AMS Letters Figure

Data

The data comes from the QSIDE dataset, accessed January 3rd. The categorical variables we are interested in are lettergroup (which letter was signed), gender, role (e.g., grad student, lecturer, assistant professor, professor, etc…), security (more or less secure) and institution (R1, R2, etc…). We did not use any ethnicity data in this analysis.

“moresecure” refers to identified associates, professors, or retired.

“lesssecure” refers to identified assistant, grad, ntt professor, staff, or undergrad.

“moreri” refers to more research intensive institutions, ie R1.

“lessri” refers to less research intensive institutions, ie non-R1.

The booleans used for comparison were:

all: security == moresecure OR security == lesssecure

non-R1: research == lessri AND (security == moresecure OR security == lesssecure)

untenured-R1: security == lesssecure AND research == moreri

tenure-track-R1: (security == moresecure OR role == assistant) AND research == moreri

tenured-R1: research == moreri AND security == moresecure

Remarks:

Note that tenured-R1 + untenured-R1 + non-R1 == all

Assistant professors are the intersection of untenured-R1 and tenure-track-R1.

Analysis

df$lettergroup <- as.factor(df$lettergroup)
df$gender <- as.factor(df$gender)
df$role <- as.factor(df$role)
df$security <- as.factor(df$security)
df$institution <- as.factor(df$institution)
df$lettergroup <- as.factor(df$lettergroup)
df$research <- as.factor(df$research)
summary(df)
##      name           affiliation         lettergroup        gender   
##  Length:1431        Length:1431        A and B:  6   man      :971  
##  Class :character   Class :character   A Only :615   nonbinary:  1  
##  Mode  :character   Mode  :character   B and C: 74   woman    :452  
##                                        B Only :600   NA's     :  7  
##                                        C Only :136                  
##                                                                     
##                                                                     
##   ethnicity            isurm             highered        
##  Length:1431        Length:1431        Length:1431       
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##         institution    research     country                 role    
##  domesticother:381   lessri:486   Length:1431        professor:630  
##  domesticr1   :676   moreri:881   Class :character   associate:210  
##  domesticr2   :105   NA's  : 64   Mode  :character   assistant:194  
##  international:205                                   grad     :119  
##  NA's         : 64                                   ntt      : 92  
##                                                      (Other)  : 85  
##                                                      NA's     :101  
##        security      field           simplefield       
##  lesssecure:422   Length:1431        Length:1431       
##  moresecure:907   Class :character   Class :character  
##  NA's      :102   Mode  :character   Mode  :character  
##                                                        
##                                                        
##                                                        
##

ALL

# A
table(na.omit(filter(df, (lettergroup=="A Only"|lettergroup=="A and B")&(security=="moresecure"|security=="lesssecure"))$field))
## 
##   comp   math mathed  other   stat 
##      7    507     34     15      9
#A cap B
table(na.omit(filter(df, (lettergroup=="A and B")&(security=="moresecure"|security=="lesssecure"))$field))
## 
## math 
##    6
#B
table(na.omit(filter(df, (lettergroup=="B Only"|lettergroup=="A and B"|lettergroup=="B and C")&(security=="moresecure"|security=="lesssecure"))$field))
## 
##   comp   math mathed  other   stat 
##      7    560      1     53      3
#B cap C
table(na.omit(filter(df, (lettergroup=="B and C")&(security=="moresecure"|security=="lesssecure"))$gender))
## 
##       man nonbinary     woman 
##        65         0         7
# C
table(na.omit(filter(df, (lettergroup=="C Only"|lettergroup=="B and C")&(security=="moresecure"|security=="lesssecure"))$gender))
## 
##       man nonbinary     woman 
##       163         0        44
#B cup C including A cap B
table(na.omit(filter(df, (lettergroup=="B Only"|lettergroup=="A and B"|lettergroup=="C Only"|lettergroup=="B and C")&(security=="moresecure"|security=="lesssecure"))$gender))
## 
##       man nonbinary     woman 
##       627         0       132
#A cup B cup C
table(na.omit(filter(df,(security=="moresecure"|security=="lesssecure"))$gender))
## 
##       man nonbinary     woman 
##       900         1       425

non-R1

# A
table(na.omit(filter(df, (lettergroup=="A Only"|lettergroup=="A and B")&(research=="lessri")&(security=="moresecure"|security=="lesssecure"))$gender))
## 
##       man nonbinary     woman 
##       160         1       172
#A cap B
table(na.omit(filter(df, (lettergroup=="A and B")&(research=="lessri")&(security=="moresecure"|security=="lesssecure"))$gender))
## 
##       man nonbinary     woman 
##         2         0         1
#B
table(na.omit(filter(df, (lettergroup=="B Only"|lettergroup=="A and B"|lettergroup=="B and C")&(research=="lessri")&(security=="moresecure"|security=="lesssecure"))$gender))
## 
##       man nonbinary     woman 
##       105         0        34
#B cap C
table(na.omit(filter(df, (lettergroup=="B and C")&(research=="lessri")&(security=="moresecure"|security=="lesssecure"))$gender))
## 
##       man nonbinary     woman 
##         1         0         0
# C
table(na.omit(filter(df, (lettergroup=="C Only"|lettergroup=="B and C")&(research=="lessri")&(security=="moresecure"|security=="lesssecure"))$gender))
## 
##       man nonbinary     woman 
##         4         0         3
#B cup C including A cap B
table(na.omit(filter(df, (lettergroup=="B Only"|lettergroup=="A and B"|lettergroup=="C Only"|lettergroup=="B and C")&(research=="lessri")&(security=="moresecure"|security=="lesssecure"))$gender))
## 
##       man nonbinary     woman 
##       108         0        37
#A cup B cup C
table(na.omit(filter(df,(security=="moresecure"|security=="lesssecure")&(research=="lessri"))$role))
## 
## assistant associate      grad       ntt professor   retired     staff 
##       106       135         6        29       177        14         5 
## undergrad 
##         3

