Theresa Lambert, the principal, and all the teachers and staff of Castro School deserve great credit for these results. They have tough jobs and are constantly under criticism from the district administration and school board yet they continue on because they are committed and care.

Why are all the teachers getting raises when test scores are dropping?

The article makes a big deal out of one or two percentage point gains and losses.

One percent of 1000 kids is ten people. That could be as simple as a single extended family moving into or out of district.

@ 25 people like - another community might not be populated by people who recognize an unsophisticated 'auto people likes' application. HA HA It is pretty obvious that someone is using VIP or other masking approaches to quickly splash a fixed number of auto-generated false likes, to many Voice comment cycles. Nice try, obvious ruse.

@ 25 people like, you might try a little less obvious approach, to Fool Us. Use a randomized seed or coefficient, to generate numbers close to, but less than 25. Then, we won't see an immediate '25 likes' or such a delta, and your artificial "likes" will be better masked.

@ 25 likes, perhaps you are fooling most of the people, nearly none of the time

[Portion removed; don't speculate about the identity of other posters.]
Rosa of Castro City may be surprised that I partially agree with her:

"Theresa Lambert, the principal, and all the teachers and staff of Castro School deserve great credit for these results. They have tough jobs."

these are my own personal opinions, and not of the MVWSD Board
these were not pre approved by the Superintendent or the Board President of the MVWSD

sorry Voice - I don't know where that old 'registered user' password of mine went. So I wasn't really speculating about the post from "@ 25 likes". If I use a pseudonym to sometimes post (makes a good header), I try to always remember to put in a signed 'claimer' at the bottom.

Wasn't there a "Publius" guy who wrote The Federalist Papers? And a pseudonym heliocentric proponent in Sidereus Nuncios (The Sidereal Messenger).

Dialogue Concerning the Two Chief World Systems, = that and "Simplicio" are the works from the history of science that I was trying to recollect!

From what I understand, students at the high school level could opt out of taking tests. Several of the top-tier juniors opted out (too busy with AP/honors coursework).

Greg is right, these differences could easily be due to random irrelevant variations (noise). While I understand that the state summarizes the data this way, I would hope to see more useful summaries from the district. Judy Crates (former Castro principal) used to show charts that plotted student scores the previous year against the current scores, so, for instance, you could see how many students who were in the Standard Nearly Met category last year, had moved to Standard Met, Standard Exceeded (or Standard Not Met) this year (yeah, I know the categories had different names back then). This removes the large source of noise Greg mentioned that you are simply testing different kids each year. I would like to see this change coupled with publication of the raw scores. A score at the high end of the “Standard Nearly Met” category is much closer to a score at the low end of the “Standard Met” category than one at the low end of the “Standard Nearly Met” category. Reporting the categories rather than the raw scores a) removes information (signal), b) increases the effect of the noise (decreases the signal to noise ratio by decreasing the signal), and c) incentivizes schools to game the system. The fact is that, scored by category, you get a lot more “bang for your buck” trying to move the kids at the high end of the “Standard Nearly Met” line over the border, than you do worrying about the students falling farther back. When you look at the raw numbers you can see how much progress every student made, not just how many crossed a border. This in turn (because you are looking at larger numbers of students) results in more stable estimates of the progress being made at the school.

Joel, there is such a built-in by the state comparison (following the cohort to the next year) but it only works up to the 8th grade. This is because the testing in high school is only one grade. At Castro, for instance, you can follow the same cohort of students from year-to-year, for instance from 4th into 5th grade, Math, Economically Disadvantaged

Web Link

Or Graham - from 7th into 8th grade Math, Economically Disadvantaged

Web Link

Using % achieving - yeah, so nice to see that many of the public /school parents/ can understand issues like signal to noise in data reporting. The State data also can show the score Average #, which binning by category only degrades (in a S/N way).

Now - how about MEDIAN of the population being tested? Joel, these are really Population Statistics, and the 50% above/50% below of the MEDIAN is what we all know how to think about (thanks to Realtors (c) and real estate reporters, who now report housing prices in the MEDIAN and not the average.) The statistics of MEDIAN and AVERAGE and when they are appropriate and inappropriate to explain a distribution are well covered now in school,

in 6th grade Common Core MATH! [standards explained by good old Sal Kahn & Co. kahnacadamy ]
Web Link

SN is a Trustee of MVWSD, these are his own opinions