##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
Spelling task.
library(gplots)
##
## Attaching package: 'gplots'
## The following object is masked from 'package:stats':
##
## lowess
a$msg.day<-as.numeric(a$msg.day)
summary(lm(spellingscore ~ cond + OS + a$own.cellphone + cell.freq+cell.keyboard + texts.per.day + use.twitter + msg.apps + msg.day + attn.check, a))
##
## Call:
## lm(formula = spellingscore ~ cond + OS + a$own.cellphone + cell.freq +
## cell.keyboard + texts.per.day + use.twitter + msg.apps +
## msg.day + attn.check, data = a)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.5055 -0.8466 0.3407 1.1709 2.5588
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.538258 0.904430 8.335 2.39e-15 ***
## condblog 0.038595 0.278419 0.139 0.8898
## condtext 0.064110 0.234600 0.273 0.7848
## condtwitter 0.491830 0.246212 1.998 0.0466 *
## OSphone -0.295914 0.310923 -0.952 0.3420
## a$own.cellphone NA NA NA NA
## cell.freq -0.151347 0.083288 -1.817 0.0701 .
## cell.keyboard 0.396305 0.244434 1.621 0.1059
## texts.per.day -0.001563 0.001620 -0.965 0.3352
## use.twitter -0.408127 0.186630 -2.187 0.0295 *
## msg.apps 0.246384 0.271112 0.909 0.3642
## msg.day 0.002421 0.006687 0.362 0.7175
## attn.check 0.092212 0.064194 1.436 0.1519
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.629 on 317 degrees of freedom
## (2349 observations deleted due to missingness)
## Multiple R-squared: 0.06048, Adjusted R-squared: 0.02788
## F-statistic: 1.855 on 11 and 317 DF, p-value: 0.04458
m1<-aov(spellingscore ~ cond, data=a)
summary(m1)
## Df Sum Sq Mean Sq F value Pr(>F)
## cond 3 14.5 4.827 1.631 0.182
## Residuals 356 1053.8 2.960
## 2318 observations deleted due to missingness
plotmeans(a$spellingscore~a$cond)
TukeyHSD(m1)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = spellingscore ~ cond, data = a)
##
## $cond
## diff lwr upr p adj
## blog-acontrol 0.16592986 -0.5472276 0.8790873 0.9318349
## text-acontrol 0.10927961 -0.4988366 0.7173959 0.9668619
## twitter-acontrol 0.52225169 -0.1064373 1.1509407 0.1412698
## text-blog -0.05665025 -0.7923636 0.6790631 0.9972159
## twitter-blog 0.35632184 -0.3964853 1.1091290 0.6134288
## twitter-text 0.41297209 -0.2411918 1.0671360 0.3633209
Sentence completion task.
summary(lm(sentencescore ~ cond + OS + a$own.cellphone + cell.freq+cell.keyboard + texts.per.day + use.twitter + msg.apps + msg.day + attn.check, a))
##
## Call:
## lm(formula = sentencescore ~ cond + OS + a$own.cellphone + cell.freq +
## cell.keyboard + texts.per.day + use.twitter + msg.apps +
## msg.day + attn.check, data = a)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.1711 -1.6341 -0.2961 1.4107 5.0629
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.063306 1.552917 2.617 0.00933 **
