The False Premise of Voucher Hype

“Vouchers are state aid for low-income students that allow them to leave public schools rated C, D or F and attend a private school,” reported Will Sentell in The Advocate (March 14, 2016).  Or so we’re supposed to believe.  If vouchers provided by the Student Scholarships for Educational Excellence Program (Louisiana Revised Statutes 17:4011-4025 are indeed used to help poor kids escape schools with C, D and F grades, it would seem reasonable to believe that there should be some observable correlation between the District Performance Score (DPS) for a school system and the use of the voucher system.  In other words, a district with many C, D and F schools would probably have a lower DPS than a district with a relatively small fraction of schools rated C, D or F.  And if the voucher program worked as hyped, the low-rated district should have a relatively high ratio of voucher students to traditional public school students, following the reasoning that in low-rated districts there is more reason to “escape” the public schools for the hypothetically superior education provided by the nonpublic schools.  And districts with high DPS values should have low utilization of the voucher program.  No such luck:

DPSvsVoucher1

Notice the outlier:  the blue box in the upper-right-hand corner.  That’s Orleans Parish, the original home to the voucher program and home district to 2742 voucher recipients in the first quarter of 2015-16.  Only 12815 attended Orleans Parish School Board-controlled schools (using February, 2015, data).  But it’s an A-rated district, at least according to the Louisiana Department of Education.  So apparently all 2742 students at some point were assigned to a C, D or F school, and then chose vouchers instead, even though the New Orleans miracle has supposedly provided a city chock-full of wonderfully innovative charter schools from which the students can choose.

It’s worth noting that Revised Statute 17:4013 defines an “eligible student” as one who “is entering kindergarten and has enrolled in the local school system in which the public school he would have otherwise attended is located or in a school under the jurisdiction of the Recovery School District.  Each local school system or the Recovery School District shall conduct its annual kindergarten enrollment process and shall report such enrollment to the department prior to the program enrollment process.”  Does anyone believe that really happens?

Also, the public-school enrollment of 12,815 is somewhat deceptive, since many thousand students attend schools operated by the so-called Recovery School District.  So let’s delete Orleans Parish and see if the graph looks better.  But first note the slope of the trend line, suggesting (but with zero statistical significance) that a higher DPS drives more students to vouchers.

DPSvsVoucher2

So now the trend line is at least going in the expected direction, but with an R-squared value of less than 4%, it’s pretty meaningless.  So what about the two outliers that have become more noticeable in the absence of Orleans?  The one on the left is Franklin Parish, a C-rated district (DPS = 79.5) with a public school enrollment of 2827 and a voucher enrollment of 172, most of whom presumably comprise the bulk of the student body at Family Community Christian School in Winnsboro.  The one on the right is St. John the Baptist (B, 85.2, 5717 public, 347 vouchers).  Maybe throwing those two districts out of the analysis will provide us some statistical significance.

DPSvsVoucher3

Nope.  Eight percent is a rather underwhelming R-squared value.  So perhaps we should eliminate the 21 districts that have zero voucher participation and the 15 that have between one and nine participants.  With Orleans, Franklin and St. John the Baptist already gone, that leaves only 30 of Louisiana’s 69 districts in the graph below.

DPSvsVoucher4

R-squared has climbed all the up to 11%.  Wow.  Compare that to R-squared of 60% when comparing at-risk population to DPS.  So how about looking at the D schools?  (There are no F schools among the 69 parish and city school districts; the only F “district” is the non-Orleans portion of the Recovery School District.  Fortunately students escaping the Recovery School District in Baton Rouge have several options within the East Baton Rouge Parish School System to which they could escape.)

There are nine districts with 2015 District Performance Scores that are in the D range:

District DPS Vouchers Traditional
East Carroll Parish 69.5 0 1033
Richland Parish 64.9 19 3122
Tensas Parish 62.0 0 631
City of Bogalusa 61.6 29 1631
Avoyelles Parish 61.5 10 5526
Morehouse Parish 60.4 14 4302
Union Parish 59.9 66 2166
St. Helena Parish 54.9 <10 1064
Madison Parish 52.5 <10 1357

East Carroll and Tensas Parishes have no voucher students.  St. Helena and Madison have between one and nine each.  The nine districts thus have a total of between 140 and 156 voucher students, and 20,832 in traditional schools (ignoring state-chartered schools for this analysis).  That puts the voucher:traditional ratio between 1:133 and 1:149, whereas statewide it’s closer to 1:90.  So students in parishes that the state says are failing (since D is apparently a failing grade in the world of John White) are less likely to “escape” to private schools than students elsewhere.  So let’s quit pretending that the voucher program is anything but a subsidy for a separate an unequal education system that distracts us from working to improve the public system in order to make it the best it can be.

Just don’t tell John White.  He’d probably give the Robertson family half a million dollars (to go with the corporate welfare by which we as taxpayers underwrite their television program) to create a Duck Dynasty Denominational School of Mediocrity in northeast Louisiana to help kill public education in that part of the state.  Sounds ridiculous?  Wait until you hear what the Department of Education already spends money on . . . .  For people wanting to do their own analyses, this file contains the Department of Education’s response to a series of public record requests.

 

 

Advertisements
This entry was posted in Uncategorized. Bookmark the permalink.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s