Student performance in Louisiana is dropping rapidly. The decline started just about the time John White became superintendent of Education and has accelerated rapidly with the introduction of Common Core in Louisiana schools. Based on a sample analysis of the very meager data LDOE finally released under threat of lawsuit it is clear that not only is student performance not increasing or staying steady, it Is in fact declining, and being masked by a lowering of the number of correct answers required to pass LEAP and iLEAP tests. Please refer to this post by Mike Deshotels and the analysis provided by Herb Bassett for the details. Below is an excerpt from Mike’s blog.
Here is the table supplied by the LDOE as a result of my public records request:
Notice that for 4th grade ELA, 4th grade Math, and 8th grade Math, there was a significant lowering of the percentage of correct answers needed to get a rating of basic. The Science and Social Studies percentages were changed very little from 2013 to 2014.
Would you like to know why such a high percentage of our students (64%) were able to reach the level of Basic this year on a more difficult 8th grade Math test? Herb Bassett calculates that using the same method of guessing described above, 8th grade students this year on the average would need to know only 20.2% of the math material on the test to reach the level of Basic.
What this means is simple terms is that Louisiana students are about 18% less prepared now in 4th grade in English Language Arts, and 28% percent less prepared in Math by the time the reach 8th grade than they were before John White and Common Core started being used in Louisiana schools. 8th grade ELA seems to be about the same. (My guess is this ability evened out as children read more books outside of school. That’s actually how I acquired my skills.) 4th Grade Math takes about a 10% hit in 4th grade, and children’s abilities seem to deteriorate going forward based on the 8th grade results.
This data actually matches up with information being supplied by teachers and parents. I can see why John White would have been so reticent to release this information except under court order and legal proceeding. This is not a local phenomenon. New York has discovered the same subterfuge in their state.
I don’t have magical powers, but I can confidently predict this is something you will find and see happening across the nation, especially in education Reformer infested territories. There is nothing standardized about the testing of Common Core, the only standardization comes in in the form of lying about it.
Proponents of Common Core, and the High Stakes testing required by it, have claimed the comparability of test scores across states will make for meaningful comparisons. To have this meaningful comparison, all states must teach the same curriculum and all must administer identical tests from one of the two federally funded consortiums (Smarter Balanced and PARCC). However neither consortium controls the cut scores; those are entirely in the control of the states. These scores can go up or down as local politics require.
Let me spell this out for you. If you want to show progress in your state you can artificially inflate the scores to show improvement. If you need to make a case for more charter schools and school closures simply lower the scores and take them over and then raise the score back to show that reform worked. That is exactly what Louisiana has done and no doubt other reform markets as well. The actual data shows the Reforms, including Common Core, have had the exact opposite effect, and a very dramatic one.
Even though the proposed tests are identical, even though the curriculum is identical, the actual scores and their meanings are left up to individual states to determine. That fact nullifies the argument for identical standardized tests and even the need for a standardized curriculum. Our scores, our levels of achievement, will not be and are not comparable to scores in other states. These tests are actually the opposite of comparable. NAEP and DEIBELS are national tests that are comparable, and neither of them requires a standardized curriculum nor extensive, expensive, technology intensive, obsessive testing, like Common Core does.
Most people won’t understand why that’s important. Even if some people do understand this shell game, most states will do whatever they can to prevent the public from learning about this practice in a timely manner (as Louisiana and John White has done). This is really a pretty important finding. It confirms anecdotal evidence many parents and teachers have experienced in school and in their homes. This finding drives a stake through the heart of the educational vampire known as Common Core.
Unfortunately Common Core vampires are very real. The corporately funded, Reformer and National teacher union embraced propaganda promoting Standardized Tests linked to Common Core as the cure-all for educational inequality and systemic and endemic poverty is the myth. Really, when you think about it, which is really the more believable reality?
Crazy Crawfish 
Appreciate yours and others’ work over last couple of years to obtain information and provide a more complete picture. For those of us at the ground working with kids, parents, & each other, the full picture and all perspectives/views do matter greatly in implementing good change and identifying what’s working for schools, kids, & professionals to know how to move forward, accurately communicate info to parents, and continue to make progress.
