Hiring for Hard to Fill Teaching Positions
Mike Mathieson
September 24, 2022
We currently have 82 teaching (including assistants) openings in the district.
There’s a strong correlation between the number of openings in a school and test scores. As test scores increase the number of openings decrease.
A highly cited teacher merit pay meta-analysis of 37 different studies shows merit pay has a statistically significant positive effect on student test scores roughly equivalent to 3 additional weeks of learning per year.
We would like to see merit pay used for hard to fill positions. Structuring this will be difficult, but not impossible.
Side note: three weeks of additional learning is important, which leads me to comment on our incredibly bad track record of summer school.
There are currently 82 teaching and teaching assistant job listings on the Decatur Public School District web site. We clearly have a hiring problem and we’ve spent the last several articles explaining how we would try to use CARES money to help with this problem, but finally Duane and I are moving on with another suggestion. As I’ve said several times, we don’t want to complain without offering solutions so here’s another possibility to attract teachers for hard to fill positions. The Venn Diagram below shows three different ways to use extra pay to attract teachers.
The first possibility is simply to pay more for hard to fill positions. The economic law of supply says that if Special Education teachers are hard to find, an easy way to attract more of them is simply to pay more money. If math and music teachers are harder to find than PE teachers, paying more for these the hard to fill jobs will eventually even things out. But, for teachers’ unions across the country, this type of pay has been untenable. So, looking further, another possibility within the realm of simply paying more money is to pay bonuses where a teacher receiving a hiring bonus must stay for a certain number of years. Again, this will be tough to get past the teachers’ union, but I like this idea better because it doesn't lock you into long term salary structures that might not make sense in five or ten years.
A third option is to use merit pay, targeted toward hard to fill positions (the “sweet spot” in the chart above), to attract teachers to those jobs. One of the reasons I like this option is because a meta-analysis of 37 different studies suggests that “the presence of a merit pay program is associated with a statistically significant positive effect on student test scores (0.043 standard deviations for U.S. based studies), roughly equivalent to 3 additional weeks of learning.” Three weeks is a lot of additional learning so despite union objections, merit pay has been implemented in Philadelphia, Cincinnati, Alabama, Arkansas and most famously in Denver. For our smaller district though, it seems to make the most sense to attempt to implement merit pay where we get the additional benefit of promoting and encouraging teachers to take the hard to fill jobs.
But how can we define which are the hard to fill jobs? After all, if you entice people to take those jobs they no longer appear to be “hard to fill” jobs! From the data on the DPS website and looking at the school building openings it seems like there’s a strong correlation between student test scores and hard to fill positions. This seems logical and the graph below helps make the case that as test scores drop the number of job openings in the school tends to increase.
So, a possible solution (there are many and I should leave this to the experts to work out, but I’ve gone this far) would be to pay a bonus to teachers based on the percentage classroom increase in test scores over the building test scores. Perhaps an incentive system where a 100% increase over the building average will allow the teacher to receive a maximum 100% of the possible bonus. This way the merit-pay bonus potential in a school where the average percentage of students that meet or exceed recommended performance levels for the IAR test is 5%, is much greater than for a building where the average is 25%. As an example, and to explain this idea better, in a classroom of 24 kids a 100% increase over the school average for the 5% school would be to have just 1 additional child meet or exceed the recommended score, where in the school with an average IAR percentage of 25% that meet or exceed, you would need 6 kids to perform better than the average for the building. (5% of 24 is 1.2 and doubling that would mean only 1 child would need to perform better. Likewise, 25% of 24 is 6, and doubling that would mean that 6 children would have to perform better – a much harder proposition.)
I realize this is a tough sell. I had a few former teachers read this article before I published and each said something like, “but it’s so hard to figure out, from looking at test scores, who the best teachers really are”. And, “it’s just not that simple – better test scores don’t mean better teachers!” I realize there are many issues with merit pay. After all, I just spent a few hours reading a lot of the research in the study I referenced above. But my response is – do you have a better plan? This will hopefully work to entice people to take some of the harder to fill jobs, and the teachers that take those jobs may be the stronger teachers. Currently the teachers in the buildings with hard to fill positions know the students don’t stand a chance getting shuffled around between other classrooms and the occasional substitute teacher. We can ignore the problem, or we can try to do something about it. We know the kids that suffer are the ones in the low socio-economic category, which are the kids that need our help the most.
So, that’s our plan. I realize this plan would be difficult to get past the union, even if it would pay more money out to teachers, it would help alleviate the ongoing problem of shuffling kids around between teachers and subs, and it helps the kids that need the most help. Does anyone have anything better? Let’s work on a solution now, before we have 100 openings in the district.
And finally, a note about the 3 weeks of extra student learning that is the result of teacher merit pay… As I read through research to write these articles and learn about things that work to help children succeed in school, I continually read studies that say, “this (fill in the blank) methodology provides the same learning advantage as a child spending an extra week in school”. And I think Wow! That’s significant. And then I read about our disastrous summer school program where (and I’m not making this up, see below) half our students perform THE SAME OR WORSE in math after 4 weeks of summer school. Wait… What? And the results for English and Language Arts are nearly identical where 46% of the kids perform THE SAME OR WORSE after 4 weeks of summer school. This is horrible. If you’ve been reading our articles, you might already know that Summer School is not a very effective way to help low SES students achieve… but could it be? Should we try harder with our summer school program, or simply give up – make summer school simply and day care program - and not publish these horrible numbers. Or at least not try to make them sound good. They are horrible.
We’d like your feedback on both the merit pay suggestion, and should we rethink summer school. If you have any ideas don’t be afraid to comment below.
