October 5, 2012

SAT and ACT: How hard are they scraping the bottom of the barrel and are they finding any diamonds in the rough?

There was recently some publicity over the national average SAT and ACT college admission test scores dropping. The Unsilenced Science has a vast post with dozens of graphs on the latest SAT and ACT college entrance scores by demographics and state. 

However, there has always been an issue with tracking changes in average SAT/ACT scores in that not everybody takes either of the two tests, and (increasingly) some students take both tests.

However, in the comments, The U.S. and I go back and forth on how to deal with this perpetual stumbling block in SAT/ACT analyses and he points out that a number of states have recently made the SAT or ACT mandatory. 
Here are the significant, single-year participation changes: Colorado’s ACT participation went from 62% to 99% in 2002, and its ACT score fell from 21.5 to 20.1. (SAT-ACT composite fell from 501 to 472.) Illinois went from 71% to 99% in 2002 with a drop from 21.6 to 20.1 (498 to 464). Kentucky went from 72% to 100% in 2009 with a drop from 20.9 to 19.4 (478 to 444). Michigan went from 70% to 100% in 2008 with a drop from 21.5 to 19.6 (492 to 448). Utah went from 73% to 97% in 2012 with a drop from 21.8 to 20.7 (493 to 469). Maine SAT participation went from 73% to 100% in 2007, and its SAT-ACT composite fell from 501 to 469. Delaware went from 74% to 100% in 2012, and its SAT-ACT composite fell from 492 to 465. In fact, Arizona’s ACT participation more than doubled between 2009 and 2011 from 15% to 34% with a similar drop from 21.9 to 19.7. (The SAT-ACT composite went from 508 to 474 as SAT participation went from 26% to 28%.)

I'd never seen that data before and it's pretty interesting. I responded, doing some top of the head calculations:
So from this we can calculate the average score of the incremental test takers. For example, in Illinois, when 71% took the SAT, the average was 21.6, but when 99% took it, the average fell to 20.1. So, all else being equal, that means the incremental 28% averaged a 17, which is 4.6 points lower than what the 71% averaged, and 4.6 is pretty close to one standard deviation.  
If the additional 28% came from the bottom of the distribution, the 2nd to 28th percentiles, then the median of the incremental test takers would be about the 14th or 15th percentiles overall, while the median of the previous 71% would be about the 64th percentile. That would imply a drop of around 6 to 6.5 points, maybe, but instead the drop was about 3/4ths of that. So, some of the incremental scorers came from above the bottom 28% -- although not too many.  
It would be interesting to see if this law made any measurable increase in high scorers (above 30 or above 25) and which race they most came from. I'm betting offhand that the new law rousted more semi-smart white kids into taking the ACT than other races. Probably kids who were set on going into the Army and had done well on the AFQT, that kind of thing.

Keep in mind that 100% participation doesn't mean 100% of all 17 year olds in the state take the SAT or ACT, just, I'm presuming, 100% of the non-dropouts.

So, I think the data is there to answer some questions about the remaining untested students, both including what are the likelihood of them succeeding in college and how many who have the right stuff to make it in college are being overlooked.

Secondly, this information on the scores of the bottom untested ranks should allow somebody to adjust the national figures for participation and track changes over time at the national level, which would be of obvious import.

24 comments:

Tertius Lydgate said...

I took both. The ACT is easier, and gives you less time per question.

Gringo said...

This is the logical consequence of the "everybody should go to college" meme. Contrast this with Switzerland and Germany, which have good technical training for those not going to college. The problem here is that the tracking involved wouldn't go over well in the US.

There are diamonds in the rough. I can think of several in my short career as a math teacher who had an unexpected great test score- which as far as I could tell didn't involve cheating. As math is cumulative, this is even more amazing.

Unfortunately, it is not that easy to keep such students going in the right direction.

Are there some bright people who do poorly on standardized tests because they blow them off? I am sure there are, but it would seem difficult to discover them, for that very reason.

There are a lot of students who are fed up to here with standardized tests. While they are probably quite justified in having such an attitude, that attitude will not help them get good scores.

