Invest1 publisher3 min readPublished
A neuroscientist dates Utah's reading decline to the year its state test required computers
Jared Cooney Horvath told the Senate Commerce Committee that Utah's NAEP scores turned down after 2014, the year a computer-adaptive state test made digital infrastructure a requirement for every school in it.
The Investor · Invest desk
What happened
- Utah's NAEP reading and math results for fourth- and eighth-graders have declined steadily for more than a decade after years of rising scores, according to Fortune's account of the state's testing data.
- Neuroscientist and former teacher Jared Cooney Horvath ties that turn to the arrival of SAGE, Utah's first computer-adaptive state test, which required every school to have digital infrastructure to administer it.
- Chromebooks accounted for more than half of the digital devices sent to schools by 2017, and Horvath said Google sold them into schools to recoup costs on the product.
Compiled by The InvestorSomething wrong?How this is made
Why it matters
- exposure The claim against classroom hardware now sits in a Senate committee record, so a vendor defending district renewals is arguing against testimony a legislator can cite.
- constraint The evidence on offer is a coincidence of dates plus a screen-time correlation across PISA. That caps how far either side can push an outcomes case in a procurement fight.
- decision Districts writing the next refresh order have to decide whether the buying trigger is the state assessment's technical requirement or the learning claim, because those two spend on different schedules.
- precedent Horvath's account, that scores were rising and gaps closing before the devices arrived, invites future hearings to treat the original sales premise as the thing needing evidence.
The two district contracts in Fortune's account price the same purchase very differently. A Florida state appropriations report from 2003 recorded a four-year, $37.2 million lease from Henrico County, Virginia, for 23,000 Apple computers for high school students [13]. That works out at about $1,617 a machine, or roughly $404 a machine-year [20]. Oklahoma City Public Schools signed a $25 million contract with Dell for 10,000 laptops and wireless carts [14], so $2,500 a laptop before anyone separates out the carts [21]. Districts were buying at discounted rates throughout, and by 2024 the United States had spent more than $30 billion putting screens in classrooms [12].
Horvath's Utah claim is really about who wrote the purchase order. "Before 2014, computers were in schools, they were just peripheral," he told Fortune. "After 2014, every school had to have digital infrastructure in order to take the state assessment" [3]. The state's first computer-adaptive test, SAGE, arrived at the point where the reading and math lines for fourth and eighth graders turned down after years of gains [1][2].
The testimony rests on a coincidence in time plus a correlation. Horvath cited PISA results from 15-year-olds in which more time on computers went with worse scores [7]. He also told the committee that this generation is the first in modern history not to outperform its parents on standardized assessments [5][6]. "Evidence indicates that indiscriminate digital expansion has weakened learning environments rather than strengthened them," he said in his testimony, adding that "this is not a debate about rejecting technology" [8]. Fortune's account does not include a comparison state that kept a paper test. A screen-hours correlation also leaves the direction of the arrow open: if weaker readers get assigned more remediation software, the same scatter appears with the causation running the other way.
The installed base being defended here is fifteen years older than the argument. Maine went first in 2002, handing 17,000 Apple laptops to seventh-graders across 243 schools, about 70 machines a school [10][22]. By 2016, 66,000 Maine students had laptops and tablets [11], roughly a fourfold expansion over fourteen years, counting devices in year one against students equipped in 2016 [23]. Chromebooks were more than half of the digital devices sent to schools by 2017 [15], and Horvath said Google sold the laptops into schools to recoup costs on the product [16]. Google did not respond to Fortune's request for comment [17].
In my view the exposure sits in the compliance channel. A district buying devices to administer a state test buys on a deadline; a district buying them because they help learning buys when someone makes the case. Horvath's own description separates the two eras exactly there [3]. The counter-thesis is that the 2010s handed Utah a long list of other candidate causes, and nothing in this record separates them. A committee hearing can move procurement politics well before it settles the science [5]. Horvath also argues the problem was invented: at the turn of the century, he said, achievement gaps across race and gender were closing and scores were rising [18]. "Everything was looking good," he said [19].
If refresh orders hold their volume in states whose assessments do not require a computer, the compliance explanation is wrong.
What to watch
- Any state that moves its assessment off computer-adaptive delivery, and what its districts then do with device refresh volumes.
- Whether the Senate Commerce Committee turns Horvath's testimony into legislative text or leaves it in the hearing record.
- Whether Google answers the claim that it sold Chromebooks into schools to recoup product costs.