innovate and refine their products I see some flaws which a new service could exploit.

I can’t share my ideas here (though I find them very exciting) because I’m working on a business plan to start a company to provide the very service that is missing in online education. Just a month into the planning I've already secured two experienced advisers (one for finance and one for legal issues and start-up advice).

I have to admit, trying to start your own business requires lots of work, but it is the most fun I’ve had in ages. Regardless of whether my idea succeeds or fails, online and computer assisted education is the way of the future.

The University that cried wolf

As I explained yesterday, UNLV officials grossly overstated the magnitude of past budget cuts. As a result, they've made students, faculty, taxpayers and politicians panic. Its all part of the game to protect their own budget. But this strategy has some adverse risks that, in my opinion, makes it a politically stupid strategy - especially in the information age.

1) If someone actually gets a hold of the revenue numbers its easy to call shenanigans.  The budget figures are public record so obfuscating the truth is an easy way to paint yourself as a liar or intellectual incompetent. I got a hold of the actual revenue figures and I showed that the $49.6 million reduction in General Fund appropriations since FY 2008 was really a reduction of $20 million (or 3.7 percent) in actual total revenue.

2) Claiming devastation doesn't build confidence in the quality of your university. University officials practically claimed Western Civilization itself will collapse if the university budget was cut. Such hyperbole only panics students (revenue) and faculty (service providers), thus reducing their confidence in the system. Of course, UNLV officials also say things like how the cuts will turn them into a "mediocre university." Aside from already being a mediocre university, what business leader would say something like this? Imagine Ford saying "if we turn down these Federal bailouts we'll become a mediocre automaker" - who would buy Ford after that? (Note: Ford is doing better than ever before...)

If I didn't know that overstating budget cuts and engaging in hyperbole wasn't part of the political strategy to protect their budget, I would have to assume UNLV is run by the worst intellectual and business incompetents in America.

3) If you claim the university will collapse with the budget cuts it better darn well collapse or you better hope the economy recovers. If the university doesn't collapse, your past predictions look foolish and you're painted as someone who is prone to overreacting. If the economy doesn't recover, you run the risk of exposing yourself to some very real and relatively large budget cuts. In other words, overstating small reductions now can backfire in the future.

University officials claimed the past cuts (claimed at $49.6 million) would devastate the university yet the universities are still functioning and no one can say with a straight face that quality has significantly declined. In other words, they cried wolf about past budget cuts. Now the public knows that a $49.6 million appropriations reduction (remember this is an overstated figure) had little to no impact on the overall operation of UNLV. Now Governor Sandoval has requested a $47.6 million reduction (a real reduction). IE, the wolf has finally arrived.

In other words, overplaying the past budget cuts backfired.


Being truthful upfront avoids these problems. It also allows you to hold on to your integrity and credibility - something our university (and other government) leaders don't seem too concerned about right now.

UNLV misleads...still

Yup, it is out...but how did it get back in there?

UNLV officials keep trotting out a bogus statistic on how large their budget reductions have been since the recession started. Recently President Smatresk claimed the loss was $49.6 million and in some cases has been expressed in percentages (usually in somewhere between 20 and 40 percent).UNLV officials calculate this figure by adding up budget appropriation reductions from each year.

This is not how you make the calculation and unfortunately no newspaper in Nevada has called UNLV (or NSHE) on the use of bogus numbers and bogus math.

First, the appropriate calculation is (Current Year Spending - Previous Year Spending)/Previous Year Spending. Looking at appropriations allows people to overstate the magnitude of the lost revenue because appropriations are plans for future spending, not actual spending. 

For example, lets say you have a budget this year of $100 and you spent your entire budget. You're feeling good about yourself and you decide to appropriate $200 to your budget next year. You've only planned to spend $100 extra, you haven't actually spent a penny yet. Let's say that windfall doesn't come through so once again you only have $100 to spend (and you spend it). Looking at appropriations ONLY allows you to claim a budget cut of $100 even though you spent the exact same amount.

Blasted personal "budget" cuts mean I still can't buy my dream car. Curse you reality!!!

Its kind of like me planning to buy an Aston Martin again this year only to find that I'm about $120,000 short. Not being able to spend money I hoped to spend doesn't mean I actually had a reduction of $120,000 in spending (I never had that money in the first place).

So if you understand this now it is time to move on to the real numbers.