Public schools do not serve well those students who are not academically inclined. The dismantling of industrial arts courses is one example.

james said...

The SAT math plots were interesting. It would be useful to compare the before/after distributions for the states that made the SAT mandatory.

Anonymous said...

The great tragedy here is that there are guys deep in the bowels of places like ETS-Princeton who know the answers to ALL of the questions that you and Charles Murray and La Griffe and everyone else are asking.

[Such as, relative to the (definitional) Caucasian standard deviation of 15, is the Negro standard deviation for IQ also 15, or is it something more like 12?!? And is Charles Murray correct in fearing that Negro mean IQ may have suddenly plummeted, in just one generation, from 83.7, to 80.2?]

But, of course, the prevailing atmosphere of Stalinistic Political Correctness makes it impossible for anyone in the know to step forward and give us the answers to our questions.

Anonymous said...

The US continues its march in making sure that the majority of graduates can talk about 'critical marxist theory in gender relations via World of Warcraft' but no one knows how to build a doghouse or solder metal.

Truth said...

" The dismantling of industrial arts courses is one example."

Not the dismantling of industrial arts courses, the dismantling of industrial arts CAREERS.

Anonymous said...

There are a lot of smart kids who just want to go to community college. I scored a 29 on the ACT(didn't take the SAT) and went to a Big 10 school for Computer Science, but two of my best friends scored a 25 and a 24... and went to a community college to become a mechanic and a EMT. Everyone takes the ACT here, but if it had not been required I'm not sure they would have have taken the test.

There's also the scholarship factor and I don't think scholarships tied to community colleges really care much about ACT or SAT scores... but they really care about high school GPAs(and local connections).

There are a lot of smart people out there that don't aim too high.

Anonymous said...

Just by anecdotal evidence, I observed that black IQ SD seems rather low.

I spent 3 months entering data that happened to be SAT scores into a unified database. This was a long time ago, around 1985.

The scores were grouped by schools and the schools were high schools around the Atlanta and other Georgia metro area.

For black high schools, I would enter the data for an entire high school, hundreds of scores sometimes, and there wouldn't be a single score above 1000. Scores above 900 jumped out, but none, in an entire school, above 1k.

For school, after school, after school. This made a pretty big impression on me at the time, I remember wondering, isn't there at least 1 diamond in the rough somewhere here in one of these obviously all black schools, just one?

After noticing the pattern, which didn't take long, a day or two, I was on the lookout for some overlooked genius or even moderately bright kid at one of these schools, but never saw a single one.

Not a single one, in literally dozens of these schools.

So I wonder if the the black IQ distribution even follows the bell curve (which is just a general model, not a prescriptive solution for this case) and even if it does, yes, what is the standard deviation.

Anonymous said...

Off topic, but this is related to Steve's observation that the hispanic population, despite being the largest minority group, has a dearth of achievement in this nation.

CBS is running a series of ads celebrating hispanic heritage month. So I assumed they would list the achievements of hispanics, which according to Merriam-Webser: "are of, relating to, or being a person of Latin American descent living in the United States."

Tonight CBS featured Gabriel García Márquez, a Columbian Nobel winner who is not even a US resident let alone citizen, in this ad .

Maybe Steve's off the cuff observation is true and there is a lack of achievement among this group. Evidently there are not even enough hispanics to fill this month long celebration.

Luke Lea said...

Noah Rosenberg, et al:

Previously, we observed that without using prior information about individual sampling locations, a clustering algorithm applied to multilocus genotypes from worldwide human populations produced genetic clusters largely coincident with major geographic regions. It has been argued, however, that the degree of clustering is diminished by use of samples with greater uniformity in geographic distribution, and that the clusters we identified were a consequence of uneven sampling along genetic clines. Expanding our earlier dataset from 377 to 993 markers, we systematically examine the influence of several study design variables—sample size, number of loci, number of clusters, assumptions about correlations in allele frequencies across populations, and the geographic dispersion of the sample—on the “clusteredness” of individuals. With all other variables held constant, geographic dispersion is seen to have comparatively little effect on the degree of clustering. Examination of the relationship between genetic and geographic distance supports a view in which the clusters arise not as an artifact of the sampling scheme, but from small discontinuous jumps in genetic distance for most population pairs on opposite sides of geographic barriers, in comparison with genetic distance for pairs on the same side. Thus, analysis of the 993-locus dataset corroborates our earlier results: if enough markers are used with a sufficiently large worldwide sample, individuals can be partitioned into genetic clusters that match major geographic subdivisions of the globe, with some individuals from intermediate geographic locations having mixed membership in the clusters that correspond to neighboring regions.

