#testscores — Public Fediverse posts

Engadget: A study shows that cellphone bans didn’t improve US students’ test scores. “The study sampled phone location data from more than 40,000 schools from 2019 to 2026 and showed that there was a temporary rise in disciplinary incidents and a short-term drop in student well-being, which was attributed to short-term disruption. However, the study revealed that student well-being improved […]

The AUSD Board of Education took a deep dive into math test scores at its meeting on January 27. "Overall, the percentage of AUSD students meeting or exceeding standards has increased and recovered from a drop caused by the COVID-19 pandemic," reports Ken Der. https://alamedapost.com/news/ausd-board-plots-future-math-instruction/

#alameda #ausd #AusdBoard #TestScores

@TomJewell

I argue that there is some understanding in the #psychometrics and professional communities that use #TestScores that evidence of quality is tentative as any statistical result.

And in my personal view, the different methodologies #COSMIN has developed (e.g., https://link.springer.com/article/10.1007/s11136-018-1798-3) or is currently pushing forward (https://doi.org/10.1186/s13643-022-01994-5) are tools to aggregate available evidence exactly for this reason: each individual study and estimate offers an incomplete picture.

@TomJewell

The introduction to Chapter 2 (reliability/precision) acknowledges that the sampling process can lead to significant differences in the determined coefficients,
that they therefore [2014, p. 37] "should be estimated separately for all relevant subgroups"
and that this is also important to assess potential consequences for #TestFairness.

And in Chapter 5 focuses especially in the discussion of #Norming of #TestScores on such aspects, including relating to our previous discussion...