There is quite some missing data, which shouldn’t be like this. For example the field U30 in excel (table RAW), which has the opr-responses for the journal “Advanced Materials” is missing. It should probably be “Not specified”.
The following are all variables, where implicit missings should be checked and converted to explicit ones (as for opr_responses), or fixed (as for publishers).
-
I don’t think this distinction is very interesting, since it masks disciplinary differences, which are more fundamental.
-
Only 86 out of 171 do have a policy for manuscript transfer after rejection. Due to policies being similar across journals of certain publishers, there are 48 distinct policies in our dataset.
@@ -1680,13 +1680,13 @@The following graph shows the relationship between to most common bigrams (only bigrams that occur at least three times).
-
The following graph shows the relationship between to most common bigrams (only bigrams that occur at least three times).
-
This doesn’t look interesting. 
This doesn’t look interesting. 
This doesn’t look interesting. 
This doesn’t look interesting. 
##
## Principal inertias (eigenvalues):
##
@@ -1801,8 +1801,8 @@ Cross tabulations
## 19 52 | -296 201 5 | 283 185 20 |
## 20 50 | -216 242 16 | -144 107 32 |
## 21 60 | 610 850 207 | 55 7 8 |
##
## Principal inertias (eigenvalues):
##
@@ -1870,7 +1870,7 @@ Cross tabulations
## 23 -510 679 <NA> | -181 85 <NA> |
## 24 -118 173 <NA> | -90 100 <NA> |
## 25 285 151 <NA> | -226 95 <NA> |
This looks intelligble, but lets try one more time with this crude variable
##
## Principal inertias (eigenvalues):
@@ -1948,8 +1948,8 @@ Cross tabulations
## 27 <NA> | 642 848 <NA> | -14 0 <NA> |
## 28 <NA> | 140 163 <NA> | -185 284 <NA> |
## 29 <NA> | -365 256 <NA> | -50 5 <NA> |
## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
@@ -1960,9 +1960,9 @@ Cross tabulations
## Warning in stats::chisq.test(tab, correct = FALSE): Chi-squared approximation
## may be incorrect
## [1] 0.2201843
-
+
First axis is distinction between SSH and sciences. Second axis is more difficult in terms of what it means. It also has not that much inertia, thus being not very influential
-
+
##
## Principal inertias (eigenvalues):
##
@@ -2079,7 +2079,7 @@ Cross tabulations
## Dim. 1 0.680514 -1.592465
## Dim. 2 1.282670 0.264554
Axes are mainly determined by peer review policy. Maybe remove this variable to bring out mor subtle and potentialy more meaningful answers regarding the key topics (that SSH have double blind and Sciences have single blind seems to be well established)
-
+
##
## Principal inertias (eigenvalues):
##
@@ -2120,9 +2120,9 @@ Cross tabulations
## 6 | LSES | 115 915 158 | 344 406 163 | -384 509 294 |
## 7 | PhyM | 105 478 103 | -8 0 0 | -316 478 181 |
## 8 | SclS | 110 299 94 | -227 284 68 | 52 15 5 |
-
+
There is not much variation left to explain here, somehow.
-
+
##
## Principal inertias (eigenvalues):
##
diff --git a/01-overview.md b/01-overview.md
index 90c6344..9af8be0 100644
--- a/01-overview.md
+++ b/01-overview.md
@@ -1,7 +1,7 @@
---
title: "Landscape Study - General Overview"
author: Thomas Klebel
-date: Last changed 2020-02-12
+date: Last changed 2020-02-25
output:
html_document:
df_print: paged
diff --git a/01-overview_files/figure-html/peer-type-publisher-1.png b/01-overview_files/figure-html/peer-type-publisher-1.png
index 2c1e30a..1d74a48 100644
Binary files a/01-overview_files/figure-html/peer-type-publisher-1.png and b/01-overview_files/figure-html/peer-type-publisher-1.png differ