4. And if 5-year survival increases over time, it is tempting to conclude that treatment has improved.

In fact, none of these inferences are correct.
www.allendowney.com/blog/2025/10...

2. Looking at the difference in survival between early and late detection, it is tempting to conclude that more screening would save lives.

3. In a case where a patient is diagnosed late and dies of cancer, it is tempting to say that they would have survived if their cancer had been caught early.

And we've got the graduates to prove we were right.

The curriculum at Olin College is our answer to this question:

1) Engineering early and often,

2) Emphasize design, entrepreneurship, teamwork, and communication,

And my focus was on

3) Use computational tools before or instead of math

The effect of this error on engineering education is like the effect of the iceberg on the Titanic.

My new auxiliary emergency backup team is taking it down to the wire...

Yes, I had not made that distinction, and you are very right.

I'm not sure I would have bothered debunking it if I had realized.

Oh, no -- it gets worse! It looks like they also including missing data in the analysis, treating it as zero. That explains the black line in the figure.

I gave a talk about that chapter here: www.youtube.com/watch?v=44D1...

Thanks! It was a fun interview to record.

Now if only the playback had speed control :(