California started CAASPP in 2014. The ELA/math assessment is administered in 11th grade. Science is a separate assessment. CAASPP has replaced CHSPE and STAR.
There are a lot of problems with offering the test in 11th grade. The kids don’t take it seriously because it isn’t meaningful to them. It was initially offered before APs and a lot of kids opted out of taking the test. The state moved the test date to after APs to try to eliminate the opt-outs.
The test has an adaptive algorithm which “rewards” a correct answer with a more rigorous question. High schoolers know this and game the test (answering incorrectly) to try to shorten the duration and rigor.
In my opinion, CAASPP has been a flop since the beginning and the state is scrambling to get high school kids to take the test seriously. The CSUs already use CAASPP for placement. Having the UCs also use CAASPP would create value for high school students, increase participation as well as test scores and make state educational data look better.
Note that my post said “future long term plans.” It is important to distinguish between short term plans and long term plans. Most larger colleges in the United States are making some type of short term admission changes in response to COVID-19. At many colleges, these short term changes include being more test flexible/optional. I would not be surprised if UC being temporarily test optional has more to do with COVID-19 than transitioning to the new system.
That said, I do agree that the authors of the report would likely have preferred a different short term transition plan, even if they had known about COVID-19. The report suggests a longer number of years before widespread implementation of the new assessment system and mentions, “Members of the Task Force differed on the question of whether to recommend that UC cease consideration of standardized test scores sooner — in all likelihood before availability of the replacement suite of assessments.”
The previous quote stated that the test “could potentially show smaller disparities than current measures along the lines of race, ethnicity, and SES” – “smaller” gaps, not zero gaps. Different standardized tests can produce different degrees of gaps between races and SES levels. For example, the Geiser UC studies found that the following correlation with parents income for the 3 SAT II subject tests. Writing appeared to be most correlated with income, while the 3rd test appeared to have little correlation with income. Math fell somewhere in between. An emphasis on writing or vocabulary might yield a higher correlation with income, probably race as well, with CA’s large Hispanic population. An emphasis on student selected field and/or planned field of study might yield a lower correlation with income. The term “assessment” also may mean more than just a traditional long multiple choice SAT-style written test.
Perhaps. But CA has millions of low income ORMS, so an income-based test is of no help. Plus, a test also has to benefit African Americans as well as Latinx.
Yes, ‘assessment’ can be anything, but as a practical matter the assessment has to be simple to evaluate (for 100+k apps to UCLA alone) as well as to submit by the applicant. (If the assessment is too much of a hassle, the poor and particularly the URM’s will find it burdensome and therefore will not apply in the first place.)
@Data10 . . . I wanted to go back and address the quote from the academic senate in your post #72:
The source of the problem is the report considers UC as a gestalt – if I am using the term correctly – rather than by its disparate, unique entities, the campuses themselves. That’s one of the reasons why I thought that making the boards test-optional should have been left to the individual campuses and essentially let UCLA and UCB keep them as requirements if they wanted (needed), but let the others go to full optional-mode if they desired. Trying to blanket this decision won’t ever work. There would be problems with the UC common app., but not so bad as not to be able to be reworked or gotten around.
The new assessment may enable UC to admit classes of students more representative of the diversity of the state.
A new system should allow the measurement of student progress over time and development of a broader range of skills and attributes relevant to an applicant’s success at UC, reflecting the principle that the University “should select those students who demonstrate a strong likelihood that they will persist to graduation.”
The quote is nebulous and too broad. It makes it sound like graduation is the only goal. If that’s the case, then UC has been doing very well as seen by the 2012 entering cohort:
1st Year Ret.…3-Year….4-Year….5-Year….6-Year….7-Year….Diff. 7th-1st Year
……92.4%|……2.8%…64.3%…81.5%…84.2%…85.3%|………….7.1%
For the cohort entering 2015, the latest year with four-year rates, it shows,
1st Year Ret………3-Year………….4-Year
……93.3%……………4.0%…69.5%
There’s a small improvement in retention and a pretty material improvement in the 4-years in just three entering cohort years.
