At the very least the WI DPI should be embarrassed by the fact it takes this long to release the test results. Personally I find it outrageous. I had an email exchange with DPI officials concerning this long delay and the loss of value, this is an excerpt from part of that response (italics mine):
... The WKCE is a large-scale assessment designed to provide a snapshot of how well a district or school is doing at helping all students reach proficiency on state standards, with a focus on school and district-level accountability. A large-scale, summative assessment such as the WKCE is not designed to provide diagnostic information about individual students. Those assessments are best done at the local level, where immediate results can be obtained. Schools should not rely on only WKCE data to gauge progress of individual students or to determine effectiveness of programs or curriculum.
Does anyone else find the fact that the state issues WKCE results to individual students surprising given the above statement?
Anyhow - the WKCE results were published in the WSJ and on the Cap Times website. As in previous years they show Monona Grove near the top in Dane County, with Waunakee out in front. We do well against tough competition, Dane County schools average the highest in the state.
Ed O'Connor notes that the MG 3rd Grade scores as printed in the paper appear to be incorrect:
The numbers reported in the WSJ (79% Reading/74% Math) are inconsistent with reports we have received from CTB McGraw-Hill and reports provided on the DPI WINSS site. These other sources both show our percent of students at Proficient/Advanced to be 84% for Reading and 81% for Math.The DPI website agrees with Ed's numbers.
I will do a little analysis when the DPI posts the results files and get back to you when I can. I neither want to put too much emphasis on a flawed assessment nor speak about what it shows without proper analysis.
28 comments:
I guess it doesn't matter all that much what it shows. These are the tests that will be used to eventually measure 100% compliance with No Child Left Behind. We have a long way to go, even if the 3rd graders are in the 80's and not the 70's, as published. I understand that short of 100% our district will begin to lose funding. Better hurry up and find some more energy savings.
Our district will be in the same boat as every other district in the state. There is an interesting study published by UTA that shows that the Texas NCLB rules are leading to dramatic increases in the drop out rate. Schools have no incentive to keep lower performing students and plenty of incentive to let them go. Despite the laudable goals of NCLB, this is clearly a flawed implementation.
If the WKCE is meant to be used more in aggregate, fine. I understand your frustration with the timeliness issue, but then how about we concentrate on the stuff we can take care of in-house? For instance, I posted a while back about the class time lost to PLAN testing at the high school when the students' were given absolutely no counseling or help in intrepeting their results. My kid has taken this test twice now and still nothing from guidance. So, if you can't use WKCE effectively, how about using the other test data we have? Otherwise, let's stop the non-mandatory testing until we have a plan in place of how to use it!
I would love to know how many school days are lost to in-service and testing.
Especially funny was the kids getting out this week early.
When they are done with school next week.
If you think a lot of time is lost to testing and in-service, ask some time about how many times per year, on average, your elementary child has a substitute teacher. Then ask if the majority of these days are due to illnes of the regular teacher.
I so completely agree with the last post. I would LOVE for the school board to take on the issue of the number of subs that are used for meetings, etc. I am very concerned about this issue, and feel very strongly that subs should be used for teacher illness, or family illness, and not for meetings. There are lots and lots of inservice days, and more of those should be dedicated to the teachers working with their peers, as opposed to the kids having a sub so that the teachers can "meet". Peter, can this be a board goal, or can we form a committee to look at this. May seem like a small thing, but it is not.
Or at least inform parents of the dates that are planned ahead for teacher absences. Some of these must be know quite a bit in advance. Dentist appointments, etc could scheduled on those days. Parents who are lucky enough to be able to take a day off of work could use those days to take their child out of school for an enriching activity rather than have them linger with a sub.
So Peter -
Do you mean to say that you aren't all that concerned about our test results because the government will have to cave in and fund all of us under-performing school districts?
Sure sounds as such from your post. As a school board member representing our communities, I would have rather heard/read you comment on what your plan is to get these test scores up! It sounds as if you are not worried in the least, because you are convinced that the government's theory is flawed.
As we are all aware, it doesn't matter what you think if the government is dictating the funding.
What is your plan to increase test scores among our students, particularly that third grade group that everyone is fumbling to refute the scores for?
The scores are abominable, no matter how you look at them. Isn't this the class that the superintendant thought should increase in class size last year? Pretty sure I heard about a group of parents approaching the board with concerns because of the low income and special education concerns with this class.
Maybe those parents with concerns were right? Seems that this is the lowest performing group in our district. Larger class sizes probably didn't help the situation.
