Burbio's Methodology

Burbio surveys over 1,200 school districts representing over 35,000 schools in 50 states in three different sample groups which generate three different margins for error. Districts are checked every 72 hours for changes. See below for details and reconciliation. 

• For 232 of the most highly populated counties in the US - representing over 28% of the US student population -  Burbio samples school districts representing an average of 90% of the students in those counties - generally it's just one district - and then categorizes the type of learning accordingly. For that group of 232 counties, margin for error at 95% confidence is 0.1% given the high concentration of sampling that occurred. 

• For  228 additional  highly populated counties - that have multiple independent districts that in theory could make different choices about learning plans -  Burbio surveys between three and five school districts per county.  We determined  a  margin of error of 3.8% based on the sample size versus the number of districts.  This group of counties makes up 47% of the US student population. 

• For the remaining counties, (25% of student population)  we surveyed a total of 130 school districts.  There is a margin for error of 8% in this subgroup based on sample size versus total number of school districts. 

• For overall margin of error, we took the square root of the sum of the squares in our initial populations to maintain our 95% confidence interval. The result gave us an overall error of 2.73%.