Our Achievement Gap in Four Charts

Results from the Wisconsin Knowledge and Concepts Exam (WKCE) administered last fall have recently been released.  I have not had a chance to review the results in any detail.  However, the district-level results do underscore the point Dan Nerad has been making at community meetings about the achievement gap plan.  Judged by WKCE results, Madison can seem like two different school districts depending on the skin color of the students taking the test.

As far as the WKCE is concerned, white students in Madison continue to do well.  WKCE test results fall into four categories:  minimal, basic, proficient and advanced.  The largest percentage of white students score in the advanced range, the highest category.  And proportionately more white students in Madison earn advanced scores in reading and math than white students statewide.

Here are two tables that compare by grade the percentage of white students scoring in the advanced range in Madison and across the state in reading and math:

 The WKCE results for our African-American students paint a different picture.  Here are two more tables, this time comparing African-American students, but here the measure is the percentage of students scoring “minimal,” the lowest of the four WKCE categories:

While the percentages seem to even out by tenth grade, in the earlier years greater percentages of Madison’s African-American students are trapped in the minimal category than African-American students statewide.

That tells our achievement gap story in a nutshell.  The achievement gap is typically measured by the difference in performance between white students and African-American students.  Our white students consistently outperform the state averages for white students while our African-American students almost as consistently underperform the state averages for African-American students.  The combination of those two facts guarantees that our achievement gap will be bigger than the state average.

The story is basically the same for our Latino students as for African-Americans.  With a couple of exceptions, our Latino students score in the minimal range in higher percentages than Latino students statewide in each of the tested grades and in both reading and math. Switching from a race-based comparison to one based on income status doesn’t change the picture much.  Non-low-income Madison students do better than the statewide average for non-low-income students; and Madison low-income students do worse than low-income students statewide.  I won’t venture into the issue of whether race or income status plays a more significant role in the disparate outcomes.

Here are a few random points that I draw from these figures.

First, the achievement gap shows up early. It’s right there when students take the WKCE for the first time in third grade.  Indeed, the achievement gap for our third grade students is slightly larger than for all our tested students as a whole.

This underscores the need for effective teaching strategies and interventions for our youngest students. This year the district has taken the single best step we could take by starting four-year-old kindergarten.  However, as far as the WKCE is concerned, we won’t have a way of knowing what impact 4K has for at least four years, since our first class of four-year-old kindergarteners won’t be in third grade until the 2015-16 school year.

We do know that if five-year-old kindergarteners come to us ready to learn, they have a better chance of being a proficient reader by third grade.  For many years the school district has given a screening test to incoming five year olds to see if they are “kindergarten-ready.”  Here is a table that shows how “kindergarten readiness” correlates with proficiency in reading by third grade:

Grade 3 Proficiency in Reading by “K-Ready” Status

Minimum/Basic Proficient/Advanced
Not K Ready 44.1% 55.9%
K Ready 7.1% 92.9

According to this relationship, if the institution of 4K is able to increase the percentage of K-Ready students from our historical average of 62% to, say, 75%, then, all else equal, when they reach third grade, the percentage of those students that test proficient in reading could be expected to increase from 78% to 84%.

So, four-year-old kindergarten is a big positive for us, though it will take a while for its impact to be felt according to traditional measures.

Second, we need better and more timely assessments.   WKCE test results have their uses but there is far too long a delay between the test and the results.  One of the proposals in the achievement gap plan is for “All students K-8 [to] undergo benchmark assessments to monitor progress in reading.”  These assessments are to be administered at least three times per year and all students who test below proficiency in reading are to receive intensive progress monitoring at least monthly.  Assuming that instruction and interventions are appropriately keyed to the assessment results, this makes sense.