Thank you again. #EthicalLeadership
Deshotel and Basset must have learned about data analysis from Pauls Pastorek and Vallas. None of these numbers mean or prove anything. Of course, none of the analyses from the LDE mean anything either. The “master” file produced by DRC has every response that every student gave on the tests. That and a breakdown of how many points each question is worth would be a starting point for this faux numerical conversation. If one wants to establish trends, the data from 2005 (first year of iLEAP) through 2012 could provide a pretty good historical perspective. Since those tests will never me administered again, the files couldn’t be used for any skullduggery as long as the student identifiers are stripped out, and because of FERPA, the school codes should also be removed. The LDE can also provide the same files for the 2013 and 2014 assessments, since they “can’t be used again” (White). These data sets would establish a set of facts. Basset’s silly “guess analysis” is a waste of time.
The most enlightening set of info would be the various data runs the LDE has done to make the RSD look better. That’s where the 10 point add on for bonus/incentive points was created. Your antireform team has been way off base since their outcry of bias (actually that was Schneider) and inflation when EOC results were first used in accountability in 2012. The rules for EOC use in SPS were written long before Bobby bought BESE. If these experts on every thing education actually followed what goes on in education, maybe their pseudoscience analyses would be more believable.
No, we can’t believe anything White and his flunkies report, but we can’t disprove it with the numbers you dumped in this posting.
Probably best case scenario would be for the public to be able to easily get information from the department no matter who’s running it. When data are used to determine how students are progressing, whether they are promoted, whether professionals keep their jobs, whether communities’ schools close or stay open which regardless of views on that practice does impact parents, kids, & communities negatively; and it’s difficult to get info, questions answered about that info, & the public is not used that nor do they have open lines of communication or a clear reason for that change, at minimum, a culture of distrust is created. And bottom line is that’s what we have…a culture of distrust.
*used to
It used to flow more freely. John White’s first action was to ban communications with the school districts and community except through very narrow, heavily monitored and censored media channels and to restrict data only to allies that supported his agenda. I can’t even reveal all the things I’ve been told he is concealing and lying about because of the small numbers of people who know about these things. His Reform is a sham designed to provide PR for the Reform agenda and him personally.
I do remember our supervisors telling us DOE employees (then) had indicated communication (as we knew it) would no longer be the practice. I didn’t know why…just that that’s what we were told. So disappointed in how this has played out. Initially had a lot of hope…and excitement re: good, sincere change.
It was by John White decree. We were told we would be subject to immediate termination if we violated this directive. I couldn’t even email folks that I worked with in the districts for almost nine years that i was leaving; or risk punitive actions. Then after deactivating my account they reactivated it and kept it open, took down my notice that I no longer worked at the department and who they could contact, and monitored all communications sent to me until I filed a formal complaint upon finding out 7 months later.
As a side note I found out because my rental agent was sending emergency emails to my old address about a leaking ceiling and I didn’t discover or respond to (because I couldn’t) My tenants looked me up on LinkedIn and contacted me about issues directly. (I found out later LDOE employees were monitoring my account, secretly. Your tax dollars ar work, folks.)
