R Notebook

ende <- read.table("results_ende.tsv", sep = '\t', header = TRUE)
head(ende)

MuCoW EN–DE

model = lm(formula = min.precision ~ newstest19 + is_student, data = ende)
summary(model)

Call:
lm(formula = min.precision ~ newstest19 + is_student, data = ende)

Residuals:
    Min      1Q  Median      3Q     Max 
-5.4074 -1.9125  0.3702  1.7782  5.4800 

Coefficients:
              Estimate Std. Error t value Pr(>|t|)    
(Intercept)     1.9228     4.8948   0.393    0.698    
newstest19      1.7783     0.1357  13.102 3.24e-13 ***
is_studentyes  -6.4706     1.0064  -6.430 6.87e-07 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 2.663 on 27 degrees of freedom
Multiple R-squared:  0.8758,    Adjusted R-squared:  0.8666 
F-statistic:  95.2 on 2 and 27 DF,  p-value: 5.891e-13

WinoMT EN–DE

model = lm(formula = min.accuracy ~ newstest19 + is_student, data = ende)
summary(model)

Call:
lm(formula = min.accuracy ~ newstest19 + is_student, data = ende)

Residuals:
    Min      1Q  Median      3Q     Max 
-14.396  -5.516   1.487   5.760  14.768 

Coefficients:
               Estimate Std. Error t value Pr(>|t|)    
(Intercept)   -151.3212    15.5601  -9.725 2.57e-10 ***
newstest19       4.9964     0.4315  11.580 5.58e-12 ***
is_studentyes   -7.5567     3.1992  -2.362   0.0256 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 8.467 on 27 degrees of freedom
Multiple R-squared:  0.8326,    Adjusted R-squared:  0.8202 
F-statistic: 67.16 on 2 and 27 DF,  p-value: 3.31e-11

enru <- read.table("results_enru.tsv", sep = '\t', header = TRUE)
head(enru)

MuCoW EN–RU

model = lm(formula = min.precision ~ newstest19 + is_student, data = enru)
summary(model)

Call:
lm(formula = min.precision ~ newstest19 + is_student, data = enru)

Residuals:
    Min      1Q  Median      3Q     Max 
-5.8919 -0.9278  0.5528  1.1199  5.1239 

Coefficients:
              Estimate Std. Error t value Pr(>|t|)    
(Intercept)    41.1290     3.6510  11.265 6.51e-12 ***
newstest19      1.3881     0.1349  10.286 5.14e-11 ***
is_studentyes  -2.1393     0.8926  -2.397   0.0235 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 2.446 on 28 degrees of freedom
Multiple R-squared:  0.8053,    Adjusted R-squared:  0.7914 
F-statistic: 57.92 on 2 and 28 DF,  p-value: 1.121e-10

WinoMT EN–RU

model = lm(formula = min.accuracy ~ newstest19 + is_student, data = enru)
summary(model)

Call:
lm(formula = min.accuracy ~ newstest19 + is_student, data = enru)

Residuals:
     Min       1Q   Median       3Q      Max 
-12.2914  -2.7848   0.7399   3.4056   8.4861 

Coefficients:
              Estimate Std. Error t value Pr(>|t|)    
(Intercept)   -45.9883     8.5200  -5.398 9.36e-06 ***
newstest19      2.5210     0.3149   8.005 1.02e-08 ***
is_studentyes  -5.2117     2.0831  -2.502   0.0185 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 5.707 on 28 degrees of freedom
Multiple R-squared:  0.7247,    Adjusted R-squared:  0.7051 
F-statistic: 36.86 on 2 and 28 DF,  p-value: 1.434e-08