Thursday, June 19, 2008

NY Times on SAT & High School grades .....
............. [ WASL?? ]

Tamar Lewin, in the today's NY Times, wrote an article called
"Study finds little benefit in new SAT".
http://www.nytimes.com/2008/06/18/education/18sat.html?ref=education

It describes how the addition of the writing test "did not substantially change how predictive the test is of first-year college performance. " She then goes on to quote in detail a SAT foe and note that many colleges have dropped the requirement for standardized test scores.

As you may know often reporters do not have the background qualifications to analyze statistics properly. The following comes from an eminently qualified research scientist viewing this situation.

Looking at the studies in question.....

It is clear that these are some of the best educational assessment studies ever conducted. Each student population is analyzed with respect to each individual institution rather than the whole: 151,000 students and 110 colleges. The results are then amalgamated and weighted by size of the institution. There is no study of high school math achievement that is in the same galaxy of methodology. No WASL study is in the same universe. The 2 studies show remarkable predictive power. SAT scores are as predictive of freshman college grades as are high school grades. The correlation of the SAT is r=0.53 and high school grades 0.54, and together are 0.62 (a high
correlation >0.6). The SAT+HS is especially predictive in the subgroups of selective colleges (0.65) and small colleges (0.67). The second study examines how close the predictions are and finds males are slightly underestimated and females overestimated by both the SAT and high school grades. All minority subgroups correlate better for the SAT than high school grades, and using both, are predictive for every subgroup above 0.54.

In summary, a 3-3/4 hour test can predict how you will do in the freshman year of college better than high school grades, in all groups except males, where they are very close, and together they are very predictive. Ms. Lewin is way off base here, as the data don't support her bias against standardized tests, specifically the SAT. The WASL would show very different findings.

These findings suggest that while colleges may not require the SAT, and may be willing to look at your math achievement notebook instead, you would be very unwise to forgo a standardized test without very good reason, if you want to get into college.

4 comments:

  1. This was our assessment of about 8,000 AVID students - SAT scores were the best predictor of college success. It may not be politically correct, but I think its better to be honest than have students attending college to take remedial classes. You would be better off enlisting in the military or going to community college and take the same remedial math you'd need at university for a lot less money.

    ReplyDelete
  2. I thought after reading the article it was misleading because Lewin was writing about the new SAT and the study's criticism was with the writing portion of the exam. The conclusion was that it wasn't any better than the old SAT at predicting student outcomes. I can understand that. It shouldn't change the fact the SAT is still a good examination, just that you don't need to build a longer test by adding a writing section which is expensive because essays have to be scored by hand. This is one of the criticisms of the WASL - why not cut costs by eliminating the writing portion of the exam, since there's little value gained by having it.

    ReplyDelete
  3. If a kid takes the SAT the expectation is there, that this person expects to enroll in college, so they've been taking classes that satisfy the college entrance requirements.

    Most kids have a realistic assessment of themselves and make realistic choices. This is not entirely true with Core plus. Most college track students only took 3 years of Core plus or 2 years and 1elective math and then found themselves in a college remedial math program.

    As far as students taking Core plus for instance and the correlation to SAT scores that would be an interesting piece of research. Because there are at least two tracks of kids - those who took Core plus in 8th grade and the regular track kids who took Core plus in 9th grade. It is also possible that an accelerated track used an entirely different book or the teacher supplemented the acc. track with more traditional materials. Studies would have to account for those differences. We'll assume only the accelerated track made it to college.

    Talking to some of the accelerated kids I learned some hadn't even taken Core 1. The principal was making sure that the next crop of kids had to take Core 1. Regressing to the lowest curriculum.

    Also, seniors could take a special elective of college prep (traditional) math.

    Then again, this was a very strange school. You could be sent to the office if you were Hispanic and wearing blue. The only Mexican-American in the Acc track was from California and he was definitely having trouble adjusting to Core plus.

    ReplyDelete
  4. This is a rather unique story, don't ask me how I read about it. It reminded me of the parent imprisoned because his daughter couldn't pass math in high school.

    http://www.lifesitenews.com/ldn/2008/jun/08062009.html

    GATINEAU, Quebec, June 20, 2008 (LifeSiteNews.com) - A Quebec youngster has used the courts to avoid parental discipline in a "landmark" case. The 12-year-old girl, who is too young to be named, went to court to force her father to overturn his decision not to allow her to go on a school trip. Her father had decided to ground her after he found out she had posted photos of herself on a dating website against his wishes.

    The sixth grader then took her father to court, arguing that his punishments were too severe.

    Madam Justice Suzanne Tessier of the Quebec Superior Court ruled today that denying the girl permission to go on the school trip was an excessive punishment. The girl's lawyer, Lucie Fortin, said, "She's becoming a big girl" and described the school trip as "a unique event in her life", the Globe and Mail reported.

    In arguing the case, Fortin cited Sections 159 and 604 of the Quebec Civil Code, which allow minors in some circumstances to initiate court proceedings relating to the exercise of parental authority. Section 159 is used in "extreme circumstances", such as cases of parental negligence.

    The father's lawyer, Kim Beaudoin said that her client is "stunned by this situation. He feels like he's lost his daughter". He is appealing the court's decision.

    Although the child got what she wanted, others are saying that court has blazed new paths into the very heart of the family, which compounds the growing threat of state interference in private life in Canada. Dr. Albert Mohler, president of the Southern Baptist Theological Seminary of the Southern Baptist Convention, called the decision "another chilling precedent", saying that it was so outrageous that it sounded like a parody.

    "This judge needs to be grounded and sent to her room," Mohler wrote. He points out that the rules the girl broke, and which the court has helped her to flout, were put in place for her own protection. Police have identified online dating sites as a common window through which sexual predators gain access to children.

    ReplyDelete