« All Eyes on the Teacher | Main | Education's Guy Problem »

The Teacher Effect


This just in: Good teaching matters. Indeed, a new study out of Pittsburgh suggests that improving teacher quality across the board may be the surest way to close the racial achievement gap. The study, which looked at Pittsburgh students’ test scores over a two-year period, found that a student’s teacher was a better predictor of his or her performance than race. The scores varied widely depending on the teachers’ ranking, the study says, regardless of students’ race. “These … are not random effects,” said Robert P. Strauss, a Carnegie Mellon University professor who led the study. “In Pittsburgh, the teachers who are successful are successful with black kids and white kids.”

Still, school officials cautioned against reading too much into the report. “Poverty is a factor that affects achievement … race is a larger factor,” said Linda Lane, Pittsburgh’s deputy superintendent of schools. “And there’s a lot of variation in African-American achievement from one school to the next … but we don’t always know what that difference might be.”


I'm not seeing any information about how the teachers were evaluated? Does anyone have any information as to what the study is calling "good teachers"? I mean, isn't that sort of the point? Please tell us what makes these teachers so "good"!

I agree, MarjeeC. The findings are interesting, but I would like to learn more details about the study.

Anthony, could you post the name of the report and/or provide a link to it?


Here is a link to his presentation to the Pgh. Board of Ed. under "#6 Current Papers/Presentations"


Thanks, James.

Ed & MarjeeC: My sense, from the article I linked to and a quick review of the study, is that they haven't gone into the question of what makes teachers "good." They just know that some teachers consistently produce better student test scores across the board. Sorry if this wasn't clear in my summary. (I admit, it wasn't.)

Also, FYI: There was a brief mention the article about one of the school board members wanting to take the study to the next level and determine what those "good" teachers were doing in the classroom to get the better results. Sounds logical ...

The power point that I gave can be found at:


The setting did not permit an econometric discussion of fixed effects regressions,and the actual paper by my doctoral student, Haijing Hao and myself, is in process.

There is one example of fixed effects analysis in the power point so that one can see that all the usual factors, from prior year's achievement as measured by Terra Nova, to SES, ethnicity, family background, disciplinary incidents, attendence etc. were in our various econometric equations.

Earlier pieces on education policy issues (see especially Vogt-Strauss and
Severino-Strauss) are on my web page.

Robert P. Strauss
Professor of Economics and Public Policy
Carnegie Mellon


Dr. Strauss' Power Point presentation shows a statistical correlation, only. His research team may have uncovered evidence that good teachers and/or good principals CAUSE better student performance on the tests; but if so that isn't shown in the presentation. Or at least, I did not see it.

To the extent that good results may follow a teacher moving from school to school, cause-and-effect is better supported by the evidence. To the extent that a particular teacher remains at the same school, there is greater question about what is responsible for better performance in that teacher's classes.

I am cxurrently mworking on my Doctoral, ED. D. in Teacher Leadership Curriculum Design & Instructional Development.

My website is down for reconstruction and updates as I relocated to FL from MD. Having taught in 3 different age groups, and serving the State Of MD as a Curricular Developer for Core Reading required courses for attaining and renewing certificate of eligibility and certification; this debate has a direct and keen interest to me. The DDP and Pass I have selective is a review of global collaboration and the lack thereof among educators to become cohorts and develop inquiry among colleagues to determine who, what and how, drectly collaborates to the development of higher ables teachers and the capacity it affords those whom they teach to demonstrate growth in literacy development.

I welcome an further debate and responses. In the meantime please email me at: [email protected]
The website should be up and running again by October 1.

Make it a powerful day!

Valrie A.Ciemielewski
Reading Specialist/MSE
Kappa Delta Pi
Guest Lecturer
Literacy Coach

Please forgive me, I should have proofread my postings. Mia Culpa.


I am very interested in the subject of closing the achievement gap, especially for African American/Black students and other students of color, and, secondarily, also ESL students. If you have any studies or papers that adress this topic and any statistical analyses that I could look at, I would appreciate that.

In addition, if you could suggest any subject specific curricula (math, reading, grammar, social studies, science, etc.) that you recommend for helping in the aforementioned issue, I'd appreciate hearing your suggestions.

Karen Campbell, parent in Bellevue School District, Bellevue, Washington

Karen Campbell states:

Here is my contact information--it didn't show up in my 12/27/07 4:21 pm posting:

Karen vanHaagen Campbell
Legal Secretary
c/o Perkins Coie LLP
10885 N.E. 4th Street, Suite 700
Bellevue, WA 98004
Fax: 425-635-2400
email: [email protected]

Comments are now closed for this post.


Recent Comments

  • Karen Campbell: Karen Campbell states: Here is my contact information--it didn't show read more
  • Karen Campbell: I am very interested in the subject of closing the read more
  • V. Ciemielewwki: Please forgive me, I should have proofread my postings. Mia read more
  • Valrie A. Ciemielewski CEO/Professor: I am cxurrently mworking on my Doctoral, ED. D. in read more
  • Dale Myers: Dr. Strauss' Power Point presentation shows a statistical correlation, only. read more




Technorati search

» Blogs that link here