Appendix B

1 RSiena without Twitter dummy for Sociology

#dependent
snet <- sienaDependent(soc_net_array)

### Step 1: define data
#gender
gender <- as.numeric(socdef_df$gender=="female")
gender <- coCovar(gender)

#Kardashian Index
ki <- as.numeric(socdef_df$ki)
ki <- coCovar(ki)

#Ethnicity
dutch <- as.numeric(socdef_df$dutch)
dutch <- coCovar(dutch)

#Twitter dummy as control variable
twitter_dum <- (socdef_df$twitter_dum)
twitter_dum <- coCovar(twitter_dum)

#Twitter followercount
#followers <- as.numeric(soc_twitterinfo$twfollowercounts)
#followers <- coCovar(followers)



#year first pub
# soc_staff_cit %>% group_by(gs_id) %>%
#   mutate(pub_first = min(year)) %>% 
#   select(c("gs_id", "pub_first")) %>%
#   distinct(gs_id, pub_first, .keep_all = TRUE) -> firstpub_df
# 
# socdef_df <- left_join(socdef_df, firstpub_df)
# 
# #if no publication yet, set pub_first op 2023
# socdef_df %>% mutate(pub_first = replace_na(pub_first, 2023)) -> socdef_df

pub_first <-  coCovar(socdef_df$pub_first)

mydata <- sienaDataCreate(snet, gender, ki, dutch, pub_first, twitter_dum)

### Step 2: create effects structure
myeffs <- getEffects(mydata)
effectsDocumentation(myeffs)
### Step 3: get initial description
print01Report(mydata, modelname = "/Users/anuschka/Documents/labjournal/results/soc_report")

myeffs1b <- getEffects(mydata)
myeffs1b <- includeEffects(myeffs1b, degPlus)
myeffs1b <- includeEffects(myeffs1b, transTriads)
myeffs1b <- includeEffects(myeffs1b, absDiffX, interaction1 = "ki")
myeffs1b <- includeEffects(myeffs1b, sameX, interaction1 = "dutch")
myeffs1b <- includeEffects(myeffs1b, absDiffX, interaction1 = "pub_first")
#myeffs1 <- includeEffects(myeffs1, sameX, interaction1 = "twitter_dum")
myeffs1b <- includeEffects(myeffs1b, sameX, interaction1 = "gender")
myAlgorithm <- sienaAlgorithmCreate(projname = "soc_report")
(ans1b <- siena07(myAlgorithm, data = mydata, effects = myeffs1b))

#Save the last model since it has the lowest maximum convergence ratio. 
save(ans1b, file="/Users/anuschka/Documents/labjournal/results/soc_model_app")