Third, we always need to bear in mind that an individual student is quite distinct from the cumulative average of all students with his or her characteristics. For example, even with the large measured gap in third grade reading, about 66 Madison African-American third-graders, or nearly one in five, were measured as advanced in reading.  The tendency to assume that all African-American students need special help can lead to diminished expectations even for our high-achieving African-American students, a source of frustration that we hear about from those students’ parents

Finally, the troubling differences in levels of student learning that give rise to our achievement gap present an enormous challenge for our teachers.  We as a District have long been committed to inclusive and heterogeneous elementary school classrooms. Consequently, given the gap, our teachers frequently lead classrooms with a number of high-achieving students and a number of struggling students. Imagine how much dedication and ingenuity it must take for our classroom teachers to provide a learning environment where all their students can thrive.  It would be helpful to hear from teachers about how they think they can be most effective in teaching all students in classes with such a wide span of developed capabilities, given our resource limitations.

Even test results as generally uninformative as the WKCE make clear the extent of our achievement gap in Madison.  From the perspective of the WKCE and based on statewide averages, our white students on the whole seem to be doing just fine while our African-American students on the whole are struggling.  This shouldn’t come as news to anyone, but it does underscore what’s at stake when over the next several weeks the School Board starts to decide what components of the superintendent’s achievement gap plan we’re actually willing to raise taxes to support.

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

9 Responses to Our Achievement Gap in Four Charts

  1. Dorothy Conniff says:

    Thanks for the very clear summary, Ed. It’s informative to see the comparisons both across racial and economic groups and within the group. It would be comforting if we had more information about the quality of instruction and the effect of it in the 4-year-old program. There are plenty of good assessment tools available. Are teachers in those programs, or the administrators using them?

    • Dorothy — I understand that 4K students are being assessed for their kindergarten readiness with the same screener the district has used for a while. We don’t have any results yet. I’m not aware of any other 4K assessments. By the way, here is a link to an interesting blog post on pre-K assessments that TJ Mertz linked to earlier: http://shankerblog.org/?p=5598

  2. Mad4Madison says:

    Ed –

    As always, thanks for the information and a place to discuss very important topics that face MMSD. I guess in this case, I have a couple of questions and / or comments.

    1) Why does it have to be that we must raise taxes in order to support some components of the superintendent’s achievement gap plan? We are still under the levy cap, so we can spend more without an impact to the community. This question is not meant to be a “don’t raise my taxes comment”, but rather a query into the position that any new program must be a net add. Are there not programs that have been implemented that have either not been effective or not effective enough? Has the district ever cut a program that is not performing as expected?

    2) If the WKCE metrics are nearly the same based on race and income, could it be that one is the cause and another simply an indicator? The example I think of in this case is the thought once held (and fairly recently I might add) that if a child was exposed to books, then test scores would increase. So a huge effort was put forth to provide books to children. Now the reality was it was not the books – it was the fact that parents were reading to the children. So all the books in the world would not matter – it was the parental reading / involvement. And the studies about race and income bear some resemblance to the same study. For example – a single mother is working two or three jobs. He / she cannot take the time to read to his / her child because they are simply using every ounce of energy to keep food on the table. That is a huge issue for the children, ragardless of their skin color.

    I hope the new board is ready to deal with all of these issues and the reality of budget and other economic strains in the community.

    • Mad –
      1) If we add new programs we’ll have to pay for them, and increasing property taxes is the one way we have to increase our revenues. Being under the levy cap means that we
      can increase taxes without bumping up against our revenue limit; it doesn’t mean we have access to free money. I try to think of the addition of new programs independently from the possibility of eliminating existing programs that don’t seem to be effective. Unless our revenue limit is binding, I don’t think there needs to be a connection between the two. We should be eliminating programs that aren’t working regardless of our other needs. Can’t say that offhand I can think of one that we’ve eliminated, though.
      2) I’m not a statistician, but I believe collinearity is what they call what you’re referring to. The district’s value-added analyses have found that race has more of an impact on test results than income status. It seems to me that this must mean that, other things equal, low-income non-African-American students tended to outperform non-low-income African-American students, but I don’t know for sure.

  3. Gary says:

    Ed:

    Thank you for another succinct summary of a complex issue. I understand your post was designed to be concise, but it provoked the following question about how data is reported generally.

    Do you think using raw numbers may more effectively inform the public about trends within our schools/district?

    For example, 2011 WINSS WKCE/WAA combined data indicate that approximately:*

    -195 out of 383 black third graders are non-proficient+ in math, compared to 101 out of their 801 white peers.

