I have finally had a chance to look at the study you have pointed out. I have a few more pages to go, but here is what I have seen so far.

First, let’s look at who this group is.

Henry M. Levin, Director
Professor Levin is the William Heard Kilpatrick Professor of Economics and Education at Teachers College, Columbia University, and the David Jacks Professor Emeritus of Higher Education and Economics at Stanford University.

Clive Belfield, Associate Director
Dr. Belfield is an economist working on education policy, the efficiency of education enterprises and the cost-effectiveness of education. He has undertaken research for the British Government, the European Parliament, and the World Bank.

Chad d’ Entremont, Assistant Director, Dissemination
Mr. d’Entremont is a doctoral candidate in Politics and Education at Teachers College. His primary interest is how ideology informs debates centered on school choice and privatization.

Casandra Gomez Arnon, Project Coordinator
Ms. Arnon has professional experience in higher education administration and the non-profit sector. She has a Masters degree in Art and Art Education from Teachers College.

Greg Elaqua, Research Assistant
Mr. Elacqua will begin his doctoral studies in the Comparative and International Education Program at Teachers College in the Fall of 2006.

Brian Gabele, Research Assistant
Mr. Gabele is a doctoral candidate in the Economics of Education Program at Teachers College.

Heather Schwartz, Iscol Research Fellow for Educational Equity and Policy
Ms. Schwartz is an EdD student in Leadership, Policy and Politics at Teachers College, supported by the Iscol Research Fellowship. She is interested in the connections between schools and neighborhoods.

Munyi Shea , Staff Associate
Ms. Shea supports the administration of the Center. She is a doctoral candidate in the Department of Counseling and Clinical Psychology at Teachers College.

Jessica Simon, Research Assistant
Ms. Simon is a doctoral candidate in Economics of Education at Teachers College.

Elisabeth Thurston, Research Assistant
Ms. Thurston is a doctoral candidate in Politics and Education at Teachers College.

Katy Bulkley, Faculty Associate
Professor, Montclair State University

Jeff Henig, Faculty Associate
Professor of Political Science and Education, Teachers College.

Luis Huerta, Faculty Associate
Professor of Education, Teachers College.

Amy Stuart Wells, Faculty Associate
Professor of Sociology and Education, Teachers College.

You’ll notice a common thread. Most of them work at Teachers College. This fact alone makes the their claim of being non-partisan suspect.

Second, let’s look at the study itself.

From page 5 – “Sample sizes for some private school types were relatively small (ranging from 78-224 schools), and reduced slightly further in the full HLM models (57-203 schools). The participation rate for 4th grade conservative Christian schools, for 8th-grade charter schools, and for “other private” schools at both grades did not meet NAEP’s stringent reporting requirements; hence, results for these particular sub-samples should be viewed as only suggestive of patterns that may exist in the entire population of such schools.“

From page 21 – “Missing data impacted the samples for conservative Christian and other private schools (73%-79% schools included in the HLM analysis) more than the samples for the remaining school types (82%-91%). However, the demographics of the reduced HLM samples were remarkably similar to the demographics of the entire dataset, and in our varied analyses we saw no indication that missing data biased the study’s results.

Additionally, there were relatively low numbers of Lutheran and conservative Christian schools at both grades, as well as charter schools at grade 8. Also, the participation rates of conservative Christian schools did not meet NCES reporting standards in grade 4, and these rates were not met for charter schools in grade 8, nor for “other private schools” in grades 4 or 8. Hence, readers are cautioned to view the results for these sub-samples as suggestive, rather than definitive, as to patterns that would hold for the full populations of these schools.”

So, they had missing data, but they don’t believe it will make a difference. They caution that these numbers can only be suggestive and not definitive. Yet, in their opening statement they claim public schools perform better. They don’t say it suggests public schools are better. This is misleading to those who take the time to actually look at the study. Did you read the study Julie?

From page 22 – “Student Demographics.
Binary variables were used for Black, Hispanic, American Indian, Asian/Pacific Islander and female students, as well as for students with “limited English proficiency” (LEP) and those with an “Individualized Education Plan” (IEP)”

From page 23 – “The percentage of students with IEPs was also included in initial models but was ultimately deleted due to lack of significance.”

This describes what they used as demographics. It also states that those on IEP’s were not significant to the study. This leaves me with questions on why you and others keep pointing your belief that these children drastically brings down the scores. According to this study, that is not true.

From page 22, 23 – “Survey information about individual students’ eligibility for free or reduced lunch posed a challenge, given that a disproportionate number of private school administrators reported that the school did not participate in the program (the percentage was over half for some types of private schools). Although these schools generally appeared to be of high SES on other measures, we did not assume that every student in such schools was ineligible for lunch. To preserve data while also being cautious in imputing eligibility status, such students were recoded to “ineligible” only if their school administrator also reported (on a separate question) that less than 5% of the school was eligible for lunch, or if the student reported having at least 5 of the 6 resources that made up the home resources composite (5 was higher than the mean of 4.1 at grade 4, and 4.7 for grade 8, for ineligible students). Overall, the recoded students’ mean achievement was higher than that of the other lunch ineligible students within their school type, providing further evidence that the recoding was warranted and did not negatively bias the performance of private schools. The final “lunch” variable was binary, with 0=ineligible, and 1=eligible for free or reduced lunch.”

