<p>There may be a state by state effect as law changes in a few states. Many Colorado colleges experienced unexpected rise in out of state applications this year even though the new law should not affect under-21 year olds. The reasons for the rise in applications are unknown (they could be related to the drug law changes in part) but legal changes and enforcement changes should obviously have some effect on college students but too early to tell exactly what.</p>
Elite colleges generally do not have a big problem with students failing out. Among elite colleges, one of the characteristics that seems to correlate best with 6-year graduation rate is percent tech majors. For example, the 6-year graduation rate for HYPSMC are below, along with the percent tech majors reported in their CDSs:</p>
<p>Harvard – 98% grad rate, 3% tech majors (using older CDS since no new ones seem to be available)
Yale – 98% grad rate, 6% tech majors
Princeton – 97% grad rate, 19% tech majors
Stanford – 95% grad rate, 26% tech majors
MIT – 92% grad rate, 56% tech majors
Caltech – 92% grad rate, 57% tech majors</p>
<p>MIT and Caltech have the best combined SAT numbers in this group, yet they have the lowest grad rate. I expect the lower grad rate than peers relates to a much larger portion of the class being tech majors. Tech majors are probably more likely to take time off from school for co-ops, internships, work, or a startup. They also generally have more course requirements within the major to graduate. </p>
<p>@Data10 How’d you get % tech majors from the CDS? That’s an astute observation, but now I want to go through colleges and see how it holds up throughout AAU colleges.</p>
<p>Section J of the CDS mentions the percentage of bachelors degrees conferred in various fields. What I was called “percentage tech majors” was the sum of the percentages of degrees conferred in the categories labeled Computer and Information Sciences, Engineering, and Engineering Technologies. Note that this relationship is not expected to hold when comparing colleges with notably varied degrees of selectivity, grad requirements, or students having trouble being able to pay for school.</p>
<p>But I don’t think anyone will complain much about 92% vs. 95% graduation rate and some might even argue that the lower number may be closer to optimal (some superstars at the Ivies drop out of college to start a business or to care for a sick relative or to enter religious life or to start an acting or music career and we shouldn’t worry too much if 1 out of 15 students does something like that) - it is when the numbers fall significantly lower that you start wondering about bad advising, pressure, substance abuse, partying, poor college culture, poor continuing student financial aid etc.</p>
<p>I remember the Engineering Dean at Columbia bragging about how high their Engineering graduation rate was (probably higher than some other majors) so problems with engineering are probably not a significant drag on the Ivies and engineers certainly do transfer to other majors and graduate as well (changing majors is relatively easier at those skills). Tech focused private schools in the next tier (Case Western, RPI were near the bottom for private research universities with completion rates in the low 80s) do ok for graduation - obviously lower than similar LAC but not too bad.</p>
As mentioned in my earlier post, what I am calling “tech majors” includes only engineering and CS, not chemistry and physics. I’d expect chemistry and physics would not show the degree of graduation rate influence as CS/engineering since fewer chem/physics majors pursue co-ops or take a break from school to pursue work/startups, and chem/physics generally require fewer credits within the major to graduate than CS/engineering. I am using the CDS for the reported percentages in these majors. As mentioned in your quote, I was unable to find a recent CDS for Harvard, so the value was a few years out of date. Searching around, I did find a more recent 2012 CDS at <a href=“http://oir.harvard.edu/files/huoir/files/harvard_cds_2011-2012.pdf”>http://oir.harvard.edu/files/huoir/files/harvard_cds_2011-2012.pdf</a> . It mentions 2% CS and 3% engineering, for a total of 5%… a significant increase. Harvard has been making a strong effort to increase their engineering presence in recent years, so the increase is not surprising.</p>
<p>
It depends how you look at it. 92 vs 95% might not sound like much, but it can be a large difference in relation to ranking of peers. For example, in the IPEDs data Caltech does not rank among the 50 colleges with best 6-year graduation rate. It ranks below many less selective colleges, such as William and Mary and Tufts. Some on this forum (and some donating alumni) care about how colleges rank in relation to their peers by almost any measure. Many college ranking lists also care about graduation rank measures, including USNWR. </p>
<p>Engineering had the strongest negative correlation with graduation rates; health profession, other technical, fine arts, physical science, and biological science also had negative correlations. Social science, English, history/political science, business, other, and humanities had positive correlations.</p>
<p>The majors with negative correlations with graduation rates tend to have longer prerequisite sequences (so missing/dropping/failing a course is more likely to delay graduation), large volume of requirements (again so that missing/dropping/failing a course is more likely to delay graduation), high workload or time commitment courses (labs, visual art studio, performing art performance) which can reduce the credit load a student can handle, particularly if working part time, and/or a minimum level of rigor that is relatively high (perhaps due to sequenced prerequisites).</p>