jody said...

my opinion is that they are not going to find any diamonds in the rough. with affirmative action, there's already A LOT to gain in life if you're a very smart NAM, and you'd already be taking the SAT or ACT if it was 1970, 1980, 1990, or 2000. there were not lots of undiscovered highly intelligent NAMs chilling, hiding out, not taking these tests 30 years ago.

one thing i did think was interesting is that as participation rates went up, "white" scores...went up. not by a lot. i think the last results i saw were that SAT math scores went up 2 points from 2010 to 2011. and the trend over the last 5 to 10 years is very slowly, but steadily up.

but that's crazy. you increase the participation rate and the average score is supposed to go down. but it went up instead. and that's before taking into account that:

1) a certain percentage of the people in the "white" category are mestizos from latin america, arabs, other middle easterners, north africans, and central asians. their scores will bring the average "white" score down, and their percentage of the "white" category keeps slowing growing.

2) the number of ashkenazi jews, who are also in the "white" category, keeps going down from year to year. their scores will bring the average "white" score up, but their percentage of the "white" category keeps dropping.

so you have 3 trends which should brings the scores down:

1) very large effect of increasing participation rate
2) small but growing effect of increasing number of non-europeans in the "white" category via immigration and high birth rates
2) small (but decreasing) effect of jews going away through low birth rates

yet the scores trend upwards. perhaps idiocracy has not set in after all when taking european americans in isolation.

jody said...

"Maybe Steve's off the cuff observation is true and there is a lack of achievement among this group."

it's not off the cuff. it's something me and steve have been exploring for years.

about 3 years ago i deliberately went looking for mexicans in all fields, and after a couple days of checking for high performers in a dozen disparate human endeavors, posted my profoundly meager list here on steve's blog.

other than democrat votes and GDP creation simply by virtue of breathing and consuming, mexicans offer the united states vanishingly little. they're a black hole of human achievement.

this does not make them unusual. in fact, they're actually quite normal in that regard. most human groups produce almost nothing of note. there are several 100 million person wide swaths of humanity on earth which are generating absolutely nothing. we could go live on another planet for 200 years and come back and their society would not have changed at all. for various european societies at a 2012 culture and tech level, that kind of societal and technologicaly stasis is an alien concept. technological and cultural change frozen for 200 years? but that's actually the norm for humans, not the exception. modern europeans are almost unique with regard to how change occurs continuously in their nations, and high human achievement in multiple fields simultaneously is the driver of this change.

the drive to continously advance our engineering, conquer disease disorders and death through medicine, improve our society through politics, achieve self actualization through the creation of masterpiece level literature music or video, pioneer new forms of entertainment and sports, and explore our planet and universe, these things are inherent in modern european psychology and almost alien to the psychology of most other human groups. even if they possess the biological equipment necessary to participate in those endeavors on a large scale, which is a matter of debate, they do not seem to be interested in the pursuit of much beyond eating, sleeping, and reproducing.

Cail Corishev said...

"There were not lots of undiscovered highly intelligent NAMs chilling, hiding out, not taking these tests 30 years ago."

True. Any teacher who thought he had such a student would have done everything he could to get the kid to take such tests, sending off for applications, helping fill them in, driving him to the test, and so on. It's exciting for a teacher to discover a one-in-a-million kid in his class, and having him be a NAM would be the cherry on top.

