BLR survey finds higher quarantine and case percentages in Arkansas districts without full mask policies; committee seeks refined data

A Bureau of Legislative Research presentation to the Senate Public Health, Welfare and Labor Committee on Sept. 8 found that Arkansas school districts without a full mask policy reported higher percentages of students and employees who were COVID‑positive or quarantined on the survey day.

Julie Holt, administrator of BLR’s Policy Analysis and Research section, told the committee BLR sent a one‑day survey to superintendents on Sept. 1 and received 80 responses, about a 70% response rate. BLR asked districts for raw counts — total students and employees, how many were COVID‑positive that day and how many were quarantined — then calculated percentages and compared school districts that had a full mask policy in place for at least two weeks against those that did not. Holt said the survey compared averages (“comparison of means”) and that three comparisons reached conventional statistical significance; the comparison for student COVID positives did not.

Lawmakers pressed BLR and Department of Health staff on limitations in the snapshot. Senator Shawn Hammer asked whether the results were tested at a 0.05 level; Holt replied that a p‑value below 0.05 means there is fewer than a 5% chance the result is due to random variation, which supports treating the finding as noteworthy. Several members — including Representatives Boyd and Bentley — asked whether BLR controlled for vaccination rates or county case rates. Holt said BLR did not collect vaccination data in the September survey and that correlating district results with county case rates was not part of the initial direction, but she offered to work with the member who requested the study to add those items in a repeat survey.

Members also focused on how quarantines were reported. Senator Bryan Sullivan and others asked whether districts could separate quarantines that stemmed from in‑school exposures from those caused by out‑of‑school contacts. Holt said the first survey did not ask that distinction; she noted some districts said their systems cannot reliably reconstruct a precise point‑in‑time and that the Department of Education and Department of Health did not collect quarantine counts centrally in the form BLR needed. BLR offered to refine future questions to try to capture whether quarantines were the result of in‑school exposure.

Representative Payton questioned why the difference in quarantine percentages (about 8.5% vs. 4.6%) was much larger than the difference in positivity (about 1.4% vs. 1.2%). Committee members and BLR staff explained that quarantine practice and policy — for example, guidance that masked close contacts may not need to quarantine — and differing reporting practices can make quarantine rates more policy‑driven than case rates. BLR said it will work with members on follow‑up surveys designed to test quarantine effectiveness and to collect additional control variables.

Holt emphasized the constraints of a rapid, one‑day snapshot. "We wanted to wait until school had been in session for two weeks to cover that incubation period," she said, and added that BLR will re‑survey districts with refined questions to improve comparability.

The committee did not take formal action on the study but members signaled interest in further BLR work that would include vaccine status, county‑level correlation and clearer separation of quarantine sources.