This is the topic of a wide-ranging discussion between Michele Kerr and Glenn Loury. They get into the highly-charged subject of racial disparities in school suspensions, and the Obama Administration’s presumption that such disparities reflected racism, which pressured schools to reduce suspensions. Here is a recent article blaming the Obama edicts for an increase in attacks on teachers.
I actually was interested in Kerr’s description of T’s, K’s, and S’s. That is, a teacher can be most focused on Teaching, on the Kids, or on the Subject. As I interpret this, a T is focused on “moving the needle” with real signs of progress for the students. K’s focus on kids and making them comfortable (think of a kindergarten teacher). An S would focus on the subject matter (think of a college professor).
I was trying to relate to this to my own experience teaching high school. I had a bit of each type of behavior.
K behaviors: On the first day, I started out with an icebreaker. Also, I told students that my goals were long term. I would value an email from a student five years later.
T behaviors: I always hoped that the AP scores would be good.
S behaviors: I had very strong views of what I wanted to get across in my statistics and economics classes.
I am not sure whether this is K or T, but I also adopted the philosophy of my favorite college professor, Bernie Saffran, which was never to let the material get in the way. If a student came in with an interesting question, even if it was totally off topic or even off subject, if I thought that I could provide an interesting take on the topic, I would go with it.
Overall, I think I was more S than K and more K than T.
Towards the end Kerr states that there is no evidence whatsoever that smarter teachers get better results.
This is blatantly false.
Eric A. Hanushek, Marc Piopiunik and Simon Wiederhold demonstrated that an increase of teacher cognitive skills of one standard deviation is estimated to improve student achievement by 11 percent of a standard deviation.
An article summarizing their work is available at: https://www.educationnext.org/do-smarter-teachers-make-smarter-students-international-evidence-cognitive-skills-performance
Hi, thanks for blogging about my conversation.
Edgar, it is not blatantly false. It’s very consistent in US education research Goldhaber has shown this more than once. A RAND meta study has discussed how consistent the finding id. As I said in the conversation, the most likely explanation for this finding in the US is that we are above a basement level of required intellect. Having exceeded that basement, we don’t find any bang for the buck by raising credential standards.
This would explain Hanushek’s research, which is global, not US, and finds an extremely small increase in results with a huge increase in teacher achievement. 11% of a standard deviation in student achievement for a full standard deviation improvement. The only way you’re going to get that full standard deviation is in 3rd world countries who went from mostly illiterate teachers to high school capabke. Western teachers are already well above that minimum. Research consistently suggests there’s a ceiling on improving teacher cognitive ability and that we’re above it.
I’ll add to this that at a certain point do you really want people that are +2, +3, +4 SD teaching anyone not in college/grad school. Seems like spending an incredibly scarce resource for a very tiny improvement. These people should be out using their vast intellect to invent things and improve society. The only defensible ROI is when they are teaching the next generation of +2, +3, +4 SD individuals to replace them in the next generation. That’s why we have grad school.
I think the current method of taking the bottom half of BA grads to teach average K-12 students makes a lot of sense from an ROI perspective. Maybe a few slightly smarter ones to teach AP Calculus.
We don’t take the bottom half of BA grads to teach K-12. Elementary school teachers are basically the third quintile (about 40-55 percentile). High school academic teachers are 55-70ile.
Respectfully, if there is a large literature out there suggesting that smarter teachers are not correlated with better student achievement, it is not easy to find.
The RAND study you referred to I am guessing is the Improving Teaching Effectiveness, 2015-2016? That document cites Michael B. Allen for the proposition that “As far as the impact of imposing more stringent requirements for entrance into teacher preparation, the research literature is inconclusive.”
The closest that I can find for support of the proposition that smarter teachers don’t matter is an article “Predicting Teacher Performance with Test Scores and Grade Point Average: A Meta-Analysis” by D’Agostino and Powers that states “it was found that test scores were at best modestly related to teaching competence and that performance in preparation programs was a significantly better predictor of teaching skill.”
In contrast, The National Council on Teacher Quality argues in favor of higher standards for teachers in a publication entitled “Increasing the Odds, How Good Policies Can Yield Better Teachers”:
“One massive study by Bradford Chaney looked at the standardized test performance of 24,000 eighth graders to determine if students did better in mathematics and science if their teachers had a degree in education. The study found that an education degree had no impact on student scores.
“Another study by economists Dan Goldhaber and Dominic Brewer found that students actually did worse on science achievement tests if their teachers had a degree in education.
“Another study indicates why it is important to rely on the findings of more than one study and to not assume that what is true for one group of teachers is equally true for another. Richard Monk found that students did better on a math test if their teachers had taken courses in math education as opposed to pure mathematics. On the other hand, Monk found the reverse was true in science; teachers who took pure physical science courses as opposed to science education courses were more effective.”
And even teaching credentials are not doing anything to assure teaching quality:
“The evidence is conclusive that master’s degrees do not make teachers more effective. In fact, the evidence strongly suggests that rewarding teachers for these degrees is an inefficient use of limited public resources.
■ Some studies have even shown that master’s degrees have a slightly negative impact on student achievement.
■ Very few studies diverge from this consensus; the findings of those that have are inconclusive. For instance, one study found that having a master’s degree modestly improved student achievement in grades 1 through 7 but had no impact in grades 8 through 12. … …
■ Only one study looked at the impact of elementary teachers earning master’s degrees in a subject area, and it found no effect.”
