Creative Destruction

August 16, 2006

The Connecticut Mastery Test. – Ampersand is Wrong.

Filed under: Education,Feminist Issues,Race and Racism — Daran @ 9:28 pm

In his recent post here (also duplicated at Alas), Barry takes issue with the claims made in this news report.


Barry:

Also, as I’ll show below, the evidence from the Connecticut Mastery Test shows that the boy crisis does not exist among “boys of every ethnic and socioeconomic group.” On the contrary, the results are consistent with my belief that without racism and poverty holding them back, boys do just as well as girls.

In fact, the article does not claim that the “boy crisis” exists among “boys of every ethnic and socioeconomic group”. The claim is that “Boys of every ethnic and socioeconomic group are falling far behind girls of similar backgrounds”. This may not be a crisis for boys in those ethnic and socioeconomic groups which, despite their lagging behind comparable girls, still achieve basic competency targets.

Barry nevertheless disputes even this:

…it’s not true that Connecticut Mastery Test found that “Boys of every ethnic and socioeconomic group are falling far behind girls of similar backgrounds.”

and argues instead that

[in] Westport, a town in which 94% of eighth graders (or, at least, of eighth graders who took the Connecticut Mastery Test) are white, and less than 3% qualified for the discount lunch program… 97% of both boys and girls are proficient in math, 96% of boys and 97% of girls are proficient in reading, and 97% of boys and 99% of girls are proficient in writing. There is effectively no difference between Westport’s boys and girls, according to the Connecticut Mastery Test.

In a comment, Barry defends his view in response to my criticism that there is in fact a large proportionate difference between the number of boys and the number of girls failing to achieve proficiency.

You’re talking about differences of one or two percent. I don’t think that’s significant – and certainly not enough to justify claims of a crisis in the education of boys among all socioeconomic and ethnic groups.

The latter claim, as I pointed out earlier, is a misrepresentation.

Using a somewhat convoluted statistical argument I argued that Barry was wrong to conclude that small differences at the margin of the distribution were not significant, and that they were entirely consistent with a large difference in the mean scores for boys and girls. It turns out that this argument was unnecessary: The CMT website will report the average scores directly. Here’s what I found:

(Edit: embedding the image didn’t display properly, at least, not in my browser. Instead, open this link in a new browser window/tab. For those unable to view the image, here’s a CSV text file with the same data.)

In both Bridgeport and Westport, boys and girls had much the same score for mathematics, with boys doing slightly better than girls in Bridgeport, and slightly worse in Westport. The pattern is very different for reading and writing. In Bridgeport, the boys lag by 6.9 and 14.7 points respectively, while in Westport, the lag is 9.3 and 20.6. Expressed as a percentage of the girl’s score, the boys in Bridgeport scored 97% and 94% for reading and writing, while in Westport, the scores were 97% and 93%. Similar results (with slightly lower overall scores) are obtained if you select white students paying the full lunch price statewide.

In conclusion then, Barry is correct to say that there is no crisis for boys “among all socioeconomic and ethnic groups”, but that was never the claim in the article. He was wrong to fault the article for claiming that “Boys of every ethnic and socioeconomic group are falling far behind girls of similar backgrounds.”. One might legitimately dispute the word “far”, but that was not his objection.

Barry can certainly claim support here for his final point:

4. Finally, as big as the 20% difference in reading achievement between Bridgeport boys and girls is, let’s not overlook the much larger differences in achievement between children in Bridgeport and children in Westport. That’s the real crisis, and that should be our main focus.

Expressed as a percentage of the scores for Westport, students in Bridgeport score 74%, 77% and 77% for mathematics, reading, and writing. Clearly these deficits dwarf the gender gaps.

Equally clearly, the gender gap has a disproportionaly large incremental effect on boys who are already disadvantaged by reason of race and social group. The 6% gender gap in average score for writing in Bridgeport translates into a 20% higher failure rate for boys. I still maintain that a focus on the gender gap is warrented on the grounds that we don’t understand it. If we did, it may turn out that improving boys performance by 6% across the board without harming girls in the process proves to be a far less intractable problem than achieving a 6% improvement for both genders within the disadvantaged classes.

Any such improvement will only come about if we have a credible theory as to what is causing the gender gap in the first place. That theory must account for the fact that the gap does indeed apply to “all socioeconomic and racial groups”. The “boy brain” theory does. Barry’s theories don’t.

2 Comments »

  1. I find it shocking that children in Bridgeport did comparatively well to children in Westport! Shocking! You maybe should recheck those facts.

    Comment by marge ingersol — August 17, 2006 @ 4:18 am | Reply

  2. I find it shocking that children in Bridgeport did comparatively well to children in Westport! Shocking! You maybe should recheck those facts.

    You should maybe recheck what I said:

    Expressed as a percentage of the scores for Westport, students in Bridgeport score 74%, 77% and 77% for mathematics, reading, and writing. Clearly these deficits dwarf the gender gaps.

    Comment by Daran — August 17, 2006 @ 6:12 am | Reply


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: