Exciting news today, class! After a four-year wait we finally have new NAEP test scores for 12th graders. Without further ado, here are reading and math test scores from the early '90s all the way to 2019:
These are the most important test scores we have. It's nice when 4th and 8th grade scores are up, but what really matters is the end point. It doesn't matter if you're leading in the 200th lap if you end up behind ten other cars when the flag comes down after the 300th.
In general, test scores in 2019 were down slightly compared to 2015. The most tragic part of this is the continuing slide in test scores for Black students. In both reading and math, their test scores are now more than 30 points below those of white students. This means that, on average, Black students who graduate from high school have reading and math abilities similar to a 9th grade white student.
This has to get turned around. Short of existential global crises like climate change, it's perhaps the most important problem our country has. But what should we do?
Slides like these make me wonder about the base line. Are we testing more students? Are the tests remaining relevant? The changes are so tiny we're probably seeing the error margins creep in.
For the 2019 math assessment, there were about 150,000 students per grade for grades 4 and 8, and about 26,000 for grade 12. I don't have the reading numbers, but they would have been about the same. There was also a science assessment for a small number of 12th graders. (The science assessment is expensive to conduct because it includes a hands-on experiment kit for each student.)
There is a minimum sample size of 2,200 students per state in order to be able to report state-level statistics, but larger states have more, determined partially by the size of the state and partially by agreements between the states and NCES. (I think Vermont consistently has the highest average sampling rate.) There are also 23 urban school districts that participate in a program called Trial Urban District Assessment* that are sampled more heavily.
The number of students is fairly constant, except that NCES is always trying to figure out ways to spend less money without compromising the statistical validity of the results. The plan for 2021 was to have a smaller sample, but I don't remember offhand what the exact numbers were. I don't think it has been decided whether or not 2022 will be a big year (normally the odd numbered years) to make up for 2021 being canceled, or if 2022 will be a normal small year and we won't have another big year until 2023. The assessments during small years are much smaller (less than 10% the size of the big years), and this is where experimental studies and such are performed.
You can find sample questions here: https://nces.ed.gov/nationsreportcard/experience/booklets.aspx
And info on the technical design of the assessments here: https://nces.ed.gov/nationsreportcard/tdw/instruments/
*It's still called "Trial" even though it's been going on for nearly 20 years.
I am jacked up to see this trumpeted on Andy Sullivan's substack.
The tests are OK. The changes from one year to the next are small and usually not significant, but over time they add up to something very significant.
Why do these test results always exclude Asian Americans, who generally crush whites?
I feel like if we're having a good faith discussion about identifying problems and creative solutions to this that break our decades long track record of failure, we should start with a chart that doesn't lead us to the conclusion that we lefties are most comfortable with... that whites are on top (they aren't), and that the reasons are most likely nefarious.
The data reported by NCES does break it out more finely: White, Black, Hispanic, Asian/Pacific Islander, and American Indian/Alaska Native. (But a lot of states have so few students in the latter two categories that the state-level tables are largely blank.)
For example, the national-level math scores for grade 4 in the above five categories were 249, 224, 231, 260, and 227, respectively.
One creative avenue for tackling this problem would be to put real effort into understanding what accounts for the Asian/White achievement gap. This would put aside our usual assumptions and hobby horses, and might lead to deeper understanding about the ingredients of student achievement.
I have to assume that closing that gap would lead to insights that could then be applied to blacks and Latinos and yield tangible positive effects.
It's all been done. We're spinning our wheels at this point.
Check out John Hattie. Plenty of research has found what "works." It's teachers in orderly classrooms, using assessment to guide instruction, providing structured practice, and offering corrective feedback.
What you'll find in schools of education is an emphasis on implicit bias, creating community, and assigning group work.
I'll have to read that. As a parent, that rings 100% true. More than true, actually, since kids doing group work often fights against creating community, since they end up getting on each others' nerves.
" using assessment to guide instruction,"
I come from a family of educators. 2 of my sisters were teachers and my brother was certified but went in another career direction.
The problem with letting assessments be a guide for instruction almost universally morphs into teaching to the test. Why? Because teachers, schools and the school districts are rewarded for higher overall scores. Heck even real estate agents use the scores as a hook to entice moving into certain neighborhoods. "This neighborhood's schools score the highest on standardized tests - you would be foolish to deny your kids and education from these local schools."
So it is Kevin who is failing to report the Asian statistics?
Unfortunately this subject is so loaded with emotion, and so complicated it seems impossible for good research into causes and experiments with interventions.
I'd like to see state or county experiments with things like Sen. Booker's baby bonds, but such an experiment would have to take 30 years or more so a non-starter. I'd like to see research into the long term effects of the various private initiatives by millionaire donors to provide college for all of a class.
I'd like everyone to live happy lives and live to 100. But my likes have no power.