The movie trope is of the minority genius who comes from a terrible family and neighborhood, so he refuses to do well because he's been told he's useless by abusive parents, or he doesn't want to stand out from the other kids, or whatever; so no one realizes he's smart until that one special Teacher With Heart comes along. (Before this trope was about minorities, it'd be a white kid from a poor, hard-luck family who couldn't take the test because he had to go to work in the mines to support his family.)

This scenario has probably existed in real life at least once somewhere, but it's mostly a myth. It's not that hard for teachers to spot intelligence; horrible families aren't that likely to produce geniuses; and geniuses are likely to look for ways to get away from their horrible families. A lot of unlikely things have to come together to cause a diamond in the rough to be missed.

The Legendary Linda said...

According to the book the bell curve there are virtually no diamonds in the rough. They argue that virtually 100% of American 17 year olds capable of scoring above 700 on either section of the old SAT did indeed take the test (and that percentage would be even higher today).

This makes sense if you think about it. The higher your IQ, the greater the probability that you will take the SAT (either because of your own ambitions or the vicarious ambitions of parents and teachers around you)until eventually, you reach an IQ level where virtually 100% of the people take the test.

Traditionally it's been assumed that the SAT distribution needs to be adjusted by a factor of 3 at the high levels. For example in 1984 only 5 people scored 1590+ on the old SAT out of the roughly one million people who took the test. However THREE MILLION Americans were in the age group that takes the test (even if only a million of them were academically ambitious enough to actually do so). Since it is assumed that 100% of the late teens CAPABLE of scoring 1590+ on the SAT actually took the test, a 1590 old SAT score is assumed to represent not just the top 5 out of the million test takers, but the top 5 out of all 3 million late teens in America in 1984 (and whatever diamonds in the rough might be missed would be roughly balanced by foreign students). Thus 1590+ is assumed to represent a 1 in 600,000 level intellect (IQ 170).

rob said...

Since ETS changed the SAT to reduce g-loading by removing analogies, thinky-type math, and whatever else, the test is less capable of finding smart but badly-educated kids. The test isn't designed to find diamonds in the rough. It's more designed to find the most highly polished turds.

Lucius, (you are Lucius, right?) you're Charles Murray link is to this very comment thread. Could you post the actual link?

The Legendary Linda said...

Just by anecdotal evidence, I observed that black IQ SD seems rather low.

I spent 3 months entering data that happened to be SAT scores into a unified database. This was a long time ago, around 1985.

The scores were grouped by schools and the schools were high schools around the Atlanta and other Georgia metro area.

For black high schools, I would enter the data for an entire high school, hundreds of scores sometimes, and there wouldn't be a single score above 1000. Scores above 900 jumped out, but none, in an entire school, above 1k.

For school, after school, after school. This made a pretty big impression on me at the time, I remember wondering, isn't there at least 1 diamond in the rough somewhere here in one of these obviously all black schools, just one?


You mentioned these students were from Georgia. Jensen found that black students in an entire school district in rural Georgia had an average IQ around 70 with an SD of 15 and the scores were normally distributed.

An SAT score of 1000 on the old SAT equates to an IQ around 117, which would be more than 3 SD above the mean Jensen found for black students in Georgia, so it's not surprising that such students would be very few and far between.

It's also likely that in these all black intercity schools where blacks are especially disadvantaged, teachers don't care, and academic achievers are considered Oreo pariahs, the few blacks with IQ's above 115 are so traumatized by being surrounded by classmates with IQ's of 70 and teachers with IQ's of 90 and getting beat up for their good grades that they go to the opposite extreme: rationally deciding to drop out of high school to become gangsters, prostitutes and rappers, rather than taking the SAT when they know the system is rigged and affirmative action doesn't benefit hardcore blacks like them.

Usually criminality is associated with lower IQ, but there was one study that found that among blacks, criminality was associated with higher IQ, though this might be have been a fluke study.

The Anti-Gnostic said...

Not the dismantling of industrial arts courses, the dismantling of industrial arts CAREERS.

Truth speaks truth. The message to young men is, forget industrial arts. We are going to offshore plants and inshore labor. Any time anybody gets too uppity, we will replace them with ever-cheaper Meso-Americans who don't share your Teutonic precision and craftsman's pride. So get busy and invent iPads or suck it up with your welfare-lifestyle at the bottom of the barrel.

