I knew that you were going to get attacked because someone would take your statement as hateful. Unfortunately, it takes longer to collect data than it does to write a dumb comment. They took your factual statement of Simpson's paradox being able to drive one global summary (lower overall scores - a bad thing) counter to the local summaries (increased scores by all groups - a good thing) explained by shifting population proportions as you "blaming immigrants" for the bad outcome when the logical extension of your statement is "hold on, we may have a good thing going on". Ignore @mmooss.
Have to admit the immigration comment also triggered my "blaming immigrants" radar. After a careful reading it was an honest attempt for more information or clarification on how the details might be better interpreted. Something people should do more of when statistics are used as evidence. Actually a very good question.
