Nice article. As a parent, I would also be curious to test out the Immersion class performance vs General Ed performance self-selection hypotheses you mentioned. Perhaps one day that data will become available...
I have no doubt that there is a selection bias in the immersion programs. I imagine one can look to the lottery prioritization to glean some of this.
I am concerned with "It would also be interesting to see if the switch to a 50:50 model has any effect on how well the English-fluent students learn Spanish but I’m not aware if this is even measured." It seems like English-fluent students are not even a consideration in the decision making.
The terrible curriculum is a big driver of the bad performance, there’s also minimal understanding of dyslexia (and none of stealth dyslexia.) There’s also an overall lack of any feeling of responsibility or urgency around kids academic abilities. There’s a lot of fluff. So much that adding in a year of substance would be a great improvement. This is also true in math where the curriculum is a minimum of a year behind where it should be. I did not use to support homework in elementary school but now (kids in 5th and 3rd) I do - the progress without hw in SFUSD is poor, though adding fluff hw like coloring obviously wouldn’t help anything either.
Agree there was a missed opportunity in not changing biliteracy pathway to 50% while keeping immersion as is - then there would have been a meaningful difference between them.
I always find your analyses well thought out and interesting to read. Has anyone looked into why disadvantaged Asians score so well relative to disadvantaged kids of other ethnicities? During the pandemic, disadvantaged Asian kids, who often come from poor, non-English speaking households, also lost a year of English access.
Thanks. I try to stay away from directly comparing ethnicities and prefer to compare students of the same ethnicity in different districts because these topics can be discussed without making anyone defensive. For example, why do disadvantaged Asian students in San Francisco do better than disadvantaged Asian students elsewhere in the state but advantaged Asian students in San Francisco do worse than advantaged Asian students elsewhere in the state?
We should stop treating compulsory education as job training. STEM should take a back seat to literacy until students hit the 70% target, or higher. It seems to me that more classroom, less activity, too much homework have all made our educational system dysfunctional.
How were the standards established? It would seem that around 25% of students fall in to each of the four categories. Kind of like a normal distribution. Would the lowest quarter Not Met be equivalent to a D and F; Exceed the standard A and B; and met and nearly met C+ to C-?
Establishing the standards was a complex, multi-step process that is documented in one of the first Smarter Balanced Technical Reports (https://portal.smarterbalanced.org/library/en/2013-14-technical-report.pdf). In practice, the percentage who meet the standard varies by grade and by subject. In other words, it's easier to meet the standard in some grades than in others and in English rather than Math.
There is no relationship to grades. With grade inflation, I would expect most students who score Standard Met to get an 'A' and, at some schools, plenty of students who score Standard Nearly Met will also get an 'A'. In part this is due to grade inflation but it's also due to confusion about what a grade represents. Is it an absolute level of achievement (like the SBAC standards) or is it relative to other students in the class? Should it reward effort or just achievement? Even when grades are test-based, the teacher often "curves" the test if enough students don't score highly enough which means the relative standards of the students in the class matters.
Nice article. As a parent, I would also be curious to test out the Immersion class performance vs General Ed performance self-selection hypotheses you mentioned. Perhaps one day that data will become available...
I have no doubt that there is a selection bias in the immersion programs. I imagine one can look to the lottery prioritization to glean some of this.
I am concerned with "It would also be interesting to see if the switch to a 50:50 model has any effect on how well the English-fluent students learn Spanish but I’m not aware if this is even measured." It seems like English-fluent students are not even a consideration in the decision making.
The terrible curriculum is a big driver of the bad performance, there’s also minimal understanding of dyslexia (and none of stealth dyslexia.) There’s also an overall lack of any feeling of responsibility or urgency around kids academic abilities. There’s a lot of fluff. So much that adding in a year of substance would be a great improvement. This is also true in math where the curriculum is a minimum of a year behind where it should be. I did not use to support homework in elementary school but now (kids in 5th and 3rd) I do - the progress without hw in SFUSD is poor, though adding fluff hw like coloring obviously wouldn’t help anything either.
Agree there was a missed opportunity in not changing biliteracy pathway to 50% while keeping immersion as is - then there would have been a meaningful difference between them.
I always find your analyses well thought out and interesting to read. Has anyone looked into why disadvantaged Asians score so well relative to disadvantaged kids of other ethnicities? During the pandemic, disadvantaged Asian kids, who often come from poor, non-English speaking households, also lost a year of English access.
Thanks. I try to stay away from directly comparing ethnicities and prefer to compare students of the same ethnicity in different districts because these topics can be discussed without making anyone defensive. For example, why do disadvantaged Asian students in San Francisco do better than disadvantaged Asian students elsewhere in the state but advantaged Asian students in San Francisco do worse than advantaged Asian students elsewhere in the state?
We should stop treating compulsory education as job training. STEM should take a back seat to literacy until students hit the 70% target, or higher. It seems to me that more classroom, less activity, too much homework have all made our educational system dysfunctional.
How were the standards established? It would seem that around 25% of students fall in to each of the four categories. Kind of like a normal distribution. Would the lowest quarter Not Met be equivalent to a D and F; Exceed the standard A and B; and met and nearly met C+ to C-?
Establishing the standards was a complex, multi-step process that is documented in one of the first Smarter Balanced Technical Reports (https://portal.smarterbalanced.org/library/en/2013-14-technical-report.pdf). In practice, the percentage who meet the standard varies by grade and by subject. In other words, it's easier to meet the standard in some grades than in others and in English rather than Math.
There is no relationship to grades. With grade inflation, I would expect most students who score Standard Met to get an 'A' and, at some schools, plenty of students who score Standard Nearly Met will also get an 'A'. In part this is due to grade inflation but it's also due to confusion about what a grade represents. Is it an absolute level of achievement (like the SBAC standards) or is it relative to other students in the class? Should it reward effort or just achievement? Even when grades are test-based, the teacher often "curves" the test if enough students don't score highly enough which means the relative standards of the students in the class matters.