For this analysis I look at several categories of revenue (I don't have actual spending, but I suspect revenues are fairly close). UNLV officials generally look only at General Fund revenues but that makes up less than 1/3rd of all revenue at UNLV. I'll look at that, but I'm also going to include the "other" revenues which includes student fees along with "other" sources of revenue like grants, contracts, special appropriations and the "other" portions of the budget like the self-supported budgets and capital projects. I will include UNLV, the Dental School, Law School, statewide programs and Athletics (this is fair because UNLV proposed cutting from each of these to meet come somewhat close to, the requested budget from Governor Sandoval). I have also adjusted for inflation to 2010 dollar values.

Figures extrapolated from NSHE historical actual budget revenues

As you can see, even looking just at general fund revenues the difference between the peak in FY 2008 and FY 2011 is just $31 million, not $49.6 million. In other words, Smatresk's figures overstate general fund figures by 60 percent!

Once you start adding in other portions of the budget the reduction shrinks considerably only to grow again, but ONLY when you include capital projects (which is not only maintenance but new building construction). Excluding capital projects is usually a fair way to compare budgets across time because capital project spending fluctuates widely from year to year. 

The fairest comparison suggests that UNLV lost $20 million in revenue since the recession began during FY 2008. This is a reduction of 3.7 percent since the peak in FY 2008. 


In other words, by ignoring the bulk of the universities revenues, UNLV officials overestimate the magnitude of past budget reductions by 123 percent!!!!

The fact is UNLV officials (and many others at UNR and NSHE) have clearly overstated the size of past budget cuts. Embarrassingly, many of them also claimed world ending levels of devastating which have not materialized.  I recognize this is all part of the political strategy - make people panic to build political pressure to encourage policymakers to protect the budgets. Of course panicking students and professors about "devastating cuts" is no way to build confidence in your university system which is necessary to attract and retain students and faculty (this fact makes me believe that education and research isn't the main focus of our universities anymore - spending as much money as possible anyway possible seems to be the main focus).

As shameful this has been, NSHE officials will continue to play this game so as long as the mainstream media fails to do some basic research and math to double check NSHE's figures. 

Btw, has anyone considered terminating the journalism school?

Higher education recap

Over the last month, I've been doing some blogging on the relationship between higher education and the economy. Make no mistake, I'm not anti-education (despite what some university professors have said about me in the press). My argument is that universities have been growing their budget rapidly and spending money on many things besides education and research. It now seems that much of the tuition and tax increases devoted to higher education didn't go to fund actual education and research - this is problematic.

Last month, Chancellor Dan Klaich claimed that the relationship between higher education and economic growth was "indisputable" He and other education insiders claimed we needed to protect the higher education investment because they are the "engine of economic growth." Through them we'd educate the populace, attract high tech jobs, reduce the unemployment rate and grow the economy.

I decided to put this "indisputable" relationship to the test.

I put up a detailed (although simple) analysis on my blog and wrote a column for the Las Vegas Review Journal on the results. This post isn't to go back into the mathematics of the findings only recap the narrative of my findings.

• There was no relationship between the percentage of people with college degrees in a state and unemployment rates in states
• There was no relationship between the percentage of people with college degrees in a state and GDP growth from 2007-2009 (latest data available, coincides with recession. Also no relationship for 2006-09 or 08-09).
• There was no relationship between the percentage of people with college degrees in a state and state budget shortfalls
• States with a Top 100 university experienced statistically higher unemployment rates on average
• States with Ivy League schools experienced a combined net migration rate of -2.5 million from 2000 to 2008 (latest data available). Ironically, defenders of higher education throw out this finding, claiming that educated people move, to which I say, DUH! College educated people move around so if we need them in Nevada we can attract them to the state (college educated people don't move to states because of colleges, but because of jobs).

Basically, the findings above means higher education isn't likely to reduce unemployment, grow the economy, or help close the budget shortfall. In other words, it seems it fails to do any of the things it claims it can do.

Michael Pravica a UNLV physics professor followed up with a letter to the editor (mostly missing the mark) asking why not look at state support for higher education and the results. I did just that in this blog post here. The results?

• State spending on higher education "education and research spending per-pupil" was not correlated with economic growth between 2007-09
• State spending on higher education "education and research spending per-pupil" was not correlated with unemployment rates in 2009 or 2010
• State spending on higher education "education and research spending per-pupil" was not correlated with college attainment rates.

Basically this means spending more tax dollars on higher education does not produce greater economic growth, more employment or even more people with college degrees!