<p>Engineering-focused schools may have lower graduation rates because students who change out of engineering as a major are more likely to have to transfer away from the school. Note that attrition out of engineering is quite high at average or lower selectivity levels – something like 40% of prospective engineering majors change to a non-engineering major by second year. That attrition is much lower at highly selective schools, but may still be noticeable in comparison to other majors at the same schools.</p>
<p>Is it really saying on page 17 and in Appendix A 42 that the #1 positive correlation with 4 year graduation rates is “Sex: female” (above grades and a lot of other factors, including being of Asian descent)? I guess I am not surprised but it is painful to see the magnitude of the difference. The female advantage is less in 6 year numbers as they note but still …</p>
<p>Also “Raising a family” is the most important personal desire in their long list (see page 41, Appendix A) that helps with graduation. Interesting. I guess it makes sense but not what I expected.</p>
Note that the coefficient variables in the table have not been rescaled to a consistent SD. Instead GPA is range 0-8, SAT is range 20-160, and Gender is range 0-1. This makes it difficult to compare them. A more meaningful measure is the percent of variation explained in 6-year graduation by the criteria</p>
<p>GPA only – 13.5% of variance explained
GPA+SAT – 16.8% of variance explained
GPA+SAT+Gender – 17.1% of variance explained</p>
<p>Gender adds little to the accuracy of the prediction, and the little is does add likely relates to correlations between gender and major selection.</p>
<p>I feel sad for those like this poster who goes to a well known high school (not a magnet school) which attracts many who want the “best” public high school (especially high concentration of Asian families). As this thread and others like it show - they are in a bad position when trying to get admission to a whole class of schools due to their parent’s decision to “pick the best public school” for their child. This poster stated she had a 3.6 (unweighted GPA) which sounds good to me (especially with good test scores etc.) but would need a MUCH higher GPA to gain admission to the flagship state school. If she had to apply without using test scores (or where class rank is the only or main criteria like UT, or a principal criteria like A&M) she would be at a HUGE disadvantage, all because her parents chose to move where they thought the “best” public high school in her area was. Seems unfair that someone this smart might not be able to go to even the huge state schools UT or A&M (or equivalent) because of her parent’s choice of HS.</p>
<p>@2018RiceParent I’m not sure if it actually hurts people. I went to a school that definitely seems to command the respect of a “top public school” since it’s got the highest ranking (USNWR and Newsweek) in the area; its best days are well behind it (it used to be Top 100 for several years) but it outperforms the nation, state, and other public schools in the area by a significant margin- i.e., people move so they can go to this school.</p>
<p>It looks like UT knows this and doesn’t penalize low GPAs as much. So does A&M. We had 585 students go to college this year- 81 to UT-Austin and 59 to A&M College Station. We had some good representation in other state flagships as well- mostly in the South but a few in UCLA as well.</p>
<p>Assuming an average yield rate (which I’d say is an overestimate), that means 29.3% (172 people) got into UT and 20.1% (117 people) got into A&M. Even if you just look at the people who opted to attend, it’s well beyond 10% (and most people in the top 10% chose other places over UT and A&M, although it looks like A&M’s National Merit money and that Italy trip certainly tempted a bunch of people).</p>
<p>Anecdotally, I know some people who got in with sub-par GPAs (and in one case, this guy got into Cockrell PetroE with a really low test score for our school). A lot of my friends outside the top 10% made it into McCombs and Cockrell, although Plan II and other Honors programs didn’t seem to be as welcoming to them. Maybe it’s because we’re a traditional UT feeder (1% of their incoming class is us) and have a strong UT/A&M community- or because we have strong emphasis on getting into college and are supplied information through our counselors and encouraged by teachers.</p>
<p>So I doubt that choosing a better high school education hurts someone’s chances of getting into UT/A&M- and I bet colleges look at the School Profile and understand that rigor varies between schools without the aid of a test score. In any case, I hope no one chooses a weaker high school so they can get a higher GPA- education matters, especially when you’re building your fundamentals; mislearning Pre-Cal might boost someone’s GPA but it’s not going to help when they actually use that math in college.</p>
<p>@2018RiceParent I’d say a good portion is the schools’ fault for not designing reasonable class rank policies. My school made it a point to only weight 4 core courses (so as not to encourage taking AP Art just to inflate rank) and provide Honors level in all 4 core courses every year. Not only did that expand the size of our Top 7% (probably going down to 6% soon as the 75-25 rule changes) but it made it class rank less stressful without destroying the useful measure. </p>