I don’t agree with the practice of an employer monitoring people’s email or in any way keeping tabs on people unless there’s a really good reason and it’s been discussed that that’s going to happen and why (e.g., there’s been a problem that’s been discussed and hasn’t been resolved another way). Destroys trust and security for actual employee it’s happening to and any other employee who knows it’s happening leading to culture of fear/distrust/insecurity. I think the most difficult part for me to understand is why promote things like empowerment, autonomy, trusting those working with kids or any employees, etc. if the practice is not something followed consistently by those promoting it. I do believe in those ideas & others, & they can work in practice; but the conditions do have to be there and that will require change. We have to have conditions where we can all be upfront about the challenges in edu from many perspectives (trust=essential); where we can communicate openly with each other across differing perspectives and not have to choose sides; feel it necessary to anonymously blog for fear of retaliation & where we can have flexibility/security to honestly ID issues, things that don’t make sense, ask questions, try interventions at individual, target, & system levels that have some sort of research/evidence supporting the initial implementation of a particular intervention for a particular problem & the flexibility to make mistakes and adjust. And we do need professionals in edu that can do that staying in edu at all levels. That needs to apply to everyone; it needs to be modeled–not used as talking points or as a “do as I say not as I do.” Accounty is fine in and of itself. Needs to be appropriate & needs input from all stakeholders. Testing/measurement is fine in and of itself. Inappropriate use, overuse, or faulty interpretation of results (spin) is not. Micromanagemt, misuse of measurement for accounty purposes, use of fear, intimidation, punishment to “motivate professionals to produce?” in education or in working with kids is not supported as a best practice anywhere I know of (but correct me if I’m wrong). How do any of those practices help to keep HQ professionals in the field working with kids & other professionals or help professionals and policy-makers understand what works at individual, target, & system levels for kids? And if what was presented is not the goal…and the real goal is “better” then why can’t we talk openly about that?Still like many concepts…just more difficult to believe that’s what we are really aiming to accomplish by action, examples, reports provided by so many.
I don’t agree with secretly spying on employee’s email. LDOE does that currently. This is worse. This is activating former employee’s email accounts, removing warning that they no longer work there, then spying on all emails that come in. My account was deactivated my last day while I was staying late to help people. They reactivated and monitored for 7 months. I emailed my old account myself to verify.
We both know that information will never be forthcoming. We also know very few people have this much insight, and fewer would understand the myriad gyrations you put these numbers through. Within that complexity is where anything could hide and arguments could be made virtually ad infinitum. However that level of detail and analysis becomes less important once the cut scores are lowered enough. Let’s use our logic processing skills. As you approach zero the actual mechanics behind this system become meaningless. If passing required getting no points or correct answers, then your complex algorithms are meaningless as are the scores. True? It’s actually basically meaningless the closer you get to 20% on a multiple choice tests with 5 possible answers. Someone choosing “C” all the way down is virtually assured of getting a passing grade with no knowledge behond how to use a pencil to bubble in a circle. Right now Math is at 40% for a passing grade. That means if a child basically only knows 20% of the material, their random answers should get them in passing territory. This is not more rigorous or evidence of academic success. Unless you want to get all crazy and say some of the questions are worth ridiculous amounts of points compared to the others, it really shouldn’t play much of a role here in the final analysis. I think your personal experience with how this should be done is impacting your opinion here. Do you really think “Baggy” knows, understands, or cares about any of the accountabilty mumbo jumbo you spewed all over my nice clean blog entry? 🙂
Jason – Your observation that “someone choosing “C” all the way down is virtually assured of getting a passing grade with no knowledge behind how to use a pencil to bubble in the circle” has yet another sinister outcome. A local parent questioned her child’s passing scores on all four components of the latest iLEAP test. The child is a special needs child and according to parent and teachers could no way pass the iLEAP test. Upon complaint to the local school district an administrator visited the LDE and pulled the child’s bubble sheet. It appears that the child circled all “D”s and acquired a passing score on all four components. THIS IS ALSO A HUGE SECURITY PROBLEM as anyone who knows either which letter is prominent OR that chosing ANY letter consistently will result in passing can inform their student(s) to follow that practice. Just another form of invalidity and unreliability in HIGH STAKES testing. Now here is the really sinister part – for this parent and child – new state law allows certain special needs children to opt out of the high stakes aspect of the test BUT NOT IF THAT CHILD HAS SHOWN THEY ARE CAPABLE OF PASSING THE TEST. Since all special needs students have to take the test, this false positive will special doom for those children and they will lose the opportunity to opt out and be given another path to achieving a diploma. EVIL!!!!! ILLEGAL!!!!!? We’ll see what U.S.Ed has to say about it.
If this strategy already works than this article is actually an understatement of the implications. We are literally wasting money giving tests kids can only fail if they read the questions and are tricked into picking a wrong answer rather than just picking a random one.
