Data behind the online education series

Education News Colorado has received a number of questions about the data behind the recent online education series completed by EdNews and the Rocky Mountain Investigative News Network. We appreciate the close reading of the lengthy stories and we’d like to respond:

Data basics: We looked at three reports a school is required to fill out each year – the October student count, CSAP in March/April and an end-of-year report in the spring. We reviewed data from these reports for a five-year span, focusing on students in full-time multi-district programs. With this information, we were able to track students over time as they moved from one school to another. To be clear – individual student data was masked. Each student had a unique number for tracking but we were not able to identify individual students and made no attempts to do so.

Mobility: The turnover numbers are based on the Oct. 1 student count from one year to the Oct. 1 student count of the next year. These are the reports that funding is based on and the numbers are subject to auditing by state staff.

At-risk:  The end-of-year reports require schools to assign codes to students indicating how the students entered or left a school during the year. For example, different codes indicate whether a student came from a public school in another district or transferred out to a private school. There are more than 20 entry codes and more than 20 exit codes.

Among other things, we reviewed these codes to get at the issue of whether online students are more “at risk” than students in other schools. Online providers we interviewed frequently cited this as a reason for the schools’ poor performances when compared to state averages. We also knew this was an issue from the 2006 state audit of online programs, which noted among its key findings that auditors “were unable to verify the at-risk status of 80 percent of the online students we reviewed.”

A review of traditional measures of “at risk” – such as free/reduced lunch status, English language learners, special needs – showed online and statewide enrollments are not markedly different. In fact, the poverty rate among online students is about 10 percent less than that of students statewide.

We then looked at the codes indicating where students were coming from before they entered online schools.  The overwhelming majority of them are coded as either being continuously enrolled in the program or as coming from another Colorado public school. Fewer than 1 percent carry codes such as entering an online program after dropping out, returning from prolonged illness or injury, or returning after an extended absence, which is defined as missing more than 10 consecutive days of school.

In addition, we looked at the schools the online students were coming from – 290 came from schools that had been identified as alternative education campuses.

Graduation rate/dropout rates: These numbers are taken from Colorado Department of Education spreadsheets showing the rates for all schools. We tallied the most recent data available, for the 2009-10 school year, for the ten largest online programs and compared it to the state as a whole. The dropout rate for the online programs was 12 percent compared to a statewide average dropout rate of 3 percent. The graduation rate for the online programs was less than 20 percent, compared to a statewide average graduation rate of 72 percent. These are straightforward apples-to-apples comparisons of data collected by the state.