#> Estimates, standard errors and convergence t-ratios
#> 
#>                                      Estimate   Standard   Convergence 
#>                                                   Error      t-ratio   
#> 
#> Rate parameters: 
#>   0.1      Rate parameter period 1    1.5888  ( 0.4323   )             
#>   0.2      Rate parameter period 2    2.6081  ( 0.7312   )             
#> 
#> Other parameters: 
#>   1.  eval degree (density)          -2.9547  ( 0.6043   )   -0.0626   
#>   2.  eval transitive triads          0.5743  ( 0.2826   )   -0.0873   
#>   3.  eval degree act+pop             0.0943  ( 0.0450   )   -0.0644   
#>   4.  eval same gender               -0.0455  ( 0.2590   )   -0.0678   
#>   5.  eval ki abs. difference        -0.0126  ( 0.0609   )   -0.0188   
#>   6.  eval same dutch                 0.2307  ( 0.3465   )   -0.0667   
#>   7.  eval pub_first abs. difference  0.0097  ( 0.0191   )   -0.0328   
#> 
#> Overall maximum convergence ratio:    0.1144 
#> 
#> 
#> Total of 2207 iteration steps.
#> 
#> Covariance matrix of estimates (correlations below diagonal)
#> 
#>        0.365        0.046       -0.019       -0.042       -0.011       -0.154       -0.003
#>        0.271        0.080       -0.008       -0.011        0.001       -0.026        0.001
#>       -0.704       -0.633        0.002        0.001        0.000        0.006        0.000
#>       -0.267       -0.148        0.063        0.067        0.000        0.009        0.001
#>       -0.295        0.068        0.038       -0.026        0.004        0.005        0.000
#>       -0.735       -0.265        0.409        0.101        0.239        0.120        0.000
#>       -0.255        0.202       -0.247        0.130        0.056        0.051        0.000
#> 
#> Derivative matrix of expected statistics X by parameters:
#> 
#>       80.644       58.536     2482.916       77.558      176.059      110.729     1689.188
#>       58.133       68.506     2079.404       61.033      107.997       83.888     1176.231
#>      900.013      764.982    30924.019      880.799     1903.424     1184.612    19370.669
#>       35.770       27.681     1106.952       61.129       81.675       48.645      706.255
#>       93.517       56.237     2671.722       92.330      752.652      105.525     1983.877
#>       60.191       44.070     1758.689       57.588      106.130      103.737     1207.985
#>      856.209      616.164    27167.909      772.180     1897.987     1135.407    22959.954
#> 
#> Covariance matrix of X (correlations below diagonal):
#> 
#>      113.090       94.049     3692.727      113.963      223.903      156.835     2361.272
#>        0.846      109.177     3440.233       98.006      164.786      134.551     1946.038
#>        0.959        0.909   131067.426     3751.435     6825.683     4981.963    78260.008
#>        0.846        0.741        0.818      160.284      228.842      159.444     2313.522
#>        0.549        0.411        0.491        0.471     1472.499      269.018     4839.142
#>        0.917        0.801        0.856        0.783        0.436      258.581     3195.689
#>        0.926        0.776        0.901        0.762        0.526        0.829    57535.477

myeffs2b <- getEffects(mydata)
myeffs2b <- includeEffects(myeffs2b, degPlus) 
myeffs2b <- includeEffects(myeffs2b, transTriads)
myeffs2b <- includeEffects(myeffs2b, altX, interaction1 = "ki")
myeffs2b <- includeEffects(myeffs2b, sameX, interaction1 = "dutch")
myeffs2b <- includeEffects(myeffs2b, absDiffX, interaction1 = "pub_first")
#myeffs2 <- includeEffects(myeffs2, sameX, interaction1 = "twitter_dum")
myeffs2b <- includeEffects(myeffs2b, sameX, interaction1 = "gender")
(ans2b <- siena07(myAlgorithm, data = mydata, effects = myeffs2b, prevAns = ans1b))

#Save the last model since it has the lowest maximum convergence ratio. 
save(ans2b, file="/Users/anuschka/Documents/labjournal/results/soc_model_app2")