Jon, I thought you were going to point out that the guessing figures were not for the LEAP specifically, but for an all multiple-choice test. Since the LEAP contains constructed response (where guessing is not as productive), the figure of certainty required is a little low. The excerpt posted above did not make the context clear, but Deshotels’ full entry makes it clearer.
You go into test equating a good bit, Deshotels brought that up briefly at the end of his blog. I get it; between versions, the raw scores required for the different cuts will vary a little. There is a standard technical process for that. However, the percent required for passing 8th grade math and 4th grade ELA dropped considerably over the three years. While this doesn’t conclusively prove that anything nefarious happened, Deshotels is justified in pointing it out and raising the question. The number of items a student has to be certain about has gotten very low.
As far as “guess analysis”, consider what DRC does. Tests where the student gets less correct than through random guessing are eliminated and not used in the test equating process you describe. DRC does not consider the effect of guessing to be silly.
Different point values for more or less difficult questions does not apply here (in this stage). The LEAP technical summary indicates that two students taking the same version of the test get the same scaled score if one gets only the 20 easiest questions right and the other gets only the 20 hardest questions right. Each question is one point in this context.
Thus, if a student is sure about only a few of the easiest test items – about one in four, maybe one in three in this case – he is likely to guess enough of the other questions right to pass. That is what I want people to understand.
I am interested in your tack on the High School SPS inflation. Do you think that the very high High School SPSs in 2012 were truly justified? If so, what about last year? The grades went down dramatically. Each was the result of changes to the SPS rules, and LDOE provided enough documentation to examine each change. Through that I certainly have learned to deconstruct Vallas, Pastorek, and White.
I agree that the rules for EOC use were in place in 2010. That doesn’t mean that they were well calibrated to produce consistent scores at the time of transition. The 2011 transition baselines vs the 2011 baselines tell that story. 7.6 points off (on average) at that point.
Changes were made after 2010. LDOE schemed to hold down the high end of the 2011 grades in the summer of 2011 by tinkering with the cohort grad rate index, but didn’t get the rule change published in time. They had the grades calculated with the new rule in place but got caught the day before the grades came out and had to change them. In January 2012, they changed it again in a way that set all high school SPSs 2.25 points higher than if they had returned to the original put in place in 2010. (That was at White’s first meeting with BESE as Supt.) If they could tinker with grad rate formulas, why didn’t they align the overall results by adjusting the EOC point values at the same time? They had the necessary data in hand.
You seem to be knowledgeable and may know the answer to this. The technical summary and the interpretive guide indicate that there were 60 multiple choice and 4 constructed response on the 8th grade math LEAP this year. I checked the released test items but found only 3 constructed response and 38 multiple choice. I was under the impression that all the questions were released. What gives?
What gives us that John White lied about the test release. I checked ELA and the released items were not “the test” nor were they anywhere near complete. White was using the “release” to avoid re-using the test while the PARCC controversy was hung up in court.
So he lied like usual, and got away with it, like usual. Jindal doesn’t care unless a camera is pointed at him.
Jason, I generally support your bog posts, but in this last paragraph your reference to the unions needs some correction. This summer both unions came out strongly against standardized testing. I can’t speak for the AFT/ LFT/CFT, but I can tell you that those in the Calcasieu ASs. of educators stand against the Common Core. Please try to watch the broad strokes, so many of us are out here working to remove the CC and frankly all of those unnecessary ed reforms. Thanks, buddy.
I specified “national” unions for that very reason. UFT, AFT and NEA all support common core.
But thank you for asking about that so I could clarify. Readers may not realize there is a difference between national and local teachers unions.
ELECT JASON FRANCE to BESE and we’ll have access to the data!!!!!
Lol. That won’t give me magical powers but I can add my voice to a chorus of requestors and hold whomever is in charge of LDOE responsible for failing to provide data.
Do you know if other grade levels’ scores were manipulated as well???
We only recently got new data after winning a lawsuit against LDOE for hiding public records. I believe it is being analyzed and it appears so. I will report on further details as they become available.