This data is not what Kent, myself and others have asked for anyway. We asked for those with IEP’s to be removed. Just because someone receives a free lunch does not mean they cannot be accepted into ACSI schools. I, for one, believe all races and economic status can learn the same. This study based much of their time in explaining why different races and those of different economic status can’t learn the same. What do you believe in this regard Julie? I know you have argued before that this is not racial, but this study you point out says otherwise.

p. 25, 26 – “We began by running a traditional null model, followed by a model with school sector variables only.20”
2020 We acknowledge that this “school sector only” model was rather unorthodox, containing no student-level variables. However, the inclusion of this model was important for allowing comparisons of achievement means by school type before and after demographic/location differences were controlled.

They are admitting to using an unorthodox method to make their comparisons work. This is just another example of why this study is biased and is just more propaganda from the education establishment.

As for “discrediting”, I’ve read the analysis and it it no way sets out to discredit; there’s a big difference between “qualify” and “discredit”. One might think you have an agenda…that wouldn’t be true, would it?

This says it all about this study. They have an agenda and that agenda is to discredit the scores of charter schools and private schools. To them, any number showing public schools are behind means they have to change the analysis.

From a very recent study comparing public to private schools:

A summary:
http://www.dallasnews.com/s/dws/dn/education/columnists/jbenton/stories/062705dnmetedcol.4b6a3351.html

An abstract and the study itself:
http://www.ncspe.org/readrel.php?set=pub&cat=126

Comment by Nan — 6/30/2006 @ 4:09 pm

Don’t get me wrong…I have nothing against ASCI schools; I think they serve a very useful purpose and produce fine results. It’s just that “site admin’s” argument (and your implication) won’t hold water relative to the test score comparison.

“Once again, proof is available to dispell the myth that money will solve the problem. It will not. If money were the issue, then ACSI schools would be providing a lower quality education than Public Government schools. The exact opposite is true; ACSI provide a higher quality education for less money.”

The test score data proves nothing of the sort. When the samples tested have nothing in common, the outcome is irrelevant. I answer your questions when they apply to the argument. I will not respond to your attempts to make something racial out of this. For that, sir, you should be ashamed.

The only area that you haven’t hit but intentionally seem to be walking around is the religious part of the prerequisite. I’m wondering if you have issues with that but just not saying it? I guess what I’m asking is do you have some kind of chip on your shoulder with this subject?

The reason I ask that is because Lennie and I have debunked your “understanding” of selectiveness. Again… if you have any question about Westlake… you need to call THEM up! If you look again at their policy… it seems to be always talking about extreme situations, grades, etc. What I am reading from their policy is that they are a case by case basis. I DO NOT read the best of the best, although they may turn out such students. (i.e.…“Premise Issue” that was talked about above by Lennie) Unless you have gone through the process and KNOW this about Westlake… then it would be good for you to stop making these assumptions about them, that aren’t true.

Again… apples to apples… FUNDING that government schools makes it “apples to bananas”. Let’s really make everything apples to apples, Julie. Let’s fund the child!

As it stands now even with its financial handicaps… “apples to bananas”… private schools are out performing government schools. But being a private school user yourself, you know that and I’m wondering why we are here discussing this? Are you dissatisfied with your private school or is this about something bigger?

Julie, I was hoping that you would answer the questions from above. Again:

“I have some questions, Julie. When you have a challenged student in government school, do they have an Aid that is assigned to them? How many cases can one Aid take? If it depends on the case, break that down for me, please. What if a child is excelling way beyond their years… what would d46 (assuming that you are from d46) do with them? What if they are just really good at Language and normal to challenged in other areas. Then what would they do?”

“Your premise is wrong from the beginning. It appears you believe that all schools teach equally whether government run or private. This premise is false.

Private schools must teach better to compete in the free market with parents. If they did not, they would lose enrollment and eventually go out of business. If they do well, the enrollment increases.

Government schools have no incentive to improve their students academically. They get the taxpayer money no matter what leaving most kids performing at a lower level than those in private schools.

With the 2 models, private schools students become the best, not because they were selected at the start, but because the private school education was better.”

If one could filter the government schools test scores the same way the WCA filters admission, the results might be interesting.

Back to your original premise: “It is conventional wisdom that private religious school test scores are higher than that of public schools. Have you every wondered what the difference actually was in test scores?”

As one who abandoned government schools long ago, I never subscribed to such “wisdom” and never felt the need to “wonder” as you did. In fact, it doesn’t appear wise to me. My selection of private schools was always based on the performance of my options. I didn’t really care how they stacked up against the option I had already eliminated.