Admittedly the analysis I've done so far is very simple. More complex analysis can and should be done, but so far the people claiming higher education is a silver bullet to our state's economic woes have failed to even do the most basic of analysis.  This is very important because we're talking about spending billions of dollars extra over the next decade on an enterprise which may not provide any additional benefit with that extra spending.

Chuck Norris is on my team

And if you're wondering why Chuck Norris keeps appearing on my blog, well, he appears when I find relationships that are random. Not only did most of the relationships above appear very weak, many were also extremely random. Some relationships were so random you'd get better odds from flipping a coin - that is random, kinda like putting up pictures of Chuck Norris on a blog about education.

Still not related to economic growth

Michael Pravica, a UNLV physics professor attacked my Sunday column in the Las Vegas Review Journal on the relationship between higher education and economic growth and unemployment. Of course Pravica fails to address almost every point I raised.

My favorite line from his letter in the Las Vegas Review Journal,

California is in deep crisis not because of its great universities but because of corporate mismanagement and outright theft from its citizens by companies such as Enron. Despite this, California is still doing far better than Nevada economically.

First, I never said California is in a crisis because of the universities, I said California is in a crisis despite their great universities. Second, its laughable to suggest that Enron, which collapsed in 2001, is the culprit of California's economic woes. Third, the theft is being done by California's own government and public employees. Finally, goodness gracious Pravica, you're a professor you should know better than assume that because one state is doing better than another the difference in results must be the difference in inputs (in this case higher education). Heck, its even likely that the difference between California and Nevada isn't even statistically significant. Besides, it is only RECENTLY that California has done better than Nevada...

It's for you Michael

Next, Pravica focuses on my comment that a top university has little impact on unemployment rates. There are 100 of them in 32 states, sure it is a small sample size, but why, oh why, Pravica did you ignore my regressions on college attainment rates within states? See this blog post for more information. So why did I look at just the Top 100 universities for that first analysis? Well, its simple, UNLV and UNR keep trying to be one and keep claiming that if they can be a top university they can help grow our economy.

Pravica's response did raise one good question - what is the relationship between higher education spending and college attainment, unemployment and economic growth. I decided to check this out. Remember these are very simple regressions and should not be taken as the final answer on the relationship between higher education and the economy. However, it should raise your eyebrow because I failed to find any of the relationships higher education officials claim exist.

ANALYSIS 1:
I'm looking at higher education per-pupil spending (page 29) on education and research ONLY by state and college attainment rates by state. The higher education spending data is from 2006 and looks ONLY at spending related to education and research and so excludes athletics, plant operations, debt repayment and capital projects. This should bias the results in favor of Pravica and all others who think more spending is related to better outcomes.

State per-pupil spending and college attainment rates 2007:  R Square 0.007757 | Adjusted R Square -0.01291 | P-value 0.543056 |  t Stat 0.612564 |Coefficients 0.000142

State per-pupil spending and college attainment rates 2008:  R Square 0.013684 | Adjusted R Square -0.00686 |   P-value 0.4185 | t Stat 0.816056 | Coefficients 0.000191

As I suspected more state spending on education and research in higher education is NOT correlated with more college graduates within a state.

ANALYSIS 2:
I'm going to look at the same 2006 state per-pupil spending on education and research in higher education and compare it to the GDP growth between 2007 and 2009 (2009 is latest data available).

State per-pupil spending and GDP growth 2007-2009:  R Square 0.007759737 | Adjusted R Square -0.012911935 | P-value 0.542978074 | t Stat -0.612682824 |   Coefficients -1.25714E-06

You don't need to know how to read a regression to know what is going on. In fact,  you may have figured it out by the appearance of Chuck Norris on my blog again. Yup, the relationship is random. It is also incredibly weak to the point of state spending on higher education being a useless predictor of economic performance. Interestingly though, the coefficient was negative which means the very weak random likely non-existent relationship we see results in lower economic growth the more state's spend on education.

For giggles I looked at total spending (State spending + student tuition) and GDP growth from 2007-2009. The relationship becomes stronger and less random (but not enough to say a relationship exists) and the coefficient remains negative.

State and pupil tuition spending and GDP growth 2007-2009:    R Square 0.035786 | Adjusted R Square 0.015698 | P-value 0.188265 |  t Stat -1.33472 | Coefficients -2.4E-06

ANALYSIS 3:
I'm going to examine both state and total spending on education and research and compare it to unemployment rates from Dec 2010. I know its not the annual average, but these numbers are fairly close to the annual average. Plus I'm hungry and I couldn't come up with a usable database without having to create my own. I'll return to this later, likely still justified by my original findings.