Kevin ignores this "problem" has been researched and beaten into the ground already. The government spent years and millions of dollars looking for the best instructional practices for poor kids and found a clear winner: Direct Instruction.
That was 40 years ago.
I agree in wishing we could all live to 100...while eating and drinking whatever we want.
For many years now I’ve been of the opinion that the only way to get the Black test scores near Whites is to break up poor Black neighborhoods. I just don’t think you can have such large concentrations of poverty and have good schools. Maybe if you doubled or tripled the budgets of those schools you really could overcome the testing gap, but even that isn’t assured.
If you want to significantly increase test scores you either need lots more low cost housing in middle and upper class neighborhoods. Or you need to gentrify the poor neighborhoods and figure out a way for poor people to continue to live there. This wouldn’t solve every problem overnight, but I strongly suspect in just 10-15 years you’d finally see the gap going away.
Instead of uprooting neighborhoods, only to have white people re-segregate themselves anyway, I wonder if we should explore new ways of funding schools that ensures that all schools at the state level are funded equally?
If funding weren't an issue, is there any reason why a mostly black school couldn't be as successful as a mostly white school, especially considering that bias in fully integrated schools is often cited as a reason for lower minority test scores anyway? I wonder if there's an implicit racism to the idea that the key to minority success is proximity to white people?
But all schools within New York City are funded equally (in fact schools in poor neighborhoods receive slightly more), and the racial achievement gap is still huge.
I had assumed that everyone was aware that quite a few places have equalized spending on school children. Even some southern locations like Mississippi have passed laws like that. Obviously it didn’t solve the problem. This is why I mentioned that doubling or tripling the spending on Black kids might help - because equal spending did very little. Of course, that will ever happen.
Nice of you to change my “class” discussion to a Black/White thing. And to talk about my “implicit racism”. Calling people racist, but doing it indirectly, is obviously the way to solve the testing gap. I suppose you also kind of imply that I’ve never really thought about the issue or looked up any real studies or discussions before, but I’ve paid attention over the years. I actually don’t mind arguing with people as long as it remains civil and intelligent points are made. Not that it matters - on Kevin’s new site there are no notifications of responses.
I must say, I haven’t seen many studies that conclude that black students in predominantly white schools are regularly discriminated against academically. You may get a teacher that discourages some black students from taking more AP classes, but _usually_ it is isolated teachers and not entire facilities. I _have_ seen several studies that noted that black students who moved from very poor to more affluent schools generally did equally well if the move happened when the kid was in grade school. If the move happened when the child was older then the move doesn’t change their scores much.
I still think my point is very good. Better than just doing what we’ve been doing - which hasn’t helped in decades. I can’t help but notice that you are also just repeating the white flight analogy from the 60s. Most white communities have minorities living in them these days. Nobody runs in terror. I repeat my 2 cents: Concentrated poverty doesn’t just make closing the Black/White gap difficult to close - it makes it impossible.
> But what should we do?
Do you think there's any correlation between adequate ventilation and percentage of black students at a school? Or test scores?
According to ...
Cracks in the Pavement: Social Change and Resilience in Poor Neighborhoods Paperback – September 2, 2008
by Martin Sanchez-Jankowski (Berkeley)
https://www.amazon.com/Cracks-Pavement-Social-Resilience-Neighborhoods/dp/0520256751
... Ghetto schools fail because half the students and many teachers (!) don't see enough remuneration waiting for them in the labor market to make the extra effort worth it.
Berkeley professor Sanchez-Jankowski spent nine years on the ground in five New York and Los Angeles poor neighborhoods researching this. He had previously spent ten years on the ground studying gangs.
There is no easy answer to this as discrimination left such a long standing feeling of inferiority on blacks that even after 2 generations they are still feeling like they are not capable of learning which is totally false.
Add to that the NUMEROUS states that still are fighting the civil war and we have a mess on our hands.
I major cities like NY and Boston inner city schools do get funded equally but no one wants to work int he inner city. The Betsy Voss model of education is to make it all private which allows for even further discrimination with little to no oversight.
Is free college education the answer? I don't know. I do believe that in too many poor households there is too much worry put on paying for health care and paying for education. For upper and middle class whites those worries are diminished somewhat by jobs with low cost healthcare insurance which eliminate ONE THING to worry about
I developed a close bond with a squad of soldiers I served with in Nam. 8 men total 5 were black. It took awhile but we all became friends and I stayed in touch with them after they left the army. Our get togethers are something, although the last one was virtual. They all agree that free college for all is a good thing but wonder how those struggling to pay college debt off would fee of the class behind you was tuition free? We agreed we have to start somewhere
It's a tough question as to why black students do worse that others.
I don't have any answers but given our nations apathy towards education (the majority still do NOT have their bachelors) perhaps its time we start allowing opportunity for all.....