A story we published July 20 about student-to-teacher ratios at the San Diego Unified School District contained several major errors. The purpose of this story is to explain what went wrong and what we now know.
To evaluate student-to-teacher ratios, I pulled student enrollment records from San Diego Unified and teacher employment records from the state Department of Education.
Regrettably, each set of records did not represent similar information and this led to incorrect findings. I basically compared apples and oranges.
For enrollment, the records represented the number of students at district-run schools. But for teachers, the records represented the number of teachers at both district-run schools and charter schools.
The distinction is important. Part of the reason for declining enrollment at district-run schools over the past decade is the growth of charter schools. I had specifically aimed to isolate trends at district-run schools, which have been a focus of financial concern in recent years.
By including charter schools, my calculations distorted the number of students per teacher. In the 2010 fiscal year, for example, including charter schools added about 700 teachers to the district for a total of 7,054.
Unfortunately, the errors didn’t end there.
I also incorrectly described how the state employment records represented the number of teachers. They counted each person as one teacher regardless of whether they worked for the district full- or part-time. In the story, I said the records represented the equivalent of full-time teachers.
This distinction matters because part-time employees can skew student-to-teacher ratios. My calculations exaggerated how much time teachers work for the district and therefore how much time they spend with students.
In the 2010 fiscal year, the number of full- and part-time teachers at district-run schools was 6,372. The number of equivalent full-time teachers was 6,158.
By including charter schools and using the wrong employment records, I inflated the number of teachers by nearly 1,000 in the 2010 fiscal year.
Examined over an entire decade, the same kind of errors skewed how student-to-teacher ratios have changed at San Diego Unified. After correcting the mistakes, I found this had led to a wrong conclusion.
My original research showed that the ratio in recent years was still lower than most years in the decade despite recent budget cuts. The implication: With all the hand-wringing over teacher layoffs, we actually have more teachers for our students than we used to.
The new numbers show that the ratios were actually similar or slightly higher than most other years in the decade.
The graphic below compares my new and old findings. The two lines show similar trends, but the new line fluctuates to a much smaller degree. The new ratio hovered around 19 students per teacher since 2000 while the old ratio fell from about 18 to 16.
The graphic does not include a new ratio for the 2011 fiscal year because San Diego Unified officials are unsure whether teacher employment records maintained by the state are accurate. The records show the district cut the equivalent of nearly 1,000 full-time teachers, but district officials say that didn’t happen.
“It makes absolutely no sense,” said Ami Shackelford, who oversees the district’s budget. “Based on our historical reporting, finance is unable to reconcile the state data at this time.”
I’m still trying to figure out what accounts for the discrepancy. San Diego Unified officials are puzzled. A state education official said the state began using a new reporting system that year, but couldn’t immediately tell whether that alone accounts for the difference. If I find out more, I’ll update this post.
Still, the graphic supports a different conclusion than the one I explained in my story two weeks ago. As San Diego Unified’s financial problems have worsened in recent years, the number of students per teacher has grown. Recent ratios have been higher than most years in the previous decade.
Reaching this conclusion didn’t need to happen this way. I failed to fully understand the state personnel records before using them, and I should have checked my research with more experts before publishing it. For some reason which I am still struggling to reconcile, I was sloppy and did not take these normal steps.
I expect more from myself and I know you expect more from me, too. I screwed up this story and I’m sorry. I’ll make every effort I can in the future to prevent this from happening again.
Like VOSD on Facebook.
Value investigative reporting? Support it. Donate Now.
Show 19 comments