Monday, July 14, 2014

100% Error Rate on Student Transfers

New Orleans Recovery District Found to Have a 100% Error Rate on Certain Transfer Data

Why is it important for student transfer data to be accurate? Because a school can falsely identify some of their dropouts as transfers and thereby boost their graduation rate. It seems that the New Orleans Recovery District has the highest rate of such "errors" in the state: 100%!

Here is how it works: The graduation rate (the percentage of students who start the 9th grade and graduate 4 years later) for each high school in the state is calculated each year and the percentage of students graduating helps determine the school performance score and also the letter grade assigned to a school. If some students who actually dropped out of school between 9th and 12th grades are misclassified as transfers, they do not count against a school in calculating the graduation rate. I personally do not agree with blaming schools for student dropouts and most of the other stuff that goes into calculating the school performance score. But I also don't believe that schools should be allowed to cheat on such statistics to improve their school letter grades.

How long has this misclassification trick been used? In a report issued by the LDOE several years ago titled: Louisiana's Turnaround Zone: Answering the Urgency of Now, our DOE touted supposedly dramatic improvements in lowering dropout rates in the RSDNO. The dropout rate in grades 9 though 12 had supposedly dropped from 11.4% to only 5.7% by the 2010 school year. If the latest dropout rates were correct the graduation rate in the RSDNO would be 77%. But the drastic drop in this latest DOE report takes it to only 59.5%. I wrote an analysis for Research on Reforms in 2011 questioning these suspicious dropout figures for the Recovery District, but it seems they have been getting away with this scheme for years.

Superintendent White and our Department of Education have taken a small step this year to reduce the cheating occurring in some schools where dropouts are being misclassified as transfers. This year the DOE audited a sampling of the three types of transfer codes most often used to cheat the system. Apparently the sampling was done before the school systems were required to submit their data for the calculation of the graduation rate. Strangely enough the graduation rate of the RSDNO immediately dropped by 8.2%. The DOE calls this sampling a pre-review and has decided to conduct it in all districts in 2014.
"For the class of 2014, the Department will conduct pre-review in every district."
 

I think the DOE decided to clamp down because this practice was becoming extremely obvious to those of us who monitor this data and there was a danger that the whole issue would blow up and be a major embarrassment to the DOE.

But the DOE only took a small step when they could have cleaned the mess up in one decisive action. The DOE could have required all schools to submit documentation of all their transfers to out of state schools, to private schools and to home schooling. Any such transfers can only be considered legitimate if the original school can produce a copy of the request for transcripts from the receiving school (for home schooling there must be a parent permission form on file). That would have cleaned up the graduation rate calculation in one fell swoop. My guess is that if decisive action had been taken this year, the graduation rates would have shown a decline instead of the slight 1% improvement.

Instead, what the DOE did was to require documentation from only  a sampling of all schools in the state prior to the actual calculation of the graduation rate for the 2012-13 school year in hopes that schools would clean up their own data. We have no way of knowing whether schools only cleaned up the sample data or did major housekeeping on all their transfer data.

The two worst offenders discovered by the sampling of transfer data were the New Orleans Recovery District and the East Baton Rouge Parish system. RSDNO could not verify any of the data sampled and therefore had a 100% error rate and EBR had a 73% error rate. The average error rate for the state was 34%.  What do you think would have happened to the graduation rate if every school would have been required to document transfers? Note: It is not necessary for schools to document transfers to other public schools in Louisiana because the Student Information System (SIS) at the DOE automatically picks those up and makes the corrections for all such transfers.

But while we're talking about cooking the books by local school systems to falsely raise their all important letter grades, lets consider one other major cooking recipe. That is the little trick called credit recovery. Several years ago the state decided to waive the minimum clock hours of instruction required for a student to be awarded a carnegie unit of credit for a course in cases where the student had failed a course. Credit recovery allows a student to take a special shortened version of the class where usually the student sits in front of a computer and reviews only the parts of the course he/she failed. At a certain point the student is tested by a monitor or teacher who determines whether or not the student deserves credit for the course. Well the word is out that some school systems use this new tool to hand out course credit like candy to many undeserving students just to boost the school's graduation rate. Guess which school systems are the biggest culprits in this scheme? Some kids have bragged to their teachers that they do not have to worry about passing a required course because the school administrators will let them take a credit recovery course and pass it anyway. Also, in EBR and some other systems, many teachers are "encouraged" to retest their students in failed areas until they pass. There is great pressure on teachers to pass kids with little accountability, thereby enlarging the bogus awarding of carnegie unit credit and graduating more students. Is this what the business community wanted when they insisted that the high school diploma actually mean something?

But wait there's more! Did you know that for the new course choice program John White has waived the seat time requirements allowing students who take such courses from private providers to get full carnegie unit credit. Also the two virtual private academies in Louisiana that are making a huge profit by receiving 90% of the MFP money for each student can also award carnegie credit without regard to instructional time requirements. As far as we know, no one monitors how much time a student spends in class before he/she is awarded credit.

All of the above comes under the heading of school reform under the leadership of John White.

3 comments:

Anonymous said...

Mike, thank you for exposing the various ways teachers and schools have been arm-twisted into playing the "book cooking" game. If schools who are reporting things the "right way" don't bend to the "wrong way", they get public embarrassment for not using the same methods as those that use underhanded methods.

I noticed this at my school in regards to how we report absences...we have certain days where we are directed to not officially report our absences in order to boost our absentee rates. This happens On days where we have mass absences specifically: job shadow day, the lat day before a holiday, heavy rain/storm days, the day of homecoming, final exams when many kids are exempt for having good grades, etc. our absentee rate was "peculiarly low" for our district until our principal was FINALLY told by colleagues this "trick" to avoid this inconvenience to the school score formula. If anyone asks, "the computers were down that day" so there is no absences to report.

Sad, isn't it?

Candyce Watsey said...

I just posted the link to this piece on Diane Ravitch's blog. She has a post today on the skewed graduation rate date in New York. See my post in the comments section.

Candyce Watsey said...

Data, not date.