Comments on: Class Size and SAT Scores By State

By: Best of FlowingData: November 2009 « Earth to Bobby

Best of FlowingData: November 2009 « Earth to Bobby — Thu, 03 Dec 2009 06:13:22 +0000

[...] Class Size and SAT Scores By State [...]

By: Andrew

Andrew — Wed, 02 Dec 2009 17:59:18 +0000

Good, but too many unaccounted for variables. School with low class size are typically private or higher economic and schools with high class sizes are typically urban and lower economic. (property tax)

Different states may have different class size averages but what about; money spent per student? who is tested? how many plan on going to college? Socio-Economics? state standards? etc.. there are just too may other variable to make this data valid.

By: Do smaller classes mean higher SAT scores? « Grockit Blog

Do smaller classes mean higher SAT scores? « Grockit Blog — Fri, 20 Nov 2009 22:28:41 +0000

[...] the question FlowingData posed last week. Nathan Yau examined the average SAT scores and high school class sizes for every state, using a [...]

By: paresh shah

paresh shah — Sun, 15 Nov 2009 15:51:52 +0000

It is a pity – it was indeed a beautiful chart!!

Will surely use the graphic in other circumstances.

By: Thomas Lotze

Thomas Lotze — Sat, 14 Nov 2009 09:12:25 +0000

Everyone complains about the weather, no one does anything about it…

Scatterplot. Enjoy!

I actually made an interactive scatterplot, with some additional education data (including percent of students who take the SAT). From that, you can definitely see the two groups (states where many students take the SAT, versus states where few do). While each of the two groups seems to have an effect due to class size, overall the self-selection seems like a bigger determinant of the average SAT score (when fewer students take the SAT, only the better students tend to take the SAT).

By: Thomas Lotze

Thomas Lotze — Sat, 14 Nov 2009 09:06:59 +0000

[...] couple of days ago, Nathan at FlowingData posted a parallel coordinates plot showing a relationship between student-to-teacher ratio and mean SAT scores. In the comments, a lot [...]

By: Michael

Michael — Fri, 13 Nov 2009 02:51:52 +0000

I’d also suggest reversing the direction of the left axis (average student-to-teacher ratio) in the parallel coordinate plots. If a correlation is assumed to exist, it would be easier to see that among a set of nearly parallel lines instead of an X shape indicating an inverse relationship.

By: Gideon

Gideon — Wed, 11 Nov 2009 19:26:57 +0000

It’s really difficult to read correlation on parallel coordinate plots.
http://bit.ly/18Ij5O

I have a feeling that regression is the strongest force at play here. What do the number say?

By: Class Size and SAT Scores By State | VizWorld.com

Class Size and SAT Scores By State | VizWorld.com — Wed, 11 Nov 2009 15:03:12 +0000

[...] Class Size and SAT Scores By State | FlowingData. Related News & Resources The State of OpenCL Montana State University’s new [...]

By: Daily interesting articles | Aww Dip

Daily interesting articles | Aww Dip — Wed, 11 Nov 2009 06:09:50 +0000

[...] helped clean teeth instead?Student Work – David Biglow All of this packaging looks gorgeous.Class Size and SAT Scores By State While all a correlation, it is interesting to visualize the links between the two factors.Pregnadel [...]