## condblog 0.413019 0.359963 1.147 0.25213
## condtext 0.319326 0.329496 0.969 0.33326
## condtwitter -0.001620 0.348219 -0.005 0.99629
## OSphone -0.089952 0.415782 -0.216 0.82887
## a$own.cellphone NA NA NA NA
## cell.freq 0.001346 0.108667 0.012 0.99012
## cell.keyboard 0.348906 0.323726 1.078 0.28199
## texts.per.day -0.007390 0.002839 -2.603 0.00970 **
## use.twitter -0.138261 0.248827 -0.556 0.57886
## msg.apps -0.619701 0.364173 -1.702 0.08985 .
## msg.day -0.015563 0.009830 -1.583 0.11441
## attn.check 0.071573 0.157578 0.454 0.65001
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.16 on 301 degrees of freedom
## (2365 observations deleted due to missingness)
## Multiple R-squared: 0.04882, Adjusted R-squared: 0.01406
## F-statistic: 1.404 on 11 and 301 DF, p-value: 0.1697
m2<-aov(a$sentencescore ~ cond, data=a)
summary(m2)
## Df Sum Sq Mean Sq F value Pr(>F)
## cond 3 14 4.677 0.968 0.408
## Residuals 337 1629 4.834
## 2337 observations deleted due to missingness
plotmeans(a$sentencescore~a$cond)
TukeyHSD(m2)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = a$sentencescore ~ cond, data = a)
##
## $cond
## diff lwr upr p adj
## blog-acontrol 0.5443769 -0.3518585 1.4406123 0.3982204
## text-acontrol 0.3517502 -0.4785374 1.1820377 0.6935685
## twitter-acontrol 0.1324721 -0.7198810 0.9848253 0.9781062
## text-blog -0.1926267 -1.0909165 0.7056630 0.9455163
## twitter-blog -0.4119048 -1.3306283 0.5068188 0.6540214
## twitter-text -0.2192780 -1.0737910 0.6352349 0.9110562
Paragraph comprehension task.
summary(lm(Parscore ~ cond + OS + a$own.cellphone + cell.freq+cell.keyboard + texts.per.day + use.twitter + msg.apps + msg.day + attn.check, a))
##
## Call:
## lm(formula = Parscore ~ cond + OS + a$own.cellphone + cell.freq +
## cell.keyboard + texts.per.day + use.twitter + msg.apps +
## msg.day + attn.check, data = a)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.6034 -1.1332 -0.1360 0.9785 4.6698
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.6443356 1.3238618 1.242 0.2153
## condblog -0.7381103 0.3174562 -2.325 0.0208 *
## condtext -0.4862757 0.2810802 -1.730 0.0848 .
## condtwitter -0.4509622 0.2718701 -1.659 0.0983 .
## OSphone -0.8080267 0.3687569 -2.191 0.0293 *
## a$own.cellphone NA NA NA NA
## cell.freq 0.1168459 0.0925633 1.262 0.2079
## cell.keyboard 0.3270605 0.2895445 1.130 0.2596
## texts.per.day -0.0029957 0.0018407 -1.628 0.1048
## use.twitter 0.2854635 0.2062839 1.384 0.1675
## msg.apps 0.2095344 0.3182134 0.658 0.5108
## msg.day -0.0009114 0.0077927 -0.117 0.9070
## attn.check 0.0898036 0.1220665 0.736 0.4625
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.661 on 273 degrees of freedom
## (2393 observations deleted due to missingness)
## Multiple R-squared: 0.08846, Adjusted R-squared: 0.05174
## F-statistic: 2.409 on 11 and 273 DF, p-value: 0.007195
m3<-aov(a$Parscore ~ cond, data=a)
summary(lm(a$Parscore ~ cond, data=a))
##
## Call:
## lm(formula = a$Parscore ~ cond, data = a)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.4412 -1.2805 -0.0377 0.9623 4.7195
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.8986 0.2062 18.908 < 2e-16 ***
## condblog -0.8608 0.3128 -2.752 0.00629 **
## condtext -0.6181 0.2798 -2.209 0.02793 *
## condtwitter -0.4574 0.2670 -1.713 0.08770 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.713 on 302 degrees of freedom
## (2372 observations deleted due to missingness)
## Multiple R-squared: 0.0276, Adjusted R-squared: 0.01794
## F-statistic: 2.858 on 3 and 302 DF, p-value: 0.03728
summary(m3)
## Df Sum Sq Mean Sq F value Pr(>F)
## cond 3 25.1 8.383 2.858 0.0373 *
## Residuals 302 885.9 2.933
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 2372 observations deleted due to missingness
plotmeans(a$Parscore~a$cond)
TukeyHSD(m3)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = a$Parscore ~ cond, data = a)
##
## $cond
## diff lwr upr p adj
## blog-acontrol -0.8608149 -1.6689782 -0.05265154 0.0317597
## text-acontrol -0.6180629 -1.3408971 0.10477131 0.1231115
## twitter-acontrol -0.4573743 -1.1470662 0.23231766 0.3185683
## text-blog 0.2427520 -0.5370849 1.02258877 0.8524445
## twitter-blog 0.4034406 -0.3457797 1.15266092 0.5058893
## twitter-text 0.1606887 -0.4955829 0.81696026 0.9214844
Analogy Task.
