Wednesday, February 08, 2012

Testing

Goffstown Area High School, where Kyle goes and sons 3 and 4 went, finished first in the state in reading on the NECAP test.  We used to finish in the middle of the pack, and we have made no grand changes in our system in the last ten years.  Anyone who knows the schools in the state knows that Goffstown is not the best school and does not have the best students.

So we are gaming the system in some way.  Some methods of gaming the system I would approve of, others not, so I don't know what my opinion is on this yet.  I do know one thing we are doing to raise our students' scores, and I approve of that one.  I doubt it is enough to explain a move from middle to top, however, so I am reserving judgment on the whole thing.

7 comments:

  1. What's the distribution of scores?

    ReplyDelete
  2. That I don't know, nor if I can get my hands on the data. That question does suggest to me one of my suspicions: getting the most out of the students who might be tempted to screw with the test or blow it off is one of the best ways to raise group scores.

    With a small sample - say, just to take a random example, a private Christian school in NH - you can count on having a disproportionate number of kids scoring well, and can even get an occasional kid who blows out the top of the scale, dragging the whole class average with him/her. If you get two of those in one year, you perch atop the state averages.

    That doesn't apply with Goffstown, which floats back and forth between the category of largest and next-largest group of schools every year.

    ReplyDelete
  3. I attended a high school meeting a few years back at which the (now thankfully departed) principal was justifying new initiatives by telling us that Sun Prairie was 3'rd in the district and we needed to be number 1. The difference between average scores for the top 4 schools was of order 0.1%. She declined to address my question.

    ReplyDelete
  4. It might be simply explained by a particularly bright group of students one year. Those sort of bumps happen coincidentally every now and then.

    ALso, I can't find the link right now, but I just read an article about improvement in students'/school's academic performance being tied to higher expectations and emphasis. The school may not be doing anything different, but if the emphasis on certain subjects and the normative expectations are higher than before, that may also cause an increase.

    I'll see if I can find the article.

    ReplyDelete
  5. I'd love to see the article, but am highly suspicious of its premise. I like good numbers, though.

    ReplyDelete
  6. That was only one facet of the study. It's main objective was to look at how class size and teacher performance affected student performance.

    It showed no strong correlation with class size but discussed some differences between teacher strategies in the classroom...high expectations and goals being seen as one of the markers of good teachers and student performance.

    It had more detail than what I can convey here.

    ReplyDelete
  7. I long believed that school size, rather than class size, was important in school culture, especially as it limited the ability of bad actors to congregate and feed off each other. I believe that cherished idea of mine hasn't held up under scrutiny, however.

    ReplyDelete