The OECD has a report, Education at a Glance 2010, that provides a shockingly flawed comparison of the amount of time U.S. teachers work relative to teachers in other countries. According to the report, U.S. teachers work 1,913 hours over a 180 day school year that is 36 weeks long. And also according to the report, the average OECD teacher only works 1,659 hours over a school year of 187 days that is 38 weeks long.
So, if we believe these OECD numbers (which the WSJ apparently did in this blog post), U.S. teachers work 15.3% more hours per year than do their colleagues in other developed countries.
But if you believe the OECD comparison I have a lovely bridge to sell to you. According to the report’s methodological appendix, the method by which the U.S. information was collected was different (and clearly less reliable) than how it was collected from all of the other countries. In every country except the U.S. the hours worked was derived from teacher contracts or laws. But in the U.S. the information was drawn from self-reported responses to a survey of teachers. (See p. 75 of the appendix).
A valid comparison would require that the information be collected in similar ways across all countries — either we rely upon self-reports in surveys of teachers for all countries or we rely on contractual hours for everyone. But using self-reports for the U.S. and contractual hours for everyone else produces obvious distortions. People may be inclined to exaggerate the hours they work in a survey. And the definition of time worked is ambiguous. If I think about my students while I am brushing my teeth or running on the treadmill am I working during that time?
We have good reason to suspect that the self-reports from U.S. teachers are over-stated. If teachers really worked 1,913 hours over 180 days, as the report claims, they would be working 10.63 hours per day. And the numbers I’ve provided are just for primary school students. For high schools, the OECD report claims U.S. teachers are working 1,998 hours over 180 days, which works out to 11.1 hours per day. I know some teachers are very conscientious and work long hours but I simply do not believe that the average high school teacher is working 11.1 hours per day.
I know this might invite the wrath of Diane Ravitch’s Army of Angry Teachers, but I suspect that the average hours worked by U.S. teachers is significantly less than the OECD says (and the WSJ repeats). And I know that the comparison between U.S. and other countries is flawed by collecting the information from self-reports in the U.S. but from contracts everywhere else.
Jay P. Greene's Blog 
I worked a LOT of overtime when I was teaching and I probably did do at LEAST 10 hours a day, but I lived at school and coached softball in the Spring and assisted with after school activities in the Fall. Most teachers are out of their buildings by whatever time is stipulated on their teacher contracts. In fact, the new principal for our local high school just relayed to me her frustration that the ‘office manager’ answering the phones would stop promptly at 2:30 and walk out of the building because her contract stipulated her work ended at 2:30. Consequently the phones either didn’t get answered after 2:30 or other people were forced to make up her work. The latter is by far more than norm than the former I assure you. This piece of ‘research’ must have been done using the “fuzzy math” method.
In Boston, I think the contract says —
6.5 hours * 180 days
minus
18 days for various sick leave/personal leave
= 1053 contractual hours per year, compared to the 1659 of contractual hours in other nations.
I wonder what the real hours are.
I’d bet it’s not a bell curve. A sizable chunk of folks who do the minimum (and even within the mandated 6.5 hours, there’s downtime). Another sizable chunk of teachers who work 12+ hours per day.
P.S. Would love your opinion about something.
What’s your take on charter school closure, and the long-term prospects of school choice?
I.e., my view is that aggressive closure of the lowest 10% of charters (500-ish schools) would, long-term, boost quality in a way that would lead to ever more choice.
But the status quo alternative is generally to let parents continue to choose those schools (which understandably may still be better than the alternative)…
The contract between the Hawaii DOE and the HSTA required about 250 minutes of face-time with students per day. For schools on regular period schedules, this meant five fifty-minute classes per day over a 180-day school year.
When I started teaching (Math), I graded homework and classwork, and had to work multiple hours outside of school every day. I eventually decided that this did not contribute to student performance, and only graded weekly tests. I guess that might be different for English or Language teachers.
[…] Data used to show how hard U.S. teachers work might not be comparable. (Jay Greene) […]
[…] Data used to show how hard U.S. teachers work might not be comparable. (Jay Greene) […]
Jay–since you were never a teacher, you would not know that teachers work substantial hours on the weekend. Evidently you don;t and never have, otherwise you would have considered this.
You simplistic analysis is every bit as bad as you claim the comparison is. Much of your “work” suffers greatly from methodological flaws as well which is what we would expect from a govt professor who does not know much about education.
I am a charter school teacher who works extremely long hours. School is from 8-4 (which in itself is an 8 hour day). If I work mandatory breakfast duty which is once a week, I work 7-4. This also does not account for the fact that I often get to work between 6:30am and 7 to prepare for the day, as well as stay after school for meetings and preparation well into the evening. I would say on average, my actual in school hours are 7-5 with many days longer than that. After I come home, I also spend at least an hour a night on the phone with parents or answering parent emails. On Sundays, I also spend at least 2 -3 hours planning for the week, writing newsletters for families, and grading work. Adding all of this up, I definitely work a 60 hour work week. Oh and I DO work summers in mandatory summer school, so that argument about teachers not working summers is out the door too. I could not imagine having a family and being a classroom teacher like I am now. It’s not sustainable, especially for the salary I am making.
You would not be a typical teacher then. So yes that argument about teachers not working summers is still very much alive.
Bill and LMK —
Do you think teachers in other developed countries also work at night and on the weekends or is the remarkable worth ethic you describe particular to U.S. teachers? If it is true of teachers everywhere, then the comparison is flawed because the non-contractual hours are not being counted for teachers in other countries.