Let’s consider at UCLA and UCB separately.
UCLA:
2012
1st Year Ret.…3-Year….4-Year….5-Year….6-Year….7-Year….Diff. 7-1st Year
………96.2%|……3.4%…76.7%…88.7%…90.4%…91.2%|……….5.0%
2015
1st Year Ret………3-Year………….4-Year
……96.8%……………3.3%…81.6%
There’s a pretty good improvement in 4-year graduations. But the cumulative for 5-year will effectively be the same as in previous years, in the high 80s, so it means that graduation for UCLA in the three cohort years will have been sped up. That’s obviously important, for the students and their families along with their post-graduation goals, but is it relevant to UC by the way they described it, or are they just concerned only about students graduating?
UCB:
2012
1st Year Ret.…3-Year….4-Year….5-Year….6-Year….7-Year….Diff. 7-1st Year
………96.1%|……3.8%…74.2%…88.6%…90.4%…91.1%|……….4.1%
2015
1st Year Ret………3-Year………….4-Year
……96.7%…………….4.5%…78.6%
Obviously, UCB is a bit more STEM oriented than UCLA, which is why it trails the latter in 4-year graduations by a few percentage points. But still the 4-year rates have improved and are very good and are higher than some really good private colleges.
The 4-year graduations especially for these two UCs will have to be carefully monitored, because, though the 5-year will probably be similar before and after this change, the time-to-degree average will definitely extend out. And the longer one stays in college, the lower the grades of those who have extended stays will be. For UCLA, the average gpa of a 4-year grad was 3.42 – I believe it was for the 2010 cohort, but this info is not currently posted in the link I provided when it was before – but the 5-year was ~ 3.15 or thereabouts. Some of the 5-years would have been Engineering majors, but it’s a truism that the longer the stay, the lower the grades across all the campuses.
Additionally, going to optional testing will affect the 1st-year retention and the differential between the 7-year graduation and the 1st year retention. These both should go up materially.
This is why the UCs objectives here are too vague. There should be numerous other considerations: getting students in and out in reasonable time, helping them to have good options vocationally and/or for grad school, and for the student to be able to have great options of study.
In particular, based on the research reviewed, the Task Force believes that UC will be able to articulate a set of skills and attributes that a) can be measured in a uniform and quantitative way, b) are likely to be predictive of success at UC, and c) could potentially show smaller disparities than current measures along the lines of race, ethnicity, and SES [Socio-economic Status].
Welcome to dreamland. Just a generalized hopeful statement without any basis of fact, because removing the tests will only exacerbate the disparities between race, ethnicity and SES.
Per your quote in #77:
You’re talking about students who for the most part don’t know what they’re going to study in college. And since when is college about admitting students who aren’t well-rounded?
I don’t think it is. Both UCLA and UCB will have to present things like “It is recommended that the student take the SAT and/or ACT for the following majors…”
There was a student who spoke before the Regents in favor of doing away with the tests; if anyone has a transcript of what he said and linked or copied it here, I’d be appreciative.
^the problem, firm, is that let’s say Cal EECS or College of Chem requires a test such as Math 2 (or any other standardized thingy)… That will only continue to ensure that such a major does not reflect the “diversity” of the state, which is (clearly, at least to me) means race & ethnicity. There is absolutely no way that UC would allow a top major to have an over abundance of ORMs. Just bad policy.
My thinking on this is the fact that Napolitano purposely did not ask the Regents to make the campuses test-optional which they would have readily approved. Instead, she & the Regents went full bore to drop the test completely (yeah, sure, they’ll develop something else).
While I appreciate your hard work collecting the great stats on post #83, UC has known for many years that economic diversity (and admission tips) will not lead to racial/ethnicity diversity. There are just too many low income ORMs (as a %) in the State who are UC qualified. And the italicized is the rub. To get more UC qualified URMs, California would have to fix K12…
You can certainly design a test that lessens or even removes gaps based on race, sex or SES but then you will also lessen predictive validity.
