What 12News Missed (and Misrepresented) in EdChoice’s ESA Report 

EdChoice research is committed to credibility and honesty. We do not change facts for short-term gain, because we know trust and reliability are some of the most valuable assets we can provide to our mission. A 12News article has attempted to besmirch those commitments by smearing our recent report on ESA transactions.  

I use the word “smear” because it attempts to stir mistrust by falsely representing what our study claimed to do, then telling the general public we didn’t stand up to those made-up claims. I also use the word “smear” because it does not build an argument of its own, opting instead for a barrage of disconnected thoughts with negative language.  

The 12News piece never explicitly states its case, so here are its three points as I understand it. I will address them individually: 

  • EdChoice lied about the data it analyzed 
  • EdChoice’s numbers are incorrect because they didn’t include everything they claimed to 
  • EdChoice may have not counted unallowable expenses correctly 

We welcome rigorous, good-faith scrutiny of our methodology, but these accusations are shallow and inaccurate. 

Whether we lied about our data is easily clarified by simply opening our report. I can give you specific pages where each accusation of “misleading” people is disproven (see especially p. 3 and pp. 5-7). We are very forthright and transparent about what data we have and what we use for each section of our analysis. The very title of the report states that we are analyzing the program in multiple stages, and 12News repeatedly compares language in the first stage to analysis in later stages and claims that is misleading the audience. This conflation on 12News’ part would be apparent to any careful reader of the report. 

The second point either neglects the content of our study or fundamentally misunderstands it. 12News seems to assume that we did not include Direct Pay or Reimbursement transactions in our overall estimate unallowable spending rate. We did (p. 13). As we explain in the report, we rigorously analyzed Marketplace transactions, providing the most thorough review of their legitimacy to date. Because we still wanted to provide an estimate of overall unallowable spending, we relied on the best available evidence for unallowable spending rates through Direct Pay and Reimbursement. That evidence comes from the Arizona Department of Education’s (ADE) random audit. Altogether, 1-2% of ESA spending was unallowable in 2024-25. Even if you want to include the phasing-out debit card, the overall estimate of 1-2% is still accurate based on the best available evidence. 

For the third point, 12News neglects to compare EdChoice’s methodology to its own. Unlike the February 12News report, EdChoice explicitly states how it approached judging allowable and unallowable expenses (pp. 6-7). Also unlike 12News, EdChoice details how incomplete ADE guidelines and lack of student identifiers sometimes makes objective judgments difficult. We recommend ADE improve those guidelines. Still, we believe transaction analyses should rely on official definitions of allowable and unallowable expenses as much as possible, as that minimizes bias and more accurately reflects what families and taxpayers should expect from the program. Considering this 12News article treated Legos as unallowable expenses despite them being listed explicitly among ADE’s allowable items, I believe that EdChoice put subjectivity and personal biases aside more than 12News did in its analysis. While some skepticism is understandable when you cannot see the data for yourself, EdChoice has proven itself a more credible source of analysis than 12News thus far.  

12News’ article also ignores other insights from our report altogether. It ignores our insight that the vast majority of unallowable expenses in the riskiest spending outlet—Amazon through Marketplace—still are very small, frequently benign purchases. It ignores our insight that focusing on spending rates is more important than transaction rates, or else you are judging paper towels and pools as equally offensive to taxpayers. It ignores our suggestions for improving administration so the weird purchases that do exist can be further minimized.  

Ultimately, 12News’ smear article appears as muddled as it does because its core belief is just too hard to support. It would be easier for them if EdChoice didn’t care about facts. On the contrary, EdChoice takes great care in producing research with credibility and integrity. We are always open to feedback, but we will stand up to baseless accusations and incomplete reporting.  

We invite both school choice supporters and opponents to read the report to inform themselves about what the best available evidence says about spending in Arizona’s ESA. 

John Kristof

Senior Research Analyst

John Kristof serves as a Senior Research Analyst with the Research and Thought Leadership team at EdChoice. John frequently authors original research and writing, studies school choice programs, designs and analyzes public opinion and user experience surveys, and oversees the organization’s choice program data collection efforts.

John has shared EdChoice’s expertise by presenting research in diverse settings, engaging with audiences ranging from state legislators to education researchers to education reformers. John’s affiliations include membership in organizations such as the Association for Education Finance and Policy, the Association for Public Policy and Management, and the Midwest Association for Public Opinion Research.

Latest from the Blog