From my paternal grandmother's 1920 University of Washington Tyee yearbook.
Note the score. Ouch!
A long pondered but only lately realized blog about economics, politics, evaluation, econometrics, academia, college football and whatever else comes to mind.
It features my friend and former Maryland colleague Bill Evans and highlights the great work that the LEO center is doing in generating compelling causal evidence directly relevant to helping the poor.
So I was writing up my referee report and wanted to encourage the author(s) of the paper I was reviewing to use leading zeros. I sought for support from the Chicago Manual of Style, 14th Edition. It said:
8.19 In scientific contexts decimal fractions of less than 1.00 are set with an initial zero if the quantity expressed is capable of equaling or exceeding 1.00.
8.20 If the quantity never exceeds 1.00, as in probabilities, levels of significance, correlation coefficients, factor loadings, and so forth, no zero is used.
where I have omitted the examples the Manual provides.
I had never heard of this surprising distinction before, having erroneously imagined that there were only two views on leading zeros: "yes" and "no".
I will confess that the Manual's take seems a bit too much like Fizbin to me, but I am surely a better person for having learned about it.
Oh, and I suggested that the authors of the paper I was reviewing use leading zeros all the time.
ROPE = Region of Practical Equivalence
This is a very useful term for when two estimates of some population object differ numerically, and maybe even statistically, but do not differ enough substantively for it to matter in a practical sense. Such estimates lie within the ROPE for the problem at hand.
I am inclined to agree with Sarah Turner and her co-author Jordan Arnold who question the conventional wisdom that going test-optional will help disadvantaged college applicants in this piece in Education Next.