#> Estimates, standard errors and convergence t-ratios
#> 
#>                                      Estimate   Standard   Convergence 
#>                                                   Error      t-ratio   
#> 
#> Rate parameters: 
#>   0.1      Rate parameter period 1    1.5837  ( 0.4417   )             
#>   0.2      Rate parameter period 2    2.5344  ( 0.7068   )             
#> 
#> Other parameters: 
#>   1.  eval degree (density)          -2.9853  ( 0.5424   )    0.0103   
#>   2.  eval transitive triads          0.5780  ( 0.2731   )   -0.0094   
#>   3.  eval degree act+pop             0.0984  ( 0.0428   )    0.0099   
#>   4.  eval same gender               -0.0624  ( 0.2592   )   -0.0404   
#>   5.  eval ki alter                   0.0324  ( 0.0952   )    0.0436   
#>   6.  eval same dutch                 0.2430  ( 0.3352   )   -0.0439   
#>   7.  eval pub_first abs. difference  0.0084  ( 0.0194   )   -0.0094   
#> 
#> Overall maximum convergence ratio:    0.1760 
#> 
#> 
#> Total of 2350 iteration steps.
#> 
#> Covariance matrix of estimates (correlations below diagonal)
#> 
#>        0.294        0.044       -0.016       -0.040       -0.014       -0.119       -0.003
#>        0.297        0.075       -0.007       -0.007       -0.001       -0.022        0.001
#>       -0.707       -0.626        0.002        0.000        0.001        0.005        0.000
#>       -0.285       -0.093        0.002        0.067       -0.001        0.002        0.001
#>       -0.262       -0.021        0.238       -0.056        0.009        0.009        0.000
#>       -0.655       -0.236        0.340        0.024        0.270        0.112        0.000
#>       -0.312        0.141       -0.165        0.176        0.101        0.042        0.000
#> 
#> Derivative matrix of expected statistics X by parameters:
#> 
#>       75.104       53.813     2301.531       76.452      -79.940      102.340     1536.228
#>       53.220       67.338     1956.522       57.615      -73.232       75.987     1059.389
#>      837.143      727.786    29263.100      859.120     -998.845     1094.097    17614.134
#>       37.305       28.401     1152.731       65.349      -37.180       51.167      692.281
#>      -34.866      -38.058    -1393.892      -33.513      255.963      -59.177     -835.868
#>       54.011       38.983     1566.972       55.169      -70.682       94.276     1083.350
#>      757.645      533.564    23953.290      711.560     -844.230     1012.111    20623.305
#> 
#> Covariance matrix of X (correlations below diagonal):
#> 
#>      110.243       92.954     3613.412      113.177     -126.006      150.172     2285.830
#>        0.830      113.726     3464.121       97.746     -128.248      128.700     1900.348
#>        0.953        0.899   130468.490     3720.857    -4513.807     4730.145    76346.169
#>        0.842        0.716        0.804      164.060     -130.357      155.735     2244.100
#>       -0.525       -0.526       -0.547       -0.445      522.482     -192.812    -2635.512
#>        0.913        0.771        0.836        0.776       -0.539      245.212     3079.986
#>        0.917        0.750        0.890        0.738       -0.486        0.828    56379.585

2 RSiena without Twitter dummy for Data Science

rm(list=ls())

#dependent
dnet <- sienaDependent(dnet_array)

### Step 1: define data
#gender
gender <- as.numeric(datadef_df$gender=="female")
gender <- coCovar(gender)

#Kardashian Index
ki <- as.numeric(datadef_df$ki)
ki <- coCovar(ki)

#Ethnicity
dutch <- as.numeric(datadef_df$dutch)
dutch <- coCovar(dutch)

#Twitter dummy as control variable
twitter_dum <- (datadef_df$twitter_dum)
twitter_dum <- coCovar(twitter_dum)

# #year first pub
# data_staff_cit %>% group_by(gs_id) %>%
#   mutate(pub_first = min(year)) %>% 
#   select(c("gs_id", "pub_first")) %>%
#   distinct(gs_id, pub_first, .keep_all = TRUE) -> firstpub_df1
# 
# datadef_df <- left_join(datadef_df, firstpub_df1)
# 
# #if no publication yet, set pub_first op 2023
# datadef_df %>% mutate(pub_first = replace_na(pub_first, 2023)) -> datadef_df

pub_first <-  coCovar(datadef_df$pub_first)

mydata <- sienaDataCreate(dnet, gender, ki, dutch, pub_first, twitter_dum)

### Step 2: create effects structure
myeff <- getEffects(mydata)
#effectsDocumentation(myeff)
### Step 3: get initial description
#print01Report(mydata, modelname = "/Users/anuschka/Documents/labjournal/results/data_report")