    -181 out of 350 black 10th graders are not proficient in math compared to 79 out 854 of their white peers.

    -1274 out of 2761 black NNSD students are non proficient in math compared to their 488 out of their 5394 of white peers.

    Assuming privacy concerns could be alleviated by pooled or combined reporting, would it be worth relaying more raw numbers to the public?

    Thank you for your continued leadership and commitment to open dialogue.

    -Gary

    *The numbers are approximate because dataset insufficiencies prevent exact calculations.
    +Not proficient includes all students scoring basic or below

    • Gary –
      Thanks for your kind words. I agree that the actual numbers of students scoring in the different proficiency ranges can seem more striking than their percentages. I’m having a little trouble tracking your precise numbers though. For example, district-wide, 36.3% of MMSD African-American students scored in the minimum range in math and 18.2% scored in the basic range. This adds up to 54.5% in the non-proficient categories. 54.5% of 2,761 tested African-American students equals 1,505 students. Similarly, 5.9% of white students in the district scored in the minimum range in math and 5.8% scored in the basic range. This adds up to 11.7% in the non-proficient categories. 11.7% of 5,394 tested white students equals 631 students. Whatever the precise numbers, though, the disparities remain quite large.

  4. kristennel says:

    Ed, are there any charts that show acheivement as it relates to grades (vs. WKCE)?

    I ask this because I have now been matched with my “Little” for 5 years. During that time, her grades have improved tremendously. Our goal is a 3.0 GPA and we are almost there.

    But her WKCE scores…not only has there been no improvement – they have fallen. We started when she was in 3rd grade and she is now in 8th grade, so we’ve got 5 years of WKCE testing data.

    If you were to judge this mentoring relationship on WKCE scores alone, you would conclude it is a dismal failure. You would conclude that BB/BS makes no difference – indeed, it has a negative impact.

    I realize that the WKCE is all we have right now as a standardized test measurement….but I seriously question its value as the sole measurement of progress. I wonder if statistics related to grades might paint a more focused picture of the acheivement gap and lead us to more meaningful actions to eliminate it?

  5. Kristen — I don’t think I’ve seen any data collected on grades at the elementary or middle school levels. At the high school level, we tend to track passing grades and credits accumulated. Since a “D” is considered a passing grade, sometimes we look at grades of “C” or above. But your experience (and thanks for all you do!) helps illustrate that standardized test scores only tell one part of the story of students’ progress.

  6. mary battaglia says:

    Ed,
    Thanks for the summary.
    I’d like to add to the summary. Again, the racial gap is not just a Madison issue. Actually the issue is really graduation rates…..and for African Americans across the nation that stat is horrible. I suggest everyone read the blackboysreport.org. I know graduation rates are defined differently but whenever this is discussed in Madison, I feel like people feel we are in a bubble. The issue is a national issue and not one specific to Madison.
    Second, I would like to suggest one of the problems for our students being measured by WKCE is our students are rarely measured in such a large fashion. This harks to Kristen’s issue above. All three of my children have attended MMSD schools. Compared to my 11 nieces and nephews that live in various states in the south, my kids rarely if ever took test. It was not until 4th grade that my kids took a spelling test, or a timed math test. The view was testing was bad, negative and stressful. While I think that is true, my Junior in HS is just now figuring out how to calmly take a test. I believe avoiding test is a nice idea, but learning to take large test, how to cope and doing it more frequently would have helped all of my children. My oldest son took Math Olympiad offered via a parent volunteer in 4th grade. The greatest asset to that program was the parent spent the first month teaching the kids “how” to take a multiple choice test. He claimed non of the kids had ever seen one.

    While I am not wanting weekly big test….I do see the need to a) have an assessment at the beginning of the year to see where your class and each student is and then a mid assessment and an end of the year assessment on the standards. I’m not sure how staff know if kids understand the content if you never assess their knowledge. b) Frequent exposure to multiple choice test could decrease stress of such an event. c) my kids did have some exposure to assessments as we had silly educational games on our computer and many books and activities at my house that were assessments. Kids of lower income families may not have assess to such items and therefore a WKCE at Fourth grade could be such a foreign event.

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