State spending and December 2010 unemployment:  R Square 0.014889 |  Adjusted R Square -0.00563 | P-value 0.398591 |  t Stat 0.851735 | Coefficients 9.32E-05

Total spending (state + tuition) and December 2010 unemployment:  R Square 0.043793 | Adjusted R Square 0.023872 | P-value 0.144698 | t Stat 1.48267 | Coefficients 0.000144

Again, nothing is going on here. The relationships are very weak and likely random. The relationship does strengthen when you throw in student tuition with the state funding, but it is still not statistically significant. Results were pretty much the same for 2009 unemployment.

BONUS ANALYSIS:
One very interesting regression did stand out however. The more tuition money was paid toward education and research the more college graduates there were in a state. The coefficient suggests that for every $1,286 tuition increases we see a 1 point gain in college attainment rates.

Tuition spending on education and research and college attainment in 2008:  R Square 0.170988 | Adjusted R Square 0.153717 | P-value 0.002837 | t Stat 3.146468 | Coefficients 0.000802

Why is college expensive?

College = Amusement Park?

College tuition is rising faster than inflation - sometimes by several factors higher than inflation. Government spending on higher education is also rising faster than inflation. Heck, even in Nevada (2000 to 2010) higher education spending has risen faster than personal income (Nevada, the fastest growing state in the nation!).

Part of the reason why education is getting so expensive is because higher education is trying to do everything but actually educate students.

Nevada higher education spending growth compared with Nevada personal income growth 2000-2010. Source: Regent Ron Knecht, handout at February 2011 board of regents meeting.

Source: Dr. Mark Perry, George Mason University

Back to school Krugman

Paul Krugman, the now famous left-of-center New York Times columnist - who also has a Nobel Prize on Economics in the field of free trade but no longer really deals in that issue, or reality - penned an article a few weeks back attacking Texas, a low tax-state, as a low performing state. This time he attacks Texas for education quality.

Krugman blasts Texas for a below average graduation rate and high drop out rate. Of course, Krugman's argument is that we need higher taxes to pay for higher quality education.

What Krugman failed to do, however, was control for socioeconomic status and demographics. The white non-Hispanic population in Texas, for example, makes up just 46.7 percent of the state's population. In Wisconsin white non-Hispanics make up 85.7 percent. The national average is 65.1 percent. In other words Texas is a majority minority state and Wisconsin is whiter than the snow in winter.

Krugman was too busy trying to hold on to dear life  his bogus theory that higher taxes result in higher standards of living and better education and couldn't be bothered with a simple analysis.

For many reasons - poverty, bad schools, racism - minority children do worse than white children nationwide. So we can expect a state with a large minority population to do worse than a state with a very white population. This isn't to say minority children can't learn, heck look at Florida.

Blogger Iowa Hawk takes Krugman to task by actually comparing Texas' minority population to minority students in Wisconsin (now the center of the taxation and public union debate) and the national average.

He finds that Texas tends to beat Wisconsin in many of the tests and also outperforms the national average. In terms of drop-out rates, Iowa Hawk found that Texas wins in some areas but not others but does tend to beat the national average.

In other words, White, Hispanic and Black kids in right-to-work Texas tend to do better than in union dominated Wisconsin AND the average student in the nation. Texas has a good, but not great, system of education.

Refer back to my previous two posts on education spending here and here, for more information on how spending is not correlated with student achievement.

Spending not related to proficiency

Last night I posted a response to "Save Our Schools" blogpost claiming that spending does result in better student achievement. SOS was merely repeating some dubious findings from another organization, the Federal Education Budget Project.

Yesterday I showed that, contrary to their claims, spending was not correlated with higher levels of student achievement. In fact, the results were random. Of course, I looked at the average score, not proficiency levels which is what the Federal Education Budget Project used.

The Federal Education Budget Project compares state's "current spending" (which excludes capital projects and debt repayment" with the percentage of students scoring proficient (the second highest possible category on the NAEP).

So how strong is the relationship:

ANALYSIS 1: Just testing the strength of the relationship the Federal Education Budget Project claims exists, exactly how they report the data.