Please do not take this as license to stray from the point. Your premis is still invalid.

Then answer this, why can’t public schools produce these results when they have the funding? If you would care to help allow vouchers we’d all see the results and find out once and for all. Care to work on that with me?

Julie said, “only the most educable”. I would still like an explanation on who you think these kids are. Are you trying to say only the least educable attend government schools?

When I see words that you write like “discriminatory” and “diverse”, it is hard not to think about the racial overtones in the their meanings. Would you please explain what you do mean if you did not intend any racial overtones?

If a private school can produce the result you speak of in this post without funding and without selective, exclusionary admission policies, your premise would be valid.

It’s not competition when one team has to play everyone and the other plays only the most educable. That does not dismiss the quality education they provide; just the false premise of this post.

Who are you claiming they are discriminating against? African Americans? Hispanics? who?

Julie said, “The true test of the mettle would be the ability to produce those results with a diverse and difficult sample. ”

That is exactly what Lisa was asking for. That is for tax money to follow the child so the parent can choose the school of their choice. Private Schools will have the funding then to expand services to Learning Disabilities (some already do). We can all see your premise is false and private schools better educate.

Again, I do have to ask what you mean by diverse? Are you saying only rich, white kids attend these private schools?

Your premise is wrong from the beginning. It appears you believe that all schools teach equally whether government run or private. This premise is false.

Private schools must teach better to compete in the free market with parents. If they did not, they would lose enrollment and eventually go out of business. If they do well, the enrollment increases.

Government schools have no incentive to improve their students academically. They get the taxpayer money no matter what leaving most kids performing at a lower level than those in private schools.

With the 2 models, private schools students become the best, not because they were selected at the start, but because the private school education was better.

“It was my understanding from Lennie’s posts that the private schools didn’t require the funding to excel..”

I think you have misinterpreted what Lennie was saying. Let me re-word it so that you can understand. It should read… despite there lack of funding, like public education has, they excel.

Surely, Julie, you cannot think that any private school doesn’t need money? They don’t get our tax dollars like government schools do.

I have some questions, Julie. When you have a challenged student in government school, do they have an Aid that is assigned to them? How many cases can one Aid take? If it depends on the case, break that down for me, please. What if a child is excelling way beyond their years… what would d46 (assuming that you are from d46) do with them? What if they are just really good at Language and normal to challenged in other areas. Then what would they do?

I outlined what they aren’t in my first post. That would mean that the demographics at WCA appear to be “the best of the best”. The definition of selective is clearly stated in WCA’s admission policy; Your site provides the link to it with advocation. Sub par is below the average. If you exclude performers below the average, your score will obviosly be above the average.

“Let’s send some of that government money over to the Private Schools for these programs, following the kids. Then let’s take these tests again. Deal?”

It was my understanding from Lennie’s posts that the private schools didn’t require the funding to excel; and he’s provided plenty of proof to that end. However, if a school received government funding, I highly doubt that their admission policy could be as selective and discriminatory as WCA’s.

I never said anything about fruit…apples, oranges or whatever. I merely stated that when a school admits only the very best, or excludes non-performers, the reasonable person would expect no less than stellar performance. The true test of the mettle would be the ability to produce those results with a diverse and difficult sample.

It really was not my intention make you so defensive. If my logic is in some way flawed, please feel free to provide direction.

You ARE missing something, Julie. Let’s really draw the bottom line. Bottom line is that these (Private School) children would be in your system had something not failed within the government school system.

Julie, what are you assuming the demographics of private school children really are? What is your definition of “selective”? What does “sub-par” mean to you?

Now let’s talk about government mandated funding. Let’s talk about where government mandated funding goes to. Government schools.

Apples to apples, Julie. Let’s send some of that government money over to the Private Schools for these programs, following the kids. Then let’s take these tests again. Deal?

The bottom line seems quite simple; if you exclude the portions of the student body that perform sub-par, your aggregate test score will be above par. The government schools can’t do that as they have to include everyone. If I’m missing something, please fill me in.

No student will be admitted who:

Shows extremely low academic performance as indicated by a standardized testing program, report cards or the WCA Admissions Assessment Test.

Has failed the most recent grade level prior to application.

Has a “present history” of emotional or disciplinary problems.

Has been suspended or expelled.

Has a court record.

Comes from a non-Christian home – a Christian home being defined as: a home where one or both parents have received Jesus Christ as their personal Lord and Savior. (1 Cor. 7: 14)

Has an identified or probable L.D. beyond our scope of expertise.

NOTE: We have a deep concern that every Christian child have an opportunity to have a Christian education, but at our present stage of development, we are unable to commit to meet the needs of all students

http://www.westlakechristian.org/AdmPAndP.htm

I believe that the private schools you refer to have the ability to “tweak” their scores with selectivity in enrollment, and as such, I’m not certain your raw data is valid. I’d like to see some demographics in this regard.