bib and knitr::opts_chunk$set(fig.retina = 2.5)

README.Rmd

@@ -8,6 +8,10 @@ csl: apa.csl
 link-citations: yes
 ---
 
+```{r, echo = F, cache = F}
+knitr::opts_chunk$set(fig.retina = 2.5)
+```
+
 # The ecologocal fallacy: Thorndike's (1939) example
 
 The ecological fallacy occurs when results from group-level analyses are wrongly extended to individuals. The fallacy is often attributed to William S. Robinson's [-@robinsonEcologicalCorrelations1950] paper, [*Ecological correlations and the behavior of individuals*](https://www.jstor.org/stable/2087176?origin=crossref&seq=1#page_scan_tab_contents), and the name itself first appeared in sociologist Hanan C. Selvin's [-@selvinDurkheimSuicideProblems1958] paper, [*Durkheim's suicide and problems of empirical research*](https://s3.amazonaws.com/academia.edu.documents/33024288/Durkheim-suicide_empirical-research-problems.pdf?response-content-disposition=inline%3B%20filename%3DDurkheims_Suicide_and_Problems_of_Empiri.pdf&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWOWYYGZ2Y53UL3A%2F20191014%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20191014T151247Z&X-Amz-Expires=3600&X-Amz-SignedHeaders=host&X-Amz-Signature=4e9eeb342fa0332cd0f7ed00a4769661cb7c6921e07dd19a763595f037d35dae). However, my fellow psychologists might be happy to learn the idea goes back at least as far as E. L. Thorndike's [-@thorndikeFallacyImputingCorrelations1939] paper, [*On the fallacy of imputing the correlations found for groups to the individuals or smaller groups composing them*](https://www.jstor.org/stable/1416673?seq=1#page_scan_tab_contents). The purpose of this project is to walk out Thorndike's original examples.

README.md

@@ -527,7 +527,7 @@ t13 %>%
   facet_wrap(~table, ncol = 2, labeller = label_both)
 ```
 
-![](README_files/figure-markdown_github/unnamed-chunk-3-1.png)
+<img src="README_files/figure-markdown_github/unnamed-chunk-4-1.png" width="768" />
 
 Thorndike asserted that “within each of the districts the correlation
 between A and B is zero” (p. 124). Let’s check.
@@ -575,7 +575,7 @@ t13 %>%
   theme(panel.grid = element_blank())
 ```
 
-![](README_files/figure-markdown_github/unnamed-chunk-5-1.png)
+<img src="README_files/figure-markdown_github/unnamed-chunk-6-1.png" width="576" />
 
 This corresponds to Thorndike’s Table XIII. If you look closely at his
 original, you’ll see he miscounted a few cells. No worry–the overall
@@ -616,7 +616,7 @@ t13 %>%
   theme(panel.grid = element_blank())
 ```
 
-![](README_files/figure-markdown_github/unnamed-chunk-7-1.png)
+<img src="README_files/figure-markdown_github/unnamed-chunk-8-1.png" width="576" />
 
 That plot corresponds directly to Thorndike’s Table XIV. Here’s the
 correlation.
@@ -715,8 +715,8 @@ Robinson, W. S. (1950). Ecological correlations and the behavior of
 individuals. *American Sociological Review*, *15*(3).
 <https://doi.org/10.2307/2087176>
 
-Selvin, H. C. (1958). Durkheim’s Suicide and Problems of Empirical
-Research. *American Journal of Sociology*, *63*(6), 607–619.
+Selvin, H. C. (1958). Durkheim’s suicide and problems of empirical
+research. *American Journal of Sociology*, *63*(6), 607–619.
 <https://doi.org/10.1086/222356>
 
 Thorndike, E. L. (1939). On the fallacy of imputing the correlations

bib.bib

@@ -60,7 +60,7 @@ @article{robinsonEcologicalCorrelations1950
 }
 
 @article{selvinDurkheimSuicideProblems1958,
-  title = {Durkheim's {{Suicide}} and {{Problems}} of {{Empirical Research}}},
+  title = {Durkheim's Suicide and Problems of Empirical Research},
   author = {Selvin, Hanan C.},
   year = {1958},
   month = may,