eminence | R Documentation |
Marco Del Giudice criticized an earlier study by Simonton for using partial regression weights to estimate the importance of various predictors of rated eminence. This is a nice example of the (mis)interpretation of beta weights of highly correlated predictors.
data("eminence")
A data frame with 69 observations on the following 9 variables.
name
a character vector
reputation
Log of rated reputation
birth.year
Year of birth
first.year
Year of first cited publicatin
last.year
Year of last cited publication
works
Log of number of publications
citations
Log of number of citations
composite
A composite index of publications
h
The 'h' index of citations
Simonton (1997, 2014) discusses various estimates of eminence among 69 psychologists born between 1842 and 1912 and reports that the regression weights are small and interprets this as meaning number of publications and citations are not very important. Del Giudice (2020) points out that citations and the number of publications are highly collinear and thus while their independent contributions are small, their joint effect is quite large (R= .69 ). These data are given here as an example of multiple correlation and partial correlation
Del Giudice (2020) links to a web page with the data.
Marco Del Giudice (2020). How Well Do Bibliometric Indicators Correlate With Scientific Eminence? A Comment on Simonton (2016). Perspective in Psychological Science, 15, 202-203.
Simonton, D. K. (1992). Leaders of American psychology, 1879-1967: Career development, creative output, and professional achievement. Journal of Personality and Social Psychology, 62, 5-17.
Simonton, D. K. (2016). Giving credit where credit is due: Why it's so hard to do in psychological science. Perspectives on Psychological Science, 11, 888-892.
data(eminence)
psych::lowerCor(eminence)
cs <- psych::cs
psych::partial.r(eminence, x= cs(reputation, works, citations),y=cs(birth.year))
psych::setCor(reputation ~ works + h + first.year,data=eminence)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.