untenured-R1

# A
table(na.omit(filter(df, ((lettergroup=="A Only"|lettergroup=="A and B")&(security=="lesssecure")&(research=="moreri")))$gender))
## 
##       man nonbinary     woman 
##        79         0        91
#A cap B
table(na.omit(filter(df, ((lettergroup=="A and B")&(security=="lesssecure")&(research=="moreri")))$gender))
## 
##       man nonbinary     woman 
##         2         0         0
#B
table(na.omit(filter(df, ((lettergroup=="B Only"|lettergroup=="A and B"|lettergroup=="B and C")&(security=="lesssecure")&(research=="moreri")))$gender))
## 
##       man nonbinary     woman 
##        83         0        18
#B cap C
table(na.omit(filter(df, ((lettergroup=="B and C")&(security=="lesssecure")&(research=="moreri")))$gender))
## 
##       man nonbinary     woman 
##         1         0         0
# C
table(na.omit(filter(df, ((lettergroup=="C Only"|lettergroup=="B and C")&(security=="lesssecure")&(research=="moreri")))$gender))
## 
##       man nonbinary     woman 
##         2         0         0
#B cup C
table(na.omit(filter(df, ((lettergroup=="B Only"|lettergroup=="A and B"|lettergroup=="B and C"|lettergroup=="C Only")&(security=="lesssecure")&(research=="moreri")))$gender))
## 
##       man nonbinary     woman 
##        84         0        18
#A cup B cup C
table(na.omit(filter(df,((security=="lesssecure")&(research=="moreri")))$gender))
## 
##       man nonbinary     woman 
##       161         0       109

tenure-track-R1

# A
table(na.omit(filter(df, ((lettergroup=="A Only"|lettergroup=="A and B")&(research=="moreri")&(security=="moresecure"|role=="assistant")))$gender))
## 
##       man nonbinary     woman 
##        58         0        58
#A cap B
table(na.omit(filter(df, ((lettergroup=="A and B")&(research=="moreri")&(security=="moresecure"|role=="assistant")))$gender))
## 
##       man nonbinary     woman 
##         1         0         0
#B
table(na.omit(filter(df, ((lettergroup=="B Only"|lettergroup=="A and B"|lettergroup=="B and C")&(research=="moreri")&(security=="moresecure"|role=="assistant")))$gender))
## 
##       man nonbinary     woman 
##       375         0        51
#B cap C
table(na.omit(filter(df, ((lettergroup=="B and C")&(research=="moreri")&(security=="moresecure"|role=="assistant")))$gender))
## 
##       man nonbinary     woman 
##        63         0         7
# C
table(na.omit(filter(df, ((lettergroup=="C Only"|lettergroup=="B and C")&(research=="moreri")&(security=="moresecure"|role=="assistant")))$gender))
## 
##       man nonbinary     woman 
##       157         0        41
#B cup C
table(na.omit(filter(df, ((research=="moreri")&(lettergroup=="B Only"|lettergroup=="A and B"|lettergroup=="B and C"|lettergroup=="C Only")&(security=="moresecure"|role=="assistant")))$gender))
## 
##       man nonbinary     woman 
##       469         0        85
#A cup B cup C
table(na.omit(filter(df,((research=="moreri")&(security=="moresecure"|role=="assistant")))$gender))
## 
##       man nonbinary     woman 
##       526         0       143

tenured-R1

# A
table(na.omit(filter(df, ((lettergroup=="A Only"|lettergroup=="A and B")&(research=="moreri")&(security=="moresecure")))$gender))
## 
##       man nonbinary     woman 
##        39         0        31
#A cap B
table(na.omit(filter(df, ((lettergroup=="A and B")&(research=="moreri")&(security=="moresecure")))$gender))
## 
##       man nonbinary     woman 
##         1         0         0
#B
table(na.omit(filter(df, ((lettergroup=="B Only"|lettergroup=="A and B"|lettergroup=="B and C")&(research=="moreri")&(security=="moresecure")))$gender))
## 
##       man nonbinary     woman 
##       341         0        43
#B cap C
table(na.omit(filter(df, ((lettergroup=="B and C")&(research=="moreri")&(security=="moresecure")))$gender))
## 
##       man nonbinary     woman 
##        63         0         7
# C
table(na.omit(filter(df, ((lettergroup=="C Only"|lettergroup=="B and C")&(research=="moreri")&(security=="moresecure")))$gender))
## 
##       man nonbinary     woman 
##       157         0        41
#B cup C
table(na.omit(filter(df, ((research=="moreri")&(lettergroup=="B Only"|lettergroup=="A and B"|lettergroup=="B and C"|lettergroup=="C Only")&(security=="moresecure")))$gender))
## 
##       man nonbinary     woman 
##       435         0        77
#A cup B cup C
table(na.omit(filter(df,((research=="moreri")&(security=="moresecure")))$gender))
## 
##       man nonbinary     woman 
##       473         0       108