Political Decisions

Lowell, Grading for Equity, and Ethnic Studies

Note: since this substack is about education, it’s fitting that it takes a long Summer holiday, starting today.

This post is about three separate topics with high political salience in San Francisco. By coincidence, they have all been subjects of San Francisco Chronicle articles within the last month.

Lowell

The Chronicle published two articles recently about Lowell. The first was about the lower number of admits from non-district (i.e. private and charter) schools and the second was about the success of the first class of “lottery kids” who just graduated. Both articles are filled with juicy data about application numbers, grades, transfer rates, average SAT scores, AP exam participation rates etc. I was a little jealous of the data the Chronicle was able to get. None of the data is public and the district has never published this level of data before. I suppose it’s possible that the Chronicle simply asked the district for the data and the district decided to forego decades of muscle memory and simply hand it over. I think it more likely that someone in the district wanted to get a story out and decided that releasing the data to the Chronicle ensured that it reached the widest audience.

Since I don’t have access to the data, I can’t reproduce the analyses. I do, however, disagree with some of the points made in the first article. Here’s the crux of the story:

Fewer private school students are sitting in the coveted seats at San Francisco’s Lowell High School than in years past, with district students taking 8 out of every 10 spots in the freshman class, according to district data. It’s a significant shift from just four years ago, when nearly a third of the 2020 ninth-graders came from non-district middle schools.

My first reaction when I read the article was that the dalliance with the lottery has tarnished the lustre of Lowell in the eyes of private school applicants. Applicants who might previously have preferred Lowell might now just see it as a backup choice. It’s not just the lottery that might have tarnished the lustre. The instability (five principals in seven years) and widely publicized budget cuts can’t have helped.

After re-reading the article, I’m not as confident in that assessment. First, let’s dispose of some alternative explanations. Here’s one, from a district spokesperson:

“In recent years, we’ve expanded outreach to SFUSD middle school students — especially in historically underserved communities — to ensure there is broader access in applying to Lowell.”

This made me smile because the district has been making this exact same claim since at least 2018. Here’s another purported explanation:

For the first time, all applicants this year were required to take the same test for admission to Lowell. In years past, district students took the state standardized test while those from outside the district took a separate test.

This is true only if you have a very short memory. It’s not the first time all applicants have had to take the same test. They all had to take the same Lowell Admission Test (called the Terra Nova test) every year from at least 2001 until 2017. Starting in 2018, SFUSD students were no longer required to take the Lowell Admission Test. Their 7th grade SBAC results would be used instead. (The theory was that the existence of the Admission Test was a hurdle that discouraged public school students from applying. By removing that hurdle, the Board hoped to increase the number of applicants from historically underserved communities.) In a bit of sleight of hand, they set different standards for public and private applicants. A public school student who scored in the 82nd percentile on the SBAC received the same number of points as a private school student who scored in the 92nd percentile on the Terra Nova test. The points system thus strongly favored public school applicants.

It stayed this way until 2021 when the lottery was introduced. When points-based admissions were re-introduced in 2023, the district couldn’t revert to its old points table because the maximum obtainable 7th grade SBAC score had increased from 2778 to 2820. The new admissions points table was probably intended to be fair but ended up slightly favoring private school students. At that point, the Terra Nova test was discontinued by its publisher, forcing the district to find another solution. It switched to the Star Test, which public school students already took, but which could also be offered to private school students. Having a common test is thus just a return to where the district was ten years ago. Unless the percentage of non-district students was as low in all those years as it is today, the common admission test can’t be the explanation.

Here are two more snippets from the article:

For the upcoming fall, the district admitted 58% of the 708 private and other non-district eighth-grade students who applied to Lowell, compared to 74% of the 584 applicants in 2019.

In 2019, 46% of those admitted from non-district schools chose to enroll. Last year, 37% of those students opted to attend Lowell.

Remember I said in the last post that I always try to double-check data. Let’s do that here for the 2019 data¹. If 74% of the 584 applicants were admitted and 46% of those admitted chose to enroll, the number of non-district enrollees was 584 * 74% * 46% = 199. Elsewhere in the article, we learn that non-district students comprised 31% of the enrolled class in 2019. That implies the total size of the 9th grade class was 199 / 0.31 =642. But Lowell’s 9th grade class in 2019-20 had 716 students in it. 716 is a long way from 642. Something doesn’t add up.

If we take the data at face value, it’s hard to tell a consistent story. The percentage of Lowell enrollees from non-district schools has been 22-25% in the two post-lottery years, exactly the same range as during the two lottery years, and significantly below the 31-32% it was in the two last pre-lottery years. The yield (the percentage of those admitted who enroll) among non-district applicants fell from 46% in 2019 to 37% in 2024. These both suggest a decline in Lowell’s prestige. But the number of non-district applicants was up 21% since 2019 (708 vs 584), which is hard to explain if Lowell’s prestige has declined.

Given the incompleteness of the data, I hope that SFUSD will release publicly the data it gave to the Chronicle so that it might be possible to form a firmer conclusion. I’m not going to hold my breath, however.

Grading for Equity

“Students should be able to retake tests and redo assignments. There should be no penalties for late work and no grades for homework. No points for good behavior, classroom participation or perfect attendance, either…Students should be graded only on their demonstrated learning of class material.”

The quote above comes from a Wall Street Journal article about a new approach to grading. SFUSD was considering adopting this new grading approach. Here’s how the district described the benefits they hoped to achieve:

With more transparent, consistent, and fair grading practices across all classrooms we believe students will be more successful in their courses and utilize greater opportunities to master content objectives each semester. As a result, students will see achievement as attainable with more time and support to demonstrate their learning and stay engaged in their courses. We look forward to improvements in GPA and attendance.