#Estimate the first model
myeffd1c <- getEffects(mydata)
myeffd1c<- includeEffects(myeffd1c, degPlus) 
myeffd1c <- includeEffects(myeffd1c, transTriads)
myeffd1c <- includeEffects(myeffd1c, absDiffX, interaction1 = "ki")
myeffd1c <- includeEffects(myeffd1c, sameX, interaction1 = "dutch")
myeffd1c <- includeEffects(myeffd1c, absDiffX, interaction1 = "pub_first")
#myeffd1 <- includeEffects(myeffd1, sameX, interaction1 = "twitter_dum")
myeffd1c <- includeEffects(myeffd1c, sameX, interaction1 = "gender")
(ansd1c <- siena07(myAlgorithm, data = mydata, effects = myeffd1c))

#Save the last model since it has the lowest maximum convergence ratio. 
save(ansd1c, file="/Users/anuschka/Documents/labjournal/results/data_model_app")

#> Estimates, standard errors and convergence t-ratios
#> 
#>                                      Estimate   Standard   Convergence 
#>                                                   Error      t-ratio   
#> 
#> Rate parameters: 
#>   0.1      Rate parameter period 1    1.6227  ( 0.4582   )             
#>   0.2      Rate parameter period 2    3.1193  ( 0.6966   )             
#> 
#> Other parameters: 
#>   1.  eval degree (density)          -2.0167  ( 0.3514   )   -0.0124   
#>   2.  eval transitive triads          1.2364  ( 0.2181   )   -0.0223   
#>   3.  eval degree act+pop             0.0309  ( 0.0323   )   -0.0121   
#>   4.  eval same gender               -0.0796  ( 0.2182   )   -0.0355   
#>   5.  eval ki abs. difference        -0.0987  ( 0.0836   )    0.0271   
#>   6.  eval same dutch                -0.0597  ( 0.2011   )    0.0145   
#>   7.  eval pub_first abs. difference -0.0155  ( 0.0130   )   -0.0156   
#> 
#> Overall maximum convergence ratio:    0.0790 
#> 
#> 
#> Total of 2380 iteration steps.
#> 
#> Covariance matrix of estimates (correlations below diagonal)
#> 
#>        0.124        0.001       -0.008       -0.025       -0.005       -0.023       -0.001
#>        0.013        0.048       -0.003       -0.001        0.001        0.001        0.000
#>       -0.672       -0.394        0.001        0.000        0.000        0.000        0.000
#>       -0.323       -0.012       -0.060        0.048        0.001       -0.002        0.000
#>       -0.169        0.078       -0.020        0.068        0.007        0.000        0.000
#>       -0.329        0.025        0.032       -0.043        0.001        0.040        0.000
#>       -0.233        0.021       -0.127        0.107        0.007       -0.101        0.000
#> 
#> Derivative matrix of expected statistics X by parameters:
#> 
#>       82.831       42.483     2320.734       92.046       89.998       97.611     1568.235
#>       70.868      126.137     2985.350       85.642       50.699       81.429     1479.550
#>      906.121      823.917    33352.260     1033.896      845.565     1053.065    18339.198
#>       48.020       33.105     1423.120       97.312       37.519       60.817      885.529
#>       42.837       -0.599      953.294       35.219      368.295       48.001      698.086
#>       49.375       23.948     1363.995       57.878       52.589      103.465      973.508
#>      797.240      413.993    23138.772      820.362      799.076     1000.179    27239.641
#> 
#> Covariance matrix of X (correlations below diagonal):
#> 
#>      137.953      189.347     5433.666      162.648      114.963      161.336     2890.142
#>        0.725      494.635    10333.200      250.020       71.071      218.689     4199.539
#>        0.899        0.903   264728.573     6635.106     3318.713     6296.559   118281.688
#>        0.824        0.669        0.767      282.383       95.380      201.712     3372.347
#>        0.342        0.112        0.225        0.198      821.167      127.008     1996.467
#>        0.834        0.597        0.743        0.729        0.269      270.970     3445.908
#>        0.844        0.648        0.789        0.689        0.239        0.718    84926.496

myeffd2d <- getEffects(mydata)
myeffd2d <- includeEffects(myeffd2d, degPlus) 
myeffd2d <- includeEffects(myeffd2d, transTriads)
myeffd2d <- includeEffects(myeffd2d, altX, interaction1 = "ki")
myeffd2d <- includeEffects(myeffd2d, sameX, interaction1 = "dutch")
myeffd2d <- includeEffects(myeffd2d, absDiffX, interaction1 = "pub_first")
#myeffd2d <- includeEffects(myeffd2d, sameX, interaction1 = "twitter_dum")
myeffd2d <- includeEffects(myeffd2d, sameX, interaction1 = "gender")
(ansd2d <- siena07(myAlgorithm, data = mydata, effects = myeffd2d, prevAns = ansd1c))