<p>Also, I heard that if you apply without submitting a rank, UT reranks you based on the entire state- which favors top quarter kids at competitive schools who didn’t make top 7% or 10%.</p>
<p>This data confirms what past posters have been saying about high graduation rates of top schools. Please checkout the figure #2 graph on page 13 of recent college selectivity research showing the direct correlation between SAT scores and graduation rates. The graph shows the higher the SAT - the higher the graduation rate. According to the research , if you have a really low SAT score like those mentioned in the Philly article about top Philly students who score in the 800s out of 1600 on the SAT having them go to a more selective college like Temple will have little affect on their graduation rate.</p>
The study found that SAT score had a somewhat notable correlation with graduation rate prior to adding any controls, but when they added individual and institutional controls for things like HS GPA, HS math curriculum, and race, then a +/- 1 standard deviation difference in SAT score only influenced grad rate by ~2%. This is essentially the same relationship that every other study mentioned in this thread found – SAT score alone is somewhat correlated with various measures of academic success in college, but when you consider the rest of the application (add in controls of the rest of the application), then SAT score adds relatively little to the prediction of academic success beyond the combined information available in the rest of the application. The author of the study came to this same conclusion by saying the following in the conclusion section of the study:</p>
<p>“Put differently, from the standpoint of an individual student, choosing to enroll at a college whose average admissions test scores are substantially higher or lower does not appear to help or harm her chances of graduating.”</p>
<p>@Data10 That is correct. A high SAT student will graduate at about the same rate at a low selective school compared to going to a highly selective school. This is true about a low SAT students as well. Their attendance at a more selective school will not substantially improve their graduation rate. </p>
<p>The data from the research shows the direct relationship between SAT scores and graduation rates. This is important from the aspect of college admissions because high SAT correlates well with high graduation rates. Low SAT also correlates well with low graduation rates. Another reason why SAT is important for ranking purposes.</p>
<p>The data also shows that when a student’s SAT is above 1 standard deviation the rate of change for the probability of graduating gets smaller the higher the SAT score. This explains why those test optional schools see little difference in freshman GPA of submitters and non-submitters. The difference is not that substantial at high average SAT levels which many of these test optional schools are. </p>
<p>For instance, Bowdoin has a mean SAT of about 1400 for submitters and about 1300 for non-submitters. Based upon the graph one would not see that much of a difference in graduation rates from submitter and non-submitter . Some how Bowdoin is able to select relatively high SAT scoring students from the non-submitters. Had Bowdoin selected high HSGPA students with average SAT scores from the non-submitters, it is clear from the data that Bowdoin would see marked differences in graduation rates of non-submitters vs submitters.</p>
When adding controls, the study found only a ~2% difference in graduation rate for a test score 1 SD away from the mean of the college. If the average student at that college had a 72% chance of graduating, the 1 SD low SAT scoring student would have a 70% chance of graduating. I certainly wouldn’t call this slight difference a strong correlation. The author of the study goes so far as to say that the correlation is too small to be statistically significant, as quoted below:</p>
<p>" The ATT indicates that there could be a small SAT effect on graduation (2.2 percentage points for a standard deviation increase in college SAT), but this does not reach statistical significance. The ATU is much smaller in magnitude and is not significantly different from zero."</p>
<p>
Note that the study compared the rate of graduation for students with test scores +/ # SDs from the mean of the college, not the mean of the overall population. It’s essentially looking at whether the low scoring students at a particular college have a reduced chance of graduating compared to the average or high scoring students at that particular college, when controls for the individual are held constant (similar HS GPA, similar HS curriculum, similar race, …). That table does not consider how low the test score is an absolute sense, only how the test score compares to other students at that college, sort of like what percentile the test score is at that college.</p>
<p>It’s also not a big mystery how Bowdoin is selecting non-submitters with high test scores compared to the overall population. Test scores add relatively little to the prediction of academic success beyond other sections of the application because they are correlated with other sections of the application. You rarely find students who excel in GPA, curriculum, LORs, essays, ECs, and other areas compared to the Bowdoin applicant pool; but totally bomb the SAT and get scores near the US mean of 1000/1600. Instead most of non-submitters have test scores that are somewhat lower than the average submitter at Bowdoin, but still have far stronger test scores than the typical US student.</p>