Can you please provide direct feedback on the likelihood of class size increases impacting the test scores as reported by the Wisconsin State Journal?
Thank you for your time and service to our communities. So many of us are so concerned about the future of our great district. How can we get these scores to improve? Obviously, taking away a class section and increasing class sizes didn't work out as well as Dr. Schumacher promised us it would.
Hopefully in the near future, you become less concerned about the timeliness of the information, and more concerned about what the test scores are indicating for the future of our district, and the students who are depending on this public education afforded them by the tax dollars that their parents and neighbors pay every year.
Thanks in advance for your answers. I wish that the rest of the Board would weigh in on these very important issues. Surely your answers here in this forum don't represent the thoughts and feelings of everyone else on the Board.
Concerning the Monona 3rd grade class: First of all the WKCE tests are administered in the beginning of November, just over 2 months into the new school year. The scores largely reflect the student’s progress from the previous years, in this case what happened in the small 2nd grade classroom). This is one of the reasons that the lack of timeliness of the data is such a problem – people are led to believe these are 3rd grade scores when they are not. Second, the results reported for 3rd grade include all 3rd grade students in the district, not just the Monona students. The Monona students are less than 22% of the total 3rd grade scores reported. The conclusion that the class size decisions for the Monona 3rd graders can be judged from this data can simply not be supported.
Of course comparing the 3rd grade scores from one group of students to the next is not really a valid way to measure progress for those students anyway. Instead we need longitudinal data for specific students compared to other cohorts: how did these particular students improve from 2nd to 3rd grade and how does that compare to the goals? The MAP data tells us that, I hope that the district can assemble and report these kinds of data in the near future. I have heard from some of the teachers of these 3rd graders that they have had “great” year- I expect their test scores will reflect this.
The research on the subject is pretty clear: results from the Tennessee STAR project (and other studies of small class size) shows that the major effect of small class sizes is in the first one or two contact years, and by 3rd grade any effects are negligible. (Word, E.R., Johnston, J., Bain, H.P., et al. The state of Tennessee’s Student/Teacher Achievement
Ratio (STAR) Project: Technical report 1985–1990. Nashville: Tennessee State Department of Education, 1994.)
MG's test scores have consistently been at the 90th percentile in the state, reading scores are 2nd in Dane County, Math scores are 4th. I don’t think that makes us "under performing" or "abominable". If we are going to have this discussion lets at least start with an accurate portrayal of the situation. The 3rd grade numbers reported in the WSJ are wrong, so why should we have a response based on them? Referring directly to the DPI official website for the correct numbers is certainly not "fumbling".
As for plans for improving test scores lets look again at the numbers since 2003 (last comparable WKCE test) WI reports 5 years of data for 4th 8th and 10th grades:
4th grade Math: increased from 87.6 to 95.3%
4th Grade Reading: increased from 83.4 to 87.6%
8th Grade Math: 79.2 to 91.6%
8th Grade Reading:88.8 to 93.2%
10th Grade Reading: 76.9 to 86.1%
On average MG has shown math and reading scores that increase faster than 90% of WI school districts. The district has consistently delivered increasing test scores as measured by the WKCE and the results are evidence we are on the right track through elementary and middle school.
The state is not going to have to “cave-in” and fund under performing schools, they are going to have to continue to fund schools like ours that have consistently delivered year over year improvements in test scores and out perform 90% of the districts in the state.
That said I would also say that NONE of these scores are good enough and that we need to keep working at it. I would like to see a goal of top 5%, not to 10% were we are now. I am not in the least concerned about test scores to maintain funding, I am VERY concerned about what the test scores show about how are students are learning and what we can do to improve that. Unfortunately the design and reporting of the WKCE makes it a very poor tool for that goal, primarily because it does a poor job of assessing students who are much above or below the “proficient” level at which is it aimed. Having a goal only of getting all students to some proficient level is nothing but a formula for mediocrity and vastly under serves most students.
If the WSJ scores were wrong, why hasn't a correction/retraction been run?
Why are you asking me? That question is properly directed to the WSJ staff.
I don't have direct information that they are indeed wrong, but it sounds as if perhaps you might?
The correct scores can be found on the DPI website.
frankly, i would think the it is the district office's responsibility to contact the paper.
Perhaps, the guys who do all the counting.
I agree anon.
Peter, you are on the Board. Why don't you take responsibility for this?
It certainly isn't attracting students to our district!
I don't have a problem with the administration or the board or even residents requesting a correction, but I don't think the error is discouraging new students, nor would a correction attact more.
People use the paper to check on their own district, to see how it's performing.