#Save the last model since it has the lowest maximum convergence ratio. 
save(ansd2d, file="/Users/anuschka/Documents/labjournal/results/data_model_app2")

#> Estimates, standard errors and convergence t-ratios
#> 
#>                                      Estimate   Standard   Convergence 
#>                                                   Error      t-ratio   
#> 
#> Rate parameters: 
#>   0.1      Rate parameter period 1    1.6141  ( 0.4426   )             
#>   0.2      Rate parameter period 2    3.0882  ( 0.7145   )             
#> 
#> Other parameters: 
#>   1.  eval degree (density)          -2.1242  ( 0.3269   )    0.0572   
#>   2.  eval transitive triads          1.2542  ( 0.2283   )    0.0215   
#>   3.  eval degree act+pop             0.0292  ( 0.0314   )    0.0368   
#>   4.  eval same gender               -0.0686  ( 0.2107   )    0.0433   
#>   5.  eval ki alter                  -0.1316  ( 0.1765   )   -0.0451   
#>   6.  eval same dutch                -0.0597  ( 0.2178   )    0.0416   
#>   7.  eval pub_first abs. difference -0.0159  ( 0.0126   )    0.0668   
#> 
#> Overall maximum convergence ratio:    0.0833 
#> 
#> 
#> Total of 2150 iteration steps.
#> 
#> Covariance matrix of estimates (correlations below diagonal)
#> 
#>        0.107        0.010       -0.006       -0.020        0.007       -0.023       -0.001
#>        0.128        0.052       -0.004       -0.003       -0.005       -0.001        0.000
#>       -0.625       -0.535        0.001        0.000        0.001        0.000        0.000
#>       -0.293       -0.060       -0.024        0.044        0.006        0.000        0.000
#>        0.124       -0.135        0.135        0.164        0.031       -0.002        0.000
#>       -0.324       -0.023       -0.014       -0.008       -0.056        0.047        0.000
#>       -0.268        0.013       -0.104        0.040        0.030       -0.103        0.000
#> 
#> Derivative matrix of expected statistics X by parameters:
#> 
#>       92.039       65.105     2785.988      104.332      -77.480      104.361     1837.033
#>       76.899      145.757     3402.557       96.206      -74.815       88.807     1680.976
#>     1022.445     1116.382    39489.092     1199.402     -966.011     1160.916    21588.164
#>       53.610       41.351     1646.171      108.696      -55.990       63.578     1095.914
#>      -45.550      -34.666    -1499.601      -60.683      140.164      -48.091     -975.785
#>       51.927       36.911     1572.825       61.512      -43.425      100.842     1097.384
#>      924.226      671.495    28880.302     1038.098     -870.893     1094.218    30937.288
#> 
#> Covariance matrix of X (correlations below diagonal):
#> 
#>      149.708      215.401     5976.673      178.875     -153.336      169.204     3117.764
#>        0.757      540.663    11353.760      274.522     -260.320      241.620     4790.380
#>        0.913        0.913   286057.643     7264.671    -6784.867     6690.639   129457.084
#>        0.826        0.667        0.768      312.946     -206.492      213.066     3810.616
#>       -0.579       -0.517       -0.586       -0.539      468.429     -169.615    -3521.787
#>        0.835        0.628        0.756        0.728       -0.473      274.033     3586.345
#>        0.851        0.688        0.809        0.720       -0.544        0.724    89600.248