I’m sure our European counterparts also work just as long and as hard as we do. However, your argument attacks American teachers in an extremely demeaning way. Do some teachers leave when the school bell rings? Sure. But most teachers go above and beyond. When was the last time someone picked apart your hours? I’m going to bet probably never. Until you know the emotional, physical, and intellectual strength it takes to put your all in the classroom everyday, you should probably stop writing on this topic. I will also extend an open invitation for you to come and spend a day with me and my students in the fall so you can get a 24 hour taste.
LMK — If you think that teachers in other countries are working as hard as those in the US, then you agree that the OECD comparison is flawed because their non-contractual hours are not included while those for US teachers are.
I’m glad to hear that you agree with the main point of my blog post.
What did Jay say that was an attack on teachers or demaning? Jay is asking very reasonable questions about social science methodology.
Treating reasonable questions about data and evidence as illegitimate is a one-way ticket to irrelevance. You’re only hurting yourself here.
“but I suspect that the average hours worked by U.S. teachers is significantly less than the OECD says ” So, you are saying teachers lied about the hours they work? On what EVIDENCE do you base this on? Are you saying surveys are flawed? And you have never, ever relied on survey data, right???
Amazing how you twist yourself in a pretzel just like all the other ideologues.
There’s an important difference between “lying” and what is called “social desirability response bias.” The latter is nicely summarized in Wikipedia as follows: “Social desirability bias is a term used in scientific research to describe the tendency of respondents to reply in a manner that will be viewed favorably by others. This will generally take the form of overreporting good behavior or underreporting bad behavior. The effect is common within the fields of medicine, psychology and the social sciences.”
http://en.wikipedia.org/wiki/Social_desirability_bias
Being more devoted to education by working longer hours is perceived by the respondents as socially desirable and thus it is likely that it is over-reported. This doesn’t mean that anyone is lying or a bad person.
And your evidence that social desirability influenced the responses of teachers in this survey?
(Bill Martin): “So, you are saying teachers lied about the hours they work? On what EVIDENCE do you base this on? Are you saying surveys are flawed? And you have never, ever relied on survey data, right?
Amazing how you twist yourself in a pretzel just like all the other ideologues.”
A few points:…
1. “Ideological” is an uncomplimentary way to say “systematic” and I try to be systematic.
2. “Hours worked” is a bogus performance measure of any job where time in place does not relate strongly to results (any job above nightwatchman or doorstop). In the education industry, paying teachers (and students, in “credit hours”) on time in place is an admission that the industry is a make-work program for dues-paying members of the NEA/AFT/AFSCME cartel.
3. Dunno ’bout Jay but I’m say teachers lie about their hours. Just stand outside the faculty parking lot from 0630 to 1600 local time and count the cars. And the line that teachers take work home? So do carpenters and cooks, but they don’t get paid to sharpe their tools on the job.
4. At Waianae HS, one special ed teacher reported colleagues who had 1/2-time classroom and 1/2 time “floating” positions, where they (were paid to) work with individual sp-ed students in regular-ed classes, who just went home after lunch, since no one was keeping track of where they were. At Campbell HS, a colleague told me, two English teachers combined classes and showed movies all year while one or the other signed in for both and the other stayed home. They gave all their student a B so no one complained.
Bill: In social science, first you discuss whether there are reasonable grounds to expect that a certain bias might have affected the data, and then you investigate whether it did. Both these steps are social processes that scientists engage in through dialogue. We’re still on step one. Science can’t progress if someone raises reasonable concerns that bias might be present and others immediately respond with “You’d better be able to prove that it did, and prove it right now, or else you’re a bad person for even raising the question and you need to be expelled from the respectable conversation!” That kind of response isn’t science, it’s about power.
Yes–and social scientists DON’T PUT FORTH CONCLUSIONS based on intuition or an expectation. Those us us who actually TEACH and DO RESEARCH in this area would know that. Not that most researchers–even those who agree with Mr. Greene–would consider him a true social scientist anyway. But its nice of you to imply it.
Let’s review, Bill. You say the following things about me:
“Much of your ‘work’ suffers greatly from methodological flaws as well which is what we would expect from a govt professor who does not know much about education.”
“Amazing how you twist yourself in a pretzel just like all the other ideologues.”
“Not that most researchers–even those who agree with Mr. Greene–would consider him a true social scientist anyway. But its [sic] nice of you to imply it.”
And you think I am the one who is demeaning to others for saying: “I know some teachers are very conscientious and work long hours but I simply do not believe that the average high school teacher is working 11.1 hours per day.”
I believe the BLS has an estimate that puts the real hours worked by teachers (this is an estimate generated by teacher survey responses) that would amount to about 1600 hours per year – they actually put the work week at about 36-38 hours per week during the school year… of course, teachers typically don’t work summers.
I believe BLS surveys the employers. But you are right to point out that another report produces a measure of annual hours worked by teachers that is much lower.
Don’t forget, most teachers overestimate the amount of time they work (most people would). There is also duty periods and grading periods in which teachers don’t actually teach and many don’t do anything productive at all.
said BLS survey – http://www.bls.gov/opub/mlr/2008/03/art4full.pdf
this is based on teacher responses and is thus likely teacher responses overestimated their own work hours. Still, you can’t reach 1900 hours during the course of a school year even with this data.