Test makers have been trying to erase these gaps and yet retain predictive validity for fifty years at least. In all areas of testing from school through military through civilian employment. Any attempt by UC to design such a test will fail.
@bluebayou . . . racial and ethnic representation in specific fields of study obviously would be an underlying goal, but it isn’t stated in the quote @data10 provided. Federal guidelines are more geared toward overall enrollment and generally not specific majors. Obviously if the boards are not brought back, or the option given to the individual campuses to recommend them, then UC will [edit: be] one step closer to being the CSU system.
UC has too much self-importance in admitting someone to a Sociology major when this student would be better going to CSU which has a more nurturing and relaxed atmosphere and studying something like marketing or accounting, something that would at least give him or her a future.
The UC system does do a good job of enrolling lower income students compared to other highly selective colleges. The economic mobility studies linked at NYT (https://www.nytimes.com/interactive/projects/college-mobility/university-of-california-los-angeles ) found that UCLA was the “elite” college with largest portion of students form the bottom 20% income, by a good margin over #2. UCSD was #1 among “highly selective publics” 4 of the top 6 “highly selective publics” with the largest portion of students from bottom 20% income were within the UC system…
However, I think the task force report was more referring to racial diversity. Race is not supposed to be considered in admission, but UC administrators and the CA population as a whole still care about how well the racial demographics of the campus reflect the racial demographics of the state, and the 2 do not resemble one another. For example, CA secondary schools are ~55% Hispanic and ~9% Asian. Hispanic students outnumber Asian students by 6 to 1 to secondary schools, .yet the UC system has far more Asian students than Hispanic students,. UC is far worse than many other highly selective colleges in terms of campus racial diversity reflecting state population .
I think the high graduation rate largely relates to general selectivity, which has been increasing over time, so it makes sense that the UC graduation rate would increase as well. The task force report and various other studies found that standardized testing was at best a weak predictor of chance of graduation.
For example, the task force report found that the combination of HSGPA + SAT only explained less than 8% of variance in 4 year graduation rate. The old Geiser UC studies found similar ~8% variance combined percentages and broke those percentage down by GPA vs SAT, which are summarized below.
Predictors of 4-year Graduation Rate in UC System
SAT + SES – Explains 4% of variance
HSGPA + SES – Explains 7% of variance
HSGPA + SAT + SES – Explains 8% of variance
There is clearly room for improvement in how well testing predicts graduation success at UCs, particularly among URMs. There are also many other factors besides HS academic/test performance ones that contribute to chance of graduating, particularly among URMs. As such, there are many additional steps the UC system can take to improve graduation rate, which have nothing to do with admission testing.
The UC application requires students to enter specific major(s) for each campus to which they apply. In some cases, students are admitted directly to a major or school, and that major/school may have very different selectivity from the campus as a whole. Some majors also have special application requirements that differ from the campus as a whole.
The part in bold is the elephant in the room, even if stated with some exaggeration. What can the state of California’s higher educational system do about it? Not really much off the top because it depends at least a great portion on improving K-12 as @bluebayou alluded. I think the Regents should just take their hands off of the UC and let things play out with the community colleges doing their work, and the CSUs theirs in conjunction with the UCs, because it is ultimately a team effort between the three.
The OOS students are melded into the various ethnicities.
Latina/os are not all that badly underrepresented at UC with a 1/4 of the students being from this ethnicity, a greater percentage than white students. By inference, you want to seemingly punish Asian students who are resourceful and hard-studying by reducing their numbers to bring Latina/os up. This isn’t a perfect society which corrects inequalities immediately, so it’s going to take time because additionally, a lot of Latina/os are recent immigrants to California.