When people are looking for a district, they'll look to the District's website, or the State Dept. of Instruction.
Ed O'Connor has been aggresively pursuing this issue with the WSJ. Apparantly the paper is using the scores for an enrollment code that only includes 38 of our 3rd grade students instead of all our 3rd graders. I don't think we have a satisfactory answer as to why.
I HAVE contacted the paper by email twice, to date I haven't got any response.
Hey Peter,
Off topic, but as soon as you can get it, how about an update on how this weather has affected the constuction plans? Will we have a problem meeting the fall start up?
Hmm. What is the enrollment code of those 38 students? Sounds awfully close to the amount of third graders in the two MONONA classes.
If those scores reflect Monona kids' performance, then it certainly would be true that your experiment of increasing the class sizes of an already-challenged cohort has failed.
I would love to have you tell me that I am wrong.
This is one of the big reasons that releasing the results from november 2007 in May 2008 is a problem. People are confused about the meaning of the tests and think they reflect the learning related to the grade being completed. When in fact these tests taken 2 months into the school year reflect primarily what the child brings into class that year, i.e. the experience of prior years.
So hypothetically low "3rd grade" scores might tell us that the small class sizes of 1st and 2nd grade were ineffective at raising the achievement of these students. So if low classes didn't work in 2nd grade why keep them in 3rd grade? We shouldn't spend money on things shown to be ineffective. Of course we can't say any of this because the WKCE is a status measure, not a growth measure.
It's a little difficult to characterize putting students in standard class sizes an "experiment". The current 3rd grade class sizes are very comparable to many prior years in Monona and to our CG class sizes. Reducing the class sizes to about 15 in 3rd grade would have been the "experiment".
Lastly the WSJ reported scores are below our Nichols only scores anyway, so none of this matters- they are still wrong.
{So if low classes didn't work in 2nd grade why keep them in 3rd grade? We shouldn't spend money on things shown to be ineffective.}
Your comment above is really sad to me as a parent in this district. I read it to say that this class group struggled as a small group, so why waste the money on keeping smaller class sizes? Goes against what all the research says about successful learning environments, Peter.
Also, I think that you are ignoring the fact that these test scores also reflect the environment in which the tests were administered. A larger class size than any of these students had ever experienced and an unsettled environment caused by particular student concerns does not make for a good testing OR learning environment.
And whether the reported test scores were wrong or not, they are still indicative of a cohort with a problem that needs to be addressed.
Pretty funny - when it is pointed out that the tests taken at the beginning of the year don't tell much about that year you change your tune and say no really the tests indicate the conditions in the room at the time they are taken. Pretty slippery, you will pretty much come up with any wild idea you can to tear down this district.
So tell me, do you have any research that shows how class size during the test effects scores?
Oh, and tell me- how do wrong scores tell us anything? How can you possibly say that wrong scores can indicate anything? If the WSJ reported the scores at 100%, would that make it the greatest class in history?
Also, please answer the question on how those 38 students were coded.
You artfully ignored that very direct question.
Was it the Monona third grade only? Or would you like someone else to call Ed Oconnor and ask for that information?
You've got a problem with this class, but you are trying really hard to skirt that issue.
Bottom line is that it is really unfortunate for every one of those kids.
I very directly answered the question above:
"Lastly the WSJ reported scores are below our Nichols only scores anyway"
So no, they aren't the nichols scores. It appears they used the "Same school FAY" code but why that only reports 38 students is a mystery. They should have used the "same district FAY" code.
I think you missed the point of my comment, which is that you can read almost anything you want into a single status score like the 3rd grade WKCE. That is why I said "hypothetically".
Any close look at the effects of class size show that it makes a difference in the first few contact years, and is most significant for low SES students. The data also show that the difference diminishes with age and that by 3rd grade it is pretty negligible. Odden (pg. 93) lists the effect size of K-3 class size of 15 as only 0.25, a number just barely considered significant. This he compares to a professional development program that has 5 to 10 times the effect!
The implication of the research is that good teaching is far more important than class size, and that class size reductions-being relatively expensive- are probably among the least efficient ways to improve achievement except in targeted circumstances.
Thank you for taking on the issue of class size and what research has demonstrated. We need to be far more concerned about quality of instruction than class size. I think sometimes parents fixate on class size because that's an easy thing to measure and observe.
Class size IS important, all things being equal I think we would all prefer smaller classes and be better for it. However I think the research shows that there are other factors that are much more significant than class size, and so long as budgets are limited we should focus on the biggest bang for the buck.
Post a Comment