The National Council on Teacher Quality also urges the ed schools to tighten up admissions in their 2018 review of teaching programs:
“While nothing in the structure of graduate or alternative route programs inherently hinders them from being selective, too many simply are not. Too many imply that teaching is easy, that anyone can do it, by admitting almost everyone with minimal application requirements. Just 14 percent of traditional and 23 percent of alternative certification programs have rigorous admissions criteria.”
This matters because the justification for all the teacher pay increase proposals is that pay increases will attract better candidates to the teaching profession and that will in turn increase student achievement.
The Equity Program Charter School in New York is a frequently cited example.
This highly successful charter pays first year teachers 6 figure salaries. However, they also only hire candidates who score at the 90th percentile or better on the GRE, GMAT, etc. The return on investment of throwing billions more at teacher’s salaries, as proposed by leading 2020 presidential candidates, may not amount to anything if the education major continues to have the lowest average admissions test scores while handing out the highest average GPAs. Teachers ed is notorious for attracting students with low test scores and for being an easy A program and the data bears that out.
And while it is true that the Hanushek study is global, covering 31 countries none of which would be judged to be 3rd world, the primary focus of the study is comparing other highly developed countries with Finland.
To match the cognitive skills of Finnish teachers, the United States would need to recruit its median math teacher from the 74th percentile of the college distribution instead of the current 47th percentile, and its median reading teacher from the 71st percentile instead of the 51st.
And the resulting increase in student achievement is not insignificant in the least.
The 15 percent of a standard deviation in the PISA test that would be expected from bringing US teachers up to the Finnish standard, would do much to improve US world standing, making up about a quarter of the difference. The US is ranked 40th, well below most major Asian and European countries. The US will never achieve Finnish outcomes, but for national pride if nothing else, it would be worth moving up in the rankings.
Most importantly, though, especially here in Null Hypothesis Land, is that after trillions in social spending, the gap between rich and poor in student achievement has not changed at all. Cognitive skills as measured by standardized achievement tests are a strong predictor of future income and economic well-being. Yet the achievement gap across the socioeconomic spectrum sends a discouraging signal about the possibilities of improved intergenerational social mobility. 15% of a standard deviation would be an astonishing feat by comparison.
As Hanushek writes in a 2019 paper entitled “The Unwavering SES Achievement Gap: Trends in U.S. Student Performance” :
“Concerns about the breadth of the U.S. income distribution and limited intergenerational mobility have led to a focus on educational achievement gaps by socio-economic status (SES). Using intertemporally linked assessments from NAEP, TIMSS, and PISA, we trace the achievement of U.S. student cohorts born between 1954 and 2001. Achievement gaps between the top and bottom deciles and the top and bottom quartiles of the SES distribution have been large and remarkably constant for a near half century. These unwavering gaps have not been offset by overall improvements in achievement levels, which have risen at age 14 but remained unchanged at age 17 for the most recent quarter century. The long-term failure of major educational policies to alter SES gaps suggests a need to reconsider standard approaches to mitigating disparities.”
If politicians don’t have the guts to stand up to the teachers unions and ed schools, they should forfeit their right to use income distribution as a campaign issue.
How you could repeat all that research and not realize it supports my statement is beyond me. Except the NCTQ nonsense, of course.
US math teachers are not recruited from the 47th percentile. The average US math teacher in the US has a score well above the median for college graduates.
Once again, I think anyone who compares US to other countries is being a tad ridiculous. Hard to take international comparisons seriously. National pride? You really think Americans are troubled by their international scores? The only people who care about them want to use the scores for their own political purposes.
Not sure why you’re going on about teacher pay and education degrees.
Arnold,
As I said in the discussion, you don’t want to overdetermine the categories. Most teachers care about kids, teaching, and their subjects.
The difference might be described as “why” you take certain actions, and “how” you respond to certain challenges.
So to look at your examples:
“K behaviors: On the first day, I started out with an icebreaker. Also, I told students that my goals were long term. I would value an email from a student five years later.”
An icebreaker could be K, T, or S. It’s the type of icebreaker. A “K” icebreaker would include asking students about their lives, what successes and failures they value. A “T” icebreaker might be a quick value-add learning activity–not necessarily in the subject, just something that the kids would find useful and not have known before. An “S” icebreaker would be about the subject.
However, any teacher of any orientation might utilize an icebreaker outsider their focus, maybe because they realize it’s a useful strategy to counteract their usual emphasis.
“T behaviors: I always hoped that the AP scores would be good. ”
I would describe that as “S”. Most “S” teachers I know who teach AP focus on how many 5s they got, and frankly think any student who couldn’t get a 5 shouldn’t take their course. A “T” teacher would be thrilled if their weakest student, the one who shouldn’t really be in the class, got a a 2. A “K” teacher would argue that the AP test isn’t as important as students stretching themselves, that a test score might unduly set them back.
“S behaviors: I had very strong views of what I wanted to get across in my statistics and economics classes.”
Yes, that’s closer. This is something “T” teachers always have to use as a counterbalance–we’re always so happy just moving the needle, but we need to also be sure that there’s a baseline achievement level that must be met.
“K” teachers are found not just in kindergarten and elementary school, but all through the k-12 spectrum. My argument is that particular high school clusters (inner city African American, for example) might actually need teachers who are strongly “K” oriented simply to decide to start taking school seriously.