summary(lm(Analogyscore ~ cond + OS + a$own.cellphone + cell.freq+cell.keyboard + texts.per.day + use.twitter + msg.apps + msg.day + attn.check, a))
##
## Call:
## lm(formula = Analogyscore ~ cond + OS + a$own.cellphone + cell.freq +
## cell.keyboard + texts.per.day + use.twitter + msg.apps +
## msg.day + attn.check, data = a)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.8132 -1.3907 -0.2791 1.0158 5.9329
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.7602284 0.9609733 2.872 0.00435 **
## condblog -0.1067808 0.3123822 -0.342 0.73271
## condtext 0.2516285 0.2668414 0.943 0.34642
## condtwitter 0.0951572 0.2829906 0.336 0.73690
## OSphone -0.4008092 0.3856550 -1.039 0.29947
## a$own.cellphone NA NA NA NA
## cell.freq 0.0180723 0.0890037 0.203 0.83923
## cell.keyboard 0.1735767 0.2816876 0.616 0.53821
## texts.per.day -0.0007655 0.0014158 -0.541 0.58910
## use.twitter -0.0578569 0.2099616 -0.276 0.78307
## msg.apps 0.0326027 0.3019718 0.108 0.91409
## msg.day -0.0032804 0.0077088 -0.426 0.67074
## attn.check 0.0549604 0.0845974 0.650 0.51639
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.8 on 311 degrees of freedom
## (2355 observations deleted due to missingness)
## Multiple R-squared: 0.01483, Adjusted R-squared: -0.02001
## F-statistic: 0.4256 on 11 and 311 DF, p-value: 0.9442
m4<-aov(a$Analogyscore ~ cond, data=a)
summary(m4)
## Df Sum Sq Mean Sq F value Pr(>F)
## cond 3 7.1 2.361 0.724 0.538
## Residuals 347 1130.9 3.259
## 2327 observations deleted due to missingness
plotmeans(a$Analogyscore~a$cond)
TukeyHSD(m4)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = a$Analogyscore ~ cond, data = a)
##
## $cond
## diff lwr upr p adj
## blog-acontrol -0.02424242 -0.8123574 0.7638726 0.9998202
## text-acontrol 0.27414330 -0.3810035 0.9292901 0.7018521
## twitter-acontrol -0.06451613 -0.7425820 0.6135498 0.9947860
## text-blog 0.29838573 -0.4748330 1.0716044 0.7516721
## twitter-blog -0.04027370 -0.8330052 0.7524578 0.9991933
## twitter-text -0.33865943 -0.9993525 0.3220336 0.5487238
Math Task.
summary(lm(math ~ cond + OS + a$own.cellphone + cell.freq+cell.keyboard + texts.per.day + use.twitter + msg.apps + msg.day + attn.check, a))
##
## Call:
## lm(formula = math ~ cond + OS + a$own.cellphone + cell.freq +
## cell.keyboard + texts.per.day + use.twitter + msg.apps +
## msg.day + attn.check, data = a)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.1042 -1.5444 -0.0417 1.2266 5.1176
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.0781140 1.5218196 0.708 0.479
## condblog 0.0606240 0.3533244 0.172 0.864
## condtext -0.0227284 0.3050734 -0.075 0.941
## condtwitter -0.2884066 0.3092722 -0.933 0.352
## OSphone -0.1854774 0.4083263 -0.454 0.650
## a$own.cellphone NA NA NA NA
## cell.freq 0.0286320 0.0980985 0.292 0.771
## cell.keyboard 0.2050974 0.3074191 0.667 0.505
## texts.per.day 0.0017316 0.0029108 0.595 0.552
## use.twitter 0.1206589 0.2288022 0.527 0.598
## msg.apps 0.3874785 0.3372841 1.149 0.252
## msg.day -0.0002972 0.0091529 -0.032 0.974
## attn.check 0.1979282 0.1632916 1.212 0.226
##
## Residual standard error: 1.934 on 287 degrees of freedom
## (2379 observations deleted due to missingness)
## Multiple R-squared: 0.02375, Adjusted R-squared: -0.01367
## F-statistic: 0.6347 on 11 and 287 DF, p-value: 0.7986
m5<-aov(math ~ cond, data=a)
summary(m5)
## Df Sum Sq Mean Sq F value Pr(>F)
## cond 3 6 1.990 0.543 0.653
## Residuals 331 1214 3.666
## 2343 observations deleted due to missingness
plotmeans(a$math~a$cond)
TukeyHSD(m5)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = math ~ cond, data = a)
##
## $cond
## diff lwr upr p adj
## blog-acontrol 0.14019608 -0.6934392 0.9738314 0.9725364
## text-acontrol 0.03405573 -0.7040901 0.7722015 0.9993946
## twitter-acontrol -0.22910217 -0.9672480 0.5090436 0.8537244
## text-blog -0.10614035 -0.9214181 0.7091374 0.9868942
## twitter-blog -0.36929825 -1.1845760 0.4459795 0.6464535
## twitter-text -0.26315789 -0.9805067 0.4541909 0.7793216