Meanwhile, they should let the CSUs do their job by educating a good portion of URMs, by giving them a more nurturing, less competitive curriculum which is helpful for those from an underfunded school system, in addition to giving them a greater amount of vocational training. I really don’t like UC’s self-important attitude by thinking it can bring someone up in the mobility index by feeding them Sociology, when the CSUs can give them accounting and marketing.
Therefore, I wish you would have addressed my specific quote as follows:
Your predictors are all very nice, but it doesn’t address time to graduation and field of study. (But I’ll look at them including the link a lot more later.) Again, for a lot of URM students at the UCs, they are relegated to some dead-end major that will help them graduate, but enables nothing as far as employment or grad school if they graduate in the lower tier.
I also said that the UCs need more of a well-rounded student. I think that’s the ideal for all colleges.
Again, the current Prez, Janet Napolitano, has been on record for UC to admit more minorities (within the rules of Prop 209) since Jerry Brown recruited her for the job. The Regents are doing her bidding. If she had asked for a test-optional policy, I have no doubt that’s what they would have approved.
@bluebayou. . . I caught that the first time you stated it. She’s a lame-duck president who wants to make a killer move before she leaves office, but she obviously still wields a lot of power unfortunately.
I tried watching some of the meetings, but the video conferencing made it unwatchable for me.
If 55% of secondary school students are Hispanic, I don’t think 1/4 of students being Hispanic is a good representation, especially when the percentages vary significantly by campus and are generally lower at the more selective UCs. My post stated, " UC administrators and the CA population as a whole still care about how well the racial demographics of the campus reflect the racial demographics of the state," and I listed an example of the difference in demographic representation. My personal opinion on the subject differs from your comments, and this is the wrong thread to discuss that opinion. Do not assume that I want to “punish Asian students.”
SAT remains a weak predictor of graduation rate after controlling for general field of study. You can find some general numbers at https://files.eric.ed.gov/fulltext/ED502858.pdf . Unfortunately, this study only lists r^2 for the combined GPA + scores + SES + other factors; but even with all of the above the combination still only explained 9% +/- 3% of variance in 4-year graduation rate. GPA also wasn’t a particularly good predictor for chance of graduation, which is all the more reason that the system could be improved. Studies at other colleges have come to similar conclusions, including test scores having a weaker predictive ability among URM students.
4-Year Graduation Rate Predictors at UCs
Overall – HSGPA + SAT + SATII + SES +… – Explains 9% of Variance
Social Sciences – HSGPA + SAT + SATII + SES +… – Explains 6% of Variance
General – HSGPA + SAT + SATII + SES +… – Explains 8% of Variance
Life Sciences – HSGPA + SAT + SATII + SES +… – Explains 10% of Variance
Math + Physics – HSGPA + SAT + SATII + SES +… – Explains 12% of Variance
I wouldn’t use the word “relegated”, and I’d expect much of the difference relates to self selection. However, there are significant differences in enrollment among different majors. Looking at degree completions by major at Berkeley, in general (non-international) White and Hispanic students are over and underrepresented in the same majors. Both groups are overrepresented in social sciences + humanities, and underrpresented in math heavy fields. However, Hispanic students tend to be more severely underrepresented in those math heavy fields than White students, particularly CS/EECS.
@firmament2x I’m not saying that the UC Regents “had to couch their words because of Prop 209”. I’m saying that there is a concerted effort to get rid of Prop 209 and allow for quotas in public universities (and perhaps elsewhere in CA government services), though obviously both the CA voters and the Supreme Court will have to weigh in at some point.
Getting rid of SATs can be viewed as part of the same effort (and for those who say this is just a transition to a new testing regime, you need to then explain why consideration of testing is forbidden rather than optional in 2023 and 2024).
Implicit (or explicit) quotas are much easier to implement if you don’t have awkward questions asked and litigation following (as per Harvard) about why you are discriminating against certain groups by forcing them to achieve higher test scores for admission.