In the associated presentation, the district compressed the benefits of the new grading approach down to “improve GPA”.

One could imagine an interesting philosophical discussion about the proper balance between summative assessment (measuring how well a student has learned the material) and formative assessment (measuring how well a student is doing during the learning period in terms of homework, participation, behavior etc.). It’s quite possible that the way schools do it today is sub-optimal. It’s certainly true that grading practices are not consistent, either between two schools or between two teachers in the same school. Efforts to make them more consistent would be welcome.

That said, it’s hard to see how a change in grading practice could lead to improved attendance, particularly if one of the tenets of the new grading practice is not to give points for attendance. It’s also not obvious why a change in grading practice should lead to higher GPAs. Is it really the case that most students score lower on formative assessments than summative assessments? My children regarded formative assessments as easy points that they could accumulate to build a buffer in case they had a bad day on the final exam.

The reason the new grading approach leads to higher GPAs is that another common feature of it is, according to the Journal article, “a policy…to set minimum grades for assignments at 50 rather than zero.”

Why stop at 50? If the minimum is set to 90, everyone will get an ‘A’.

The name of this new grading approach is Grading for Equity. I refrained from mentioning it until now so that you could judge the idea on its merits rather than have a visceral reaction to its name. After Commissioner Ray highlighted some of the features described here, and the mayors of both San Francisco and San Jose chimed in, the district quickly announced that it was abandoning its plan to adopt Grading for Equity.

A few things were surprising to me:

• Although I’ve described it as a new grading approach and everyone has treated it as such, it’s not new for San Francisco. Here’s Board Policy 5121, adopted in 2017:

The Superintendent or designee shall establish a uniform grading system that shall be applied to all students in that course and grade level…A teacher shall base a student's grades solely on the quality of the student's academic work and his/her mastery of course content based on approved district standards. Students shall have the opportunity to demonstrate this mastery through a variety of methods, including, but not limited to, tests, projects, portfolios, and/or class discussion as appropriate. Other elements that are not a direct measure of knowledge and understanding of course content, such as attendance, effort, student conduct, and work habits, shall not be factored into the academic grade.

• the district did not explicitly seek the board’s approval for the new grading policy. The only reason it reached the board is that the district listed implementing the new policy as one of the steps they were taking to meet their college and career readiness goals.

• SFUSD did not anticipate that something called Grading for Equity would be controversial. When I met with one of the Commissioners² in early March to discuss various matters, I was asked at the end of the meeting whether I had any advice for the board, and responded with something like “fix the budget and stay away from Grading for Equity”. I don’t have finely tuned political antennae so, if it was obvious to someone like me that this was radioactive, it should have been obvious to them.

• In particular, it doesn’t take special insight to realize that most ordinary parents are going to react against the idea of minimum scores of 50. The district should have been prepared to either disavow that aspect of Grading for Equity or defend it.

Ethnic Studies

No sooner was Grading for Equity abandoned than attention was focused on SFUSD’s Ethnic Studies course. I’m not going to wade in to the discussion about the content of SFUSD’s course because I haven’t read it. Instead, I want to talk about the data.

Ethnic studies in SFUSD was originally developed as an optional course targeted at students who might be at risk of dropping out. It was made a graduation requirement in 2021 (when schools were still closed, by the board that would go on to be recalled). Part of the justification at the time was that

Research has shown that Ethnic Studies curriculum in high schools has demonstrated an increase in GPA across disciplines, high school graduation rates, college-going rates, and a sense of belonging.

One of the key research papers was based on SFUSD’s own experience. Here’s the original 2016 paper and a 2021 update. Now, you might assume that a paper with the title “Ethnic studies increases longer-run academic engagement and attainment” compares students who took ethnic studies with similar students who did not. That’s not really what the paper does at all. The paper is evaluating the policy of assigning students with 8th grade GPAs below 2.0 to an ethnic studies course (and giving them an option to switch out) by comparing them with students with GPAs of 2.0 or higher who have the option to enroll in ethnic studies. When a statewide Ethnic Studies mandate was being considered, a bunch of professors wrote a detailed report on the misinterpretation of the paper’s data.

One of the paper’s authors is quoted in an interesting EdSource article about ethnic studies. Here are a few excerpts:

Dee³ and the co-researchers caution, however, not to generalize the results of a study of 1,405 ninth-graders, only 13% of whom — about 180 students — took the ethnic studies course…

Whether the positive impact would apply to all students, especially those already thriving academically, is “an important, open question,” Dee said.

The study itself draws no conclusion about whether the particular course content contributed to the positive results.

What’s more essential, Dee said, is “culturally relevant” instruction that motivates and engages students. “So it’s not clear to me those mediating mechanisms require really doctrinaire, inflammatory content,” he said.

“There’s something I found impoverishing about the public debate over ethnic studies and, more recently, critical race theory,” he said. “And it’s because it has a cultural war frame” instead of focusing on “what’s going on with teaching and learning and student motivation and engagement.”

In other words, the data may support offering ethnic studies as an option to a segment of students but it does not support mandating that every student take the course. That was a political decision. Time will tell if opposition to ethnic studies as a mandate dies down over time or stays as a festering issue in the same way that abolition of middle school algebra did.

1. I didn’t cherry pick 2019. It happens to be the only year for which the article has sufficient data to enable me to do the calculation.

2. not Commissioner Ray

3. Thomas Dee was also a co-author of the paper that showed that abolishing middle school algebra in SFUSD did not lead to more people taking advanced math classes.