This chart above is from the Federal Education Budget Project. Note, they've zoomed in to make the graph appear more dramatic. They've also failed to include a regression line to show the trend. This leads me to believe they may be trying to mislead the public. Below is my graph - same data, same X and Y axis scale but I've included a regression line.

Finally, a second graph - same data, but I've included the full range of possible X and Y axis values and a regression line. (Remember the possible percentage of students who can score proficient is not 10 percent through 45 percent but 0 percent through 100 percent) Note: be wary of any graph that does not include 0 as the starting point for the X and Y axis.

Not only does a more honest graph show that the seemingly dramatic results are kinda flat, but the math shows nothing may be going on at all.


Current Spending and the Percentage of Students Scoring Proficient:  R Square 0.046311 | Adjusted R Square 0.026848 | P-value 0.129378 | t Stat 1.542535 | Coefficients 0.000477

When looking at proficiency rather than scale score the R2 jumps from 0.034 to 0.046. This is still very weak. The P-value and t-stat still suggest there is not enough certainty to declare that a relationship exists.

ANALYSIS 2: for this analysis I look at current spending and the percentage of students scoring proficient or better (why stop at just proficient when students can also score advanced.

Current Spending and the Percentage of Students Scoring Proficient or Better:  R Square 0.065282 |       Adjusted R Square 0.046207 | P-value 0.070358  | t Stat 1.849933 | Coefficients 0.000639

The relationship grows stronger but the P-value and t-stat still fall short of us being confident enough that the relationship did not occur at random.

ANALYSIS 3: For this analysis I control for poverty, ELL students and students with learning disabilities and compare current spending with the percentage of students scoring proficient or better.

Current Spending and the Percentage of FRL-NonELL-Non-SD Students Scoring Proficient or Better:   R Square 0.004456 | Adjusted R Square -0.01586 | P-value 0.641637 | t Stat 0.468314 |Coefficients 0.000118

Any hope you may have had for finding a relationship between spending and student proficiency on the 8th grade NAEP reading exam should be tossed out the window. Once you control for the differences in student poverty, the number of English language learners and students with disabilities the relationship disappears into random nothingness.

Random, like that...

Spending is NOT related to student achievement

"Save Our Schools" a local group of concerned Nevadans, students and education insiders recently posted a blog titled "Proven: Funding Equals Success in Education"  The group posts the following chart and links to some group named the "Federal Education Budget Project"

From this chart, and others like it (including some scatter plots) the group claims that,

In general, students in states where per pupil spending was highest in 2007-08 outperformed their peers in states with lower per-pupil expenditures on the National Assessment of Academic Progress (NAEP) tests in 2009. In the 10 states that spent the most per pupil, an average of 34 percent of 8th grade students scored proficient in reading on the NAEP and 36 percent scored proficient in math. In the 10 states that spent the least per pupil, an average of only 28 percent of 8th grade students scored proficient in reading and 31 proficient scored proficient in math.

But the claim isn't even supported by the data they site. Sure it looks somewhat impressive when you split the graph up the way it is (above) but when you run a simple regression analysis we see the results are not statistically significant. In fact, they're random.

Yes, random like me putting a picture of Chuck Norris on a blog about education spending

ANALYSIS 1:

For this simple analysis I used the same criteria as above 1) Per-pupil spending by state (I have both current and total spending) and 2) NAEP 8th grade reading scores.

Current Education Spending and 8th Grade Reading:  R Square 0.034755 | Adjusted R Square 0.015056 |  P-value 0.190239 |   t Stat 1.328281 |  Coefficients 0.000473

Basically, current spending only explains 3.4 percent of the difference in 8th grade reading scores between states. This is a VERY WEAK relationship Not that it matters - the P-value and t-stat suggest the relationship is probably random anyway. A p-value of 0.19 means we're only 81 percent certain that the relationship did not occur at random. For scientific analysis we generally require at least a 95 percent certainty to claim a relationship exists.

Total Education Spending and 8th Grade Reading:  R Square 0.018978 | Adjusted R Square -0.00104 |      P-value 0.335034  |  t Stat 0.973607  |   Coefficients 0.000325

Yup, when you include money spent on building schools, renovating schools and paying back the debt, the relationship becomes even weaker and more random. Nothing going on here and I'm not surprised, buildings don't teach children (someone please alert CCSD!!!!)