Correct, but I should have added that I don’t believe that Janet would have gone so far if she did not have the consent of the current Gov and the state (Dem) leaders. (They could have told her, ‘no’, if for no other reason, they still control a big part of the budget.) My point is that the decision much bigger than just Janet and the Regents making a decision, i.e., putting their hands on UC. The whole state leadership is supportive, as I imagine so are many mayors.
Exactly. While my brain is now foggy, back when I attended, I remember a study of math prep and UC. At the time it was focused on women, as they were (and still are) under-represented in quant fields. But the gist of it was that women entering Cal without at least mastery of precalc were essentially eliminated from a huge portion of the classes – and more importantly majors – taught on that campus. (The number that sticks in my mind was 30-40% of all courses offered.)
However that would require ensuring proper teaching, including over two years, with summer recap, etc, of precalculus and calculus courses in all high CA schools, which would require a specific budget for expansion and heavy academic support (smaller classes/TA’s/labs/extra hours) as well as for teachers, who’d need to be hired and properly paid, which in turn would require lifting whatever proposition was voted 20 or 30 years ago that greatly reduced the ability to increase taxes for educational purpose. Since that’d be a no-go, that’s where we’re at.
Note that another path could be: smart kids who show they have the ability and intelligence to succeed but don’t have the math background can be admitted, conditional to succeeding in a special summer bridge program, free of charge and with housing/food included (otherwise lower income kids simply couldn’t attend - they’d already be forfeiting summer earnings, which would have to be waived and made up through a grant for freshman year only). It’d be costly but probably less costly than adding properly paid teachers and courses with proper support to all high schools and it’d probably encounter fewer budget impediments.
I know UCSD has one but it doesn’t specifically include shoring up math or English skills and it’s fairly short.
Here’s our conversation related to how it came to race:
I highlighted the bold in your post #91 because I noted that I thought it was a little exaggerated. The part you left out that you addressed was as follows:
The point of my exercise here is that all quotes should be viewed in light of the things stated which surround it, i.e., to present them in full context, which are an equally important part of my explanation (besides my wanting to see a quadruple-nested quote – it doesn’t look very good). (Hopefully, too, the nested quotes fill fold up until expanded to be read.)
Regarding your predictors, again, give me a chance to when I can to look over the things you presented in your posts and two or three links. It may be a little while to when I can get to them, but I am interested in them, and I’ll subsequently tag you if I want to state something and/or if i were to request your response.
And let me add that I don’t disagree with the abstract of the studies that gpa is the absolute best predictor of a student continuing to graduation. But high test scores do add the topping: a candidate graduating with honors, which gives him/her a better chance at, say, a top-tier professional school and/or better offers of employment. A public university has to weigh these in its mission of educating the cross-section of the state’s students concomitant with trying to produce top-tier graduates who then feed its coffers. I think the University of Texas has perhaps the biggest burden regarding this. The UC, though, has nine campuses which offer undergrad instruction which can ease the burden at whatever proportion of the two enroll at each, with the 10th being primarily a graduate institution.
The second bold in your post #95 is off limits when it comes to the UC. You’re really not supposed to bring up demographics of those which are thought of as the top-tier UCs v. those of the supposed bottom-tier. The University thinks of itself as a whole, a gestalt (again if using the word correctly). This is why Napolitano led a systemwide vote to eliminate the SAT and ACT, when she should have left it to the individual campuses and their chancellors and academic senates, as the latter have to work with the students.
It can be both, self-selecting and “guided.” There is no accident that a lot of URM students major in Sociology at UCLA and the same with athletes. And again, please reference my completed statement from above involving all three tiers of public higher education in California.
Your disputation with the poster that brought you into the thread in favor of the UC dropping the testing requirements is very well-noted. By the way the task force voted 51-0 to keep testing for the interim even if their report didn’t necessarily disagree with those in favor of removing their requirement. But, if you don’t mind, what is your solution to equalling the playing field with respect to overall numbers and in the specific majors at UC?