Anonymous said...


Public schools do not serve well those students who are not academically inclined. The dismantling of industrial arts courses is one example.


In other words, schools don't well serve those who don't like school.

For such people, there were apprenticeships and on the job training, also good old military service. The first two are gone and the third only wants the people who have done well in school.

Statsquatch said...

There is definitely overlap between the scores of new and old takers in Illinois. An average score of 17 would correspond to a sample from normal distribution truncated above the 45th percentile. The other part of that distribution would have an average of 25.

To anonymous, I do not think their are any bowels in the ETS. It is located in a lovely farm type atmosphere.

Anonymous said...

Lucius, (you are Lucius, right?) you're Charles Murray link is to this very comment thread. Could you post the actual link?

Sorry, it's Footnote #44:

"Did the children do better? A total of 716 of them were tested with a highly g-loaded verbal test, the Peabody Picture Vocabulary Test (revised). The mean of the subset of mothers whose children were tested was 83.7. The mean of their children was 80.2. The mothers and children were tested with different instruments, so it should not be concluded that the black mean actually went down in the new generation. But these data certainly give no reason to think it went up."

Cail Corishev said...

"It's also likely that in these all black intercity schools where blacks are especially disadvantaged, teachers don't care, and academic achievers are considered Oreo pariahs, the few blacks with IQ's above 115 are so traumatized by being surrounded by classmates with IQ's of 70 and teachers with IQ's of 90 and getting beat up for their good grades that they go to the opposite extreme: rationally deciding to drop out of high school to become gangsters, prostitutes and rappers, rather than taking the SAT when they know the system is rigged and affirmative action doesn't benefit hardcore blacks like them."

Except that affirmative action does benefit them. You could even say the smart NAM from a bad background is the greatest possible beneficiary from AA. There are people desperately looking for him, and if they find him, he can sail right through a degree into a nice sinecure in government or academia somewhere.

As I said earlier, what you describe is certainly the standard movie trope. I just doubt that it exists very often in real life. It requires that the kid not want out of the 'hood, that none of his teachers care about him or want the considerable credit for finding a diamond in the rough (remember, we're blank slates, so a teacher who finds a swan gets a sort of credit for causing it), and that none of the people out there looking for smart NAMs to prove the naysayers wrong ever run across him.

I'm not saying it can't happen, but I think it's about as likely today as a kid growing to 6'8" by his senior year and never being handed a basketball.

Statsquatch said...

meant to say "there are no bowels"

The Legendary Linda said...

"

Except that affirmative action does benefit them. You could even say the smart NAM from a bad background is the greatest possible beneficiary from AA. There are people desperately looking for him

Nobody goes looking in the worst schools in the country.  Most whites are terrified to set foot in there.

As I said earlier, what you describe is certainly the standard movie trope. I just doubt that it exists very often in real life. It requires that the kid not want out of the 'hood, that none of his teachers care about him or want the considerable credit for finding a diamond in the rough

Teachers assume all their ghetto students are stupid and the students live down to their expectations, and the teachers don't get any credit for finding a diamond in the rough, they get blamed for wasting time on just one student when dozens of others can't read.


(remember, we're blank slates, so a teacher who finds a swan gets a sort of credit for causing it), and that none of the people out there looking for smart NAMs to prove the naysayers wrong ever run across him.

Teachers get credit if the class average goes up; they don't get credit for ONE student.

I'm not saying it can't happen, but I think it's about as likely today as a kid growing to 6'8" by his senior year and never being handed a basketball.

Which is very likely if he never stands up because he'll be beat up for being tall, and if the school has no basketball team because kids at that school are a priori  considered unable to compete. 

Anonymous said...

Teachers get credit if the class average goes up; they don't get credit for ONE student.

A single IQ 115 student, in a school district full of IQ 70-ish students, will immediately send his classroom to the very top of the district mean [average].

On the other hand, though, he won't do diddly squat for his classroom's median...