ANALYSIS 2:

In this second analysis I use the same spending data but conduct a cross-sectional analysis to control for student poverty (eligibility for Free and Reduce Lunch (FRL)), English language learners (ELL) and students with learning disabilities (SD). This comparison looks at the 8th grade reading scores for ONLY low-income students who are also non-ELL and non-SD. Basically, I'm trying to create a fair comparison between states, since each state will have different numbers of low-income, learning disabled and English language learners.

Current Education Spending and FR-non-ELL-non-SD 8th Grade Reading:  R Square 0.002722 |       Adjusted  R Square -0.01763 | P-value 0.716134 |       t Stat 0.365739 | Coefficients 0.000106

Not only does current spending fail to explain even 1 percent of the variance in reading scores once you control for income, learning disabilities and language barriers, but this is an incredibly random relationship. In fact, we can only be 28 percent certain that the relationship isn't random - you get better chances from flipping a coin! Oh yeah, the negative adjusted r-square means spending is a useless predictor of student achievement.

Total Spending and FR-non-ELL-non-SD 8th Grade Reading:     R Square 0.000115 | Adjusted R Square -0.02029 |   P-value 0.940547 |   t Stat 0.074965 | Coefficients 2.02E-05

Again, even weaker and more random. There is NOTHING going on here. This is a very simple analysis and it took me just thirty minutes to collect the data, run the numbers and write this up. I wonder why they didn't bother running the numbers? It is not that hard to do.

CONCLUSION

States which spend more on K-12 education DO NOT see better results on the NAEP 8th grade reading exam on average because the relationships are not statistically significant.

I'm also wondering if Dr. Eric Hanushek knows his name is being used on that site....because he can do a regression analysis in his sleep with his hands tied behind his back.

Finally, put your peepers on this chart:

Education Town Hall

I went to the education town hall meeting at Green Valley High last night and I survived. I'd estimate it was about 90 percent teacher union, judging from the hoots, hollers, and applause.

I also had the opportunity to speak - for two whole minutes!

My key points:

1) We need to set the record straight - $275 million is a large sum of money to take from the Clark County School District, but if that is the money they lose in the biennium it is just 6.5 percent of their CURRENT (not future fantasy land where I drive an Aston Martin) operating expenditures - a very manageable cut (especially for a guy as smart and talented as Dwight Jones). Operating expenditures excludes debt repayment and capital projects.

Spending on K-12 education and student achievement. Any Questions?

2) Stop focusing on how much we spend - the correlation between spending and student achievement is growing weaker and weaker and in some cases is no longer existent. If spending more money produced results then D.C. and Newark New Jersey would have some of the best public schools.

For example, the State of Wyoming (I think I may have said Montana during my 2 minutes - everyone makes mistakes) lost a adequacy lawsuit with the teacher union back in the mid-90s. Per pupil spending rose from about $9,000 per pupil ($2007) to $14,700 per pupil by 2007 - an increase of over 40 percent. Student achievement remained flat.

3) We need to focus on using resources effectively. For example, I pointed to the tiny Baltic nation of Estonia - a nation that beats the average American student in math and science on the PISA exam. For the same cost it takes Nevada to get a child through the fifth grade, Estonia can graduate a student through high school and with a statistically better education! And that is after adjusting for purchasing power parity (cost of living difference between the U.S and Estonia).

Paris France

Mogadishu Somalia... see the difference?

4) Stop the Hyperbole -  Budget cuts will NOT devastate education, send us back to the dark ages or make Nevada a third world country. In fact, if Nevada became a world nation, Nevada's per pupil spending would be on par with France putting us around the top 10 public education spenders in the world!

Besides, why talk about devastating budget cuts when currently less than 50 cents out of every dollar spent on education pays for teachers, textbooks and school supplies? Obviously there are ways to make spending more effective.

Why talk about firing 3,800 teachers when less than half of the district employees are actually classroom teachers? Again, lets focus on effective use of resources, not just spending more and more money.

The hyperbole is meant to scare and anger people and it prevents us from having an adult conversation on how to reform a broken system.

5) The education system is broken so lets fix it - It is not the fault of teachers, administrators or parents so lets stop blaming each group. The problem lies with the system.

• Give parents the right to choose the best school for their child - no more zoning children to schools based on how much house their parents can afford
• Make the dollars follow the student to the school so schools compete for children
• Give principals, not central office bureaucrats - the power to decide how to use those resources
• Give teachers the power to innovate in their own classrooms

This isn't a silver bullet, but it does fundamentally change the incentives which currently cripple the public education school system.