Thanks David for the helpful list of studies to check out!
I couldn’t help but wonder whether the G4G community might be a good place for designing, sharing, or collaborating on future assessments. Might be interesting to develop an instrument that could be used across different institutions.
Also wonder if you know of any other communities having these conversations about what things to assess and best practices for designing assessments to evaluate the impact of alternative grading practices.
There have definitely been groups trying to put together consistent surveys, in the style you describe. One example of successfully doing this in math is here: https://doi.org/10.1080/10511970.2018.1488317 . I was part of a group trying to analyze data from Spring 2020 alt-graded classes, but that fizzled out, alas.
A high priority, in my opinion, would be to replicate the Lewis results described here. Having the ability to say that "test anxiety is reduced in a consistent, cross-class/discipline/institution/implementation way" would be a real boon for alt. grading. And/or, to be able to see differences based on some of those features. Plus, Lewis's methods don't require a separate non-alt-graded class, so somewhat easier to implement.
Thanks David for the helpful list of studies to check out!
I couldn’t help but wonder whether the G4G community might be a good place for designing, sharing, or collaborating on future assessments. Might be interesting to develop an instrument that could be used across different institutions.
Also wonder if you know of any other communities having these conversations about what things to assess and best practices for designing assessments to evaluate the impact of alternative grading practices.
There have definitely been groups trying to put together consistent surveys, in the style you describe. One example of successfully doing this in math is here: https://doi.org/10.1080/10511970.2018.1488317 . I was part of a group trying to analyze data from Spring 2020 alt-graded classes, but that fizzled out, alas.
A high priority, in my opinion, would be to replicate the Lewis results described here. Having the ability to say that "test anxiety is reduced in a consistent, cross-class/discipline/institution/implementation way" would be a real boon for alt. grading. And/or, to be able to see differences based on some of those features. Plus, Lewis's methods don't require a separate non-alt-graded class, so somewhat easier to implement.