There is potentially a sampling bias in who is reporting data. The author links to a NYT article they wrote that says it is volunteers for the study. It could w

There is potentially a sampling bias in who is reporting data. The author links to a NYT article they wrote that says it is volunteers for the study. It could well be districts without problems that willingly choose to share data.