LDOE’s Graduation and College Enrollment Statistics in Context

Fortunately tonight I was sent a blog post from a new anonymous Louisiana education blogger  (at least new to me) named ULYankee.   I will reference some of the material contained in this entity’s blog post, because they seem to know what they are talking about, but I will build off of what they have written.

According to LDOE our grad numbers are going up and up and into the stratosphere.  What’s more, our students enrolling in colleges is going up too!  However, John White is the Prince of Education Deception and these press releases are mostly BS.

First let me explain how this all works.  There are several data systems these numbers come from. 

The Student Information System (SIS) collects basic data on enrollments, and exit codes (including exit code 04 (which gets updated to a “Y” on a statistical aggregation file) which means the student graduated – theoretically.)  The information from SIS gets used to generate the cohort graduation and dropout rates as well as the graduate counts.

Calculations used below were based on the following criteria: 1) Total Graduates were calculated based on LDOE Student Information System (SIS) End-of-Year data where "GraduateFlg" = "Y"; 2) Number enrolled 1st Fall after graduation was based on Enrollment Beginning Date <= 10/31/2014.

The Student Transcript System (STS) collects actual class info, grades, Carnegie Units earned, and can determine if a student has met all the class related requirements to graduate.

In the past, when I worked there, we would check the numbers and differences between these systems.  There should not be a large discrepancy.  STS is probably the more reliable, but SIS is easier to use for calculations and more timely in some cases and easier to update.  In the times when we cared about data accuracy, LDOE would audit the use of codes to determine if they were being used correctly.  In a recent small audit of RSD, LDOE determined that 100% of the codes they used were not used correctly and not documented.  One RSD school in New Orleans reported as much as 50% of their student cohort transferred out of the country. (If those students were really dropouts – which is likely -  by transferring them out of the country on paper they boosted they graduation rate and lowered their dropout rate by removing them from the equation.)  LDOE did nothing with their findings and did not later their figures for the years they audited.  LDOE started incorporating the Graduates in the School Performance Score calculation about 4 years ago.  Immediately after doing this the graduation rate jumped 4%. 


The magic of expectations works here just as it did in Atlanta and DC. 

The increase in 2010 was not real.  It was not real in the sense that 4% more students suddenly decided to graduate because they knew it was important to their school’s performance score. 


This increase was due to better data.  It was largely due to school districts entering data correctly for the first time because it impacted their school performance scores, and we gave them several opportunities to fix their data they did not have in years past.  It was certainly not due to staying the course on Common Core and more rigor – which had not been introduced yet.  (It might also have been related to switching over easier to pass End of Course testing instead of the pervious Graduate Exit Exam.)

Now our grad rate is creeping up every year and we have a large discrepancy between STS and SIS LDOE does nothing about.

So, according to yesterday’s 2014 college going report, 38,785 students graduated from Louisiana’s public schools last year. But according to the Regents’ STS Core 4 report, there are 38,326 public school graduates. Hmm, that’s a difference of 459 students, or a little over 1%. Not a huge number, but they should match. There’s even more of a problem with last year’s data – LDoE reported 37,655, where I only see 36,424 public grads in the Regents’ report, for a difference of 1231 students or a 3.3% difference.

So over the past two years graduate counts may be overstated by over 1600 students.  LDOE did not tie their numbers to the Student Transcript System.  We are showing quite a few more graduates in SIS (which only requires LEA’s to submit a single value of 04 on an exit record to identify a student as a graduate.)  Like the miracle workers before John White in Atlanta and Washington DC, LDOE does not audit or verify numbers they want to see increasing.  They don’t care how schools get those increases, as long as they get them so they have the pleasing numbers to back their Reform initiatives. 

The increase in college enrollments is a function of a new, more complete, matching system LDOE was planning to implement right after I left and right after John White arrived.  In previous years we identified fist time freshman based on students we found at the Board of Regents (our State College Board.)  We called this our First Time Freshman or FTF Report.  The Board of Regents only had data on instate colleges and universities – and not even all of them.  The new method is using  the National Student Clearing House.

College enrollment data is provided by the National Student Clearinghouse which includes more than 3,600 colleges and universities, enrolling 98% of all students in public and private U.S. institutions. For more information, see http://www.studentclearinghouse.org/about/

The increase in first time college enrolling students is not Common Core.  It is not rigor, or charter schools, or higher expectations, or magic pixie dust.  The increase from years past  is entirely due to including the rest of the students from Louisiana that Board of Regents did not have, using a better source file, and including the out of state student records we never had access to.  These are not apples to apples comparisons. 

This is more smoke and mirrors.  This is John White claiming credit for doing something that is quite likely entirely a factor of using a more expansive and inclusive matching method. 





6 thoughts on “LDOE’s Graduation and College Enrollment Statistics in Context

  1. Thanks so much for this! And as we know from public higher ed, we are NOT seeing these increases in public college and university enrollment. Some were up and some were way down mostly as a function of the new admission criteria that were fully implemented (almost) last fall. But overall new FTF enrollment shave been flat and overall enrollments down by about 10,000 students since Jindal took office. So where did the increased numbers go? According to this, no where because the increases aren’t real to begin with. And a large percentage of students that John White is saying is qualified for college really isn’t (43%) because an 18 ACT composite does not get you state university admission in Louisiana.

    1. Yup. But he likes to say college and “career” ready for a reason. I suspect we will find a great many more of our students going to 2 year colleges and technical institutes than in years past. Our best students from Baton Rouge High, Ben Franklin, Luscher, LSU lab, go out of state at much much higher rates because they can’t find quality options in state anymore thanks to Jindal’s incessant college cutting. Our best and brightest are fleeing and droves. I can’t blame them.

  2. Another tidbit. . . graduation rates are calculated using a 4-year cohort. However, LDOE sends all 12th grade graduates from SIS in for college-enrollment data. The graduates in May 2014, for example may not have belong to the 2014 cohort. They could have been comprised of 2012, 2013, and possible 2015 cohorts. Without using the actual cohort graduates for college-going data, it’s comparing apples to bananas! even if we account for the discrepancy between SIS and STS, there is never going to be a time when the graduates in the grad rate are the ones reflected in the college-going data because it’s not the same group of students.

    1. Wow. I had not considered that, but you’re right, Squashie. It’s even more complicated that i thought. That complexity was one of the reason’s i was kinda against the cohort calculation. There is no easy way to validate it or directly use it with other stats, but people will anyways.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s