How well are Utah students doing?

Click photo to enlarge

Are Utah's national test scores really above average or does our unusually high proportion of white students only make us appear to be above average compared with other states? Jay Blain is a math teacher at Cottonwood High School who's teaches a diverse group of students.

• «
• 1
• 2
• 3
• »

At first glance, Utah students seem academically superior to those in most other states.

Utah students have a higher high school graduation rate than the nation on average; they have a higher average ACT score; and they meet or beat national averages on nationwide math, reading, writing and science tests.

But a deeper look at the data shows Utah students might not be doing as well as they appear to be.

When statewide results are broken down by race, Utah's racial groups, including white students, sometimes perform below national averages for their peers, a Tribune analysis shows.

It's a statistical paradox that seems to counter some assertions that Utah schools are besting those in most other states despite the lowest spending per student in the country.

"As we fall farther behind in funding it should be no surprise to anyone that student achievement follows," said State Superintendent Larry Shumway. "Our teachers are doing the best they can, but we aren't providing the support for student learning that we ought to be providing."

The contradictory results seem to defy common sense. But it's a concept well-known to statisticians as Simpson's Paradox, which is when overall results can seem reversed when those results are broken down by group. Large differences in the sizes of groups can lead to the paradox.

In short: Utah does well on national tests partly because the state has a lot of white kids. Even though Utah's white students don't always perform as well as white students nationwide, they still perform better as a group than many groups of minority students. And Utah has an unusually high proportion of white students compared with most other states.

"Utah has a higher proportion of higher scoring groups," said Andrew Kolstad, a senior technical advisor at the National Center for Education Statistics.

That means Utah's white students pull up Utah's averages even though they don't always measure up to other white students nationwide.

It's a paradox that's true of Utah's scores on the National Assessment of Educational Progress (NAEP) tests in reading and math. And it's somewhat true of Utah's high school graduation rates as measured by the U.S. Department of Education.

"Don't rush to judgment when you see that first average score," said Darvin Winick, former chair of the National Assessment Governing Board and now a senior research fellow at the University of Texas at Austin. "Look behind it; it can be extremely important."

--

More needs » Stephen Kroes, president of the Utah Foundation, agrees it's important to look beyond simple state averages.

"Obviously, the danger is deluding ourselves into comfort with where we are," Kroes said. "It's too easy to rest on our laurels and say we're doing fine when in reality, with this population we have, we should be doing much better."

About two years ago, the Utah Foundation released its own report that compared Utah's test results with eight states with similarly high percentages of white students. Utah eighth-graders ranked second from last among those states on math, reading and science NAEP tests. Only Kentucky ranked lower.

"I hear people say that we are really above average, and the truth is, when you look at the scores, we're just pretty close to average," Kroes said.

It's not something that Jay Blain, a math teacher at Cottonwood High in Salt Lake City, finds surprising.

Each day, Blain teaches a mosaic of students. On a recent school day, white, Latino and Polynesian students filled his first period algebra 1 class.

Blain said his students simply have different needs than a classroom of middle-class white students. Decades ago, Utah's education spending might have been enough for Utah's then-homogenous student population, but it doesn't cover the needs of the state's present-day student population, he said.

Minorities now make up more than one-fifth of all Utah students. Experts say minority groups tend to score lower than white students as a group on tests because of unique challenges, such as language barriers, poverty, their parents' education levels, stereotyping in schools, and other factors.

"We're trying to meet the needs of everybody without the resources," Blain said. "If [policymakers] think it's the same state it was even 10 years ago, they're very wrong."

According to Education Week reports, Utah had the highest high school graduation rate in the country in 2004. By 2006, Utah had slipped to 26th in the country.

--

Does money matter? » Blain believes Utah's relatively low per-pupil funding and large class sizes are the main reasons Utah students are falling behind.

"Resources matter," Blain said. "Tell me that it wouldn't matter to have 30 kids in an algebra II class instead of 40."

Utah has the lowest base per pupil spending and the highest student-teacher ratio in the nation. But not everyone agrees that putting more money toward education is the answer.

Sen. Howard Stephenson, co-chair of the Utah Legislature's Education Interim Committee, said part of the reason Utah's per pupil spending is low compared with other states is because Utah doesn't have as many tiny school districts as some states and spends the least on administration of any state.

He said the way to improve Utah education is by attracting more quality teachers to classrooms. But to do that, he wants to boost teacher pay by putting schools on more efficient year-round schedules to save money.

Putting more money toward education would "require higher taxes," he said. In the past, Stephenson has said Utah should be a model for other states when it comes to eliminating waste in education spending.

Rep. Greg Hughes, who co-chairs the Education Interim Committee with Stephenson, said he doesn't think funding is necessarily tied to Utah students' test results.

"If test scores were directly tied to funding then the District of Columbia would have the highest student test scores in America," Hughes said, referring to the troubled Washington, D.C., school system, which spends the third-highest amount of money per student in the country.

Though he said he's not opposed to increasing education funding, Utah simply faces funding challenges other states don't. Utah has the highest proportion of school-age children of any state in the nation, and about 65 percent of Utah land is federally-owned, meaning it can't be taxed for schools, he said.

"I don't know how you ever overcome that," Hughes said.

--

Lies and statistics » Still, Hughes said, it's unacceptable for Utah students to score below national averages when divided by race.

"It's a cop-out to say, 'Well, we fund the least so we should be accepting lower student test scores,'" Hughes said. "There are some hard realities to those test scores you have to pay attention to."

Tony Romanello, a physics teacher at Alta High in Sandy, said he's wary of how test scores can be broken down and used to advocate certain points of view.

He worries that some might point to average state scores as a way to show that Utah's school funding is adequate, but he acknowledges that others might point to the scores divided by racial groups to make the opposite argument.

"There's the saying that statistics never lie, but liars use statistics," Romanello said. "You can make those numbers say whatever you want them to say based on what you're looking for."

Still, he said, sometimes, there are good reasons to look at scores as divided by race and other factors.

"There is value in looking at where our weaknesses are and not masking those weaknesses," Romanello said. "If we're trying to do what's best for kids then don't we want to know how they're really doing?"