Important Reflections On Education Research

Over the past two weeks or so, there have been number of important tweets sent-out by education researchers who I respect.

I thought I’d collect them into this one post and add them to The Best Resources For Understanding How To Interpret Education Research.

Here they are:

My rule of thumb is that any study proclaiming that a light-touch, low-cost intervention dramatically reduces a long-standing social problem based on entrenched inequities is not likely to replicate, eg pic.twitter.com/lnZ9CBhsA3

— Matt Barnum (@matt_barnum) December 7, 2018

Also: Most ed studies are not controlled experiments, but look-backs at real policy interventions in real schools. This should affect how we think about ‘replication’ — the same policy might not ‘replicate’ elsewhere because it’s not really the same or the context is different.

— Matt Barnum (@matt_barnum) December 7, 2018

I used to point out that, within the logic of null hypothesis significance testing, if the result is statistically significant, you are not allowed to complain that the sample is too small. You are, however, allowed to complain that it may be unrepresentative…

— Dylan Wiliam (@dylanwiliam) December 5, 2018

SAMPLE SIZE IN QUALITATIVE RESEARCH
This is a thread on how to address criticism of low sample size in qualitative research. I hope you will read and share. #AcademicTwitter #QualitativeResearch #critqual #phdchat #phdadvice #educolor #PhD #acwri

— Dr. QueenB (@drkakali) December 2, 2018

The ongoing debate about the usefulness of effect sizes. — Evidence-based school leadership and management: A practical guide https://t.co/2XvQxuvA8b

— Stephen Dinham (@sdinhamunimelb) December 2, 2018