library(mapproj) # map
## Warning: package 'mapproj' was built under R version 3.4.2
## Loading required package: maps
## Warning: package 'maps' was built under R version 3.4.2
library(reshape2) # melt
library(nparcomp) # gao_cs
## Warning: package 'nparcomp' was built under R version 3.4.2
## Loading required package: multcomp
## Warning: package 'multcomp' was built under R version 3.4.2
## Loading required package: mvtnorm
## Loading required package: survival
## Loading required package: TH.data
## Warning: package 'TH.data' was built under R version 3.4.2
## Loading required package: MASS
##
## Attaching package: 'TH.data'
## The following object is masked from 'package:MASS':
##
## geyser
library(car) # leveneTest and Anova Type III
## Warning: package 'car' was built under R version 3.4.2
library(heplots) # etasquared
## Warning: package 'heplots' was built under R version 3.4.2
library(MASS) # lda
library(psy) # cronbach
library(igraph) # network graphs
## Warning: package 'igraph' was built under R version 3.4.2
##
## Attaching package: 'igraph'
## The following objects are masked from 'package:stats':
##
## decompose, spectrum
## The following object is masked from 'package:base':
##
## union
library(lsr) # partial eta squared
library(psych) # KMO
##
## Attaching package: 'psych'
## The following object is masked from 'package:psy':
##
## wkappa
## The following object is masked from 'package:car':
##
## logit
library(biotools) # M Box test
## Warning: package 'biotools' was built under R version 3.4.2
## Loading required package: rpanel
## Warning: package 'rpanel' was built under R version 3.4.2
## Loading required package: tcltk
## Package `rpanel', version 1.1-3: type help(rpanel) for summary information
## Loading required package: tkrplot
## Loading required package: lattice
## Loading required package: SpatialEpi
## Warning: package 'SpatialEpi' was built under R version 3.4.2
## Loading required package: sp
## Warning: package 'sp' was built under R version 3.4.2
##
## Attaching package: 'SpatialEpi'
## The following object is masked from 'package:igraph':
##
## normalize
## ---
## biotools version 3.1
##
##
## Attaching package: 'biotools'
## The following object is masked from 'package:heplots':
##
## boxM
library(vcd) # goodfit
## Warning: package 'vcd' was built under R version 3.4.2
## Loading required package: grid
library(agricolae)
## Warning: package 'agricolae' was built under R version 3.4.2
##
## Attaching package: 'agricolae'
## The following object is masked from 'package:igraph':
##
## similarity
library(lavaan) # SEM4
## Warning: package 'lavaan' was built under R version 3.4.2
## This is lavaan 0.5-23.1097
## lavaan is BETA software! Please report any bugs.
##
## Attaching package: 'lavaan'
## The following object is masked from 'package:psych':
##
## cor2cov
library(semPlot) # SEM graph
## Warning: package 'semPlot' was built under R version 3.4.2
library(Hmisc) # correlation matrix
## Warning: package 'Hmisc' was built under R version 3.4.2
## Loading required package: Formula
## Loading required package: ggplot2
##
## Attaching package: 'ggplot2'
## The following objects are masked from 'package:psych':
##
## %+%, alpha
##
## Attaching package: 'Hmisc'
## The following object is masked from 'package:psych':
##
## describe
## The following objects are masked from 'package:base':
##
## format.pval, round.POSIXt, trunc.POSIXt, units
library(MVN) # multivariate normality
## Warning: package 'MVN' was built under R version 3.4.2
## sROC 0.1-2 loaded
##
## Attaching package: 'MVN'
## The following object is masked from 'package:psych':
##
## mardia
library(mvoutlier) # multivariate outlier
## Warning: package 'mvoutlier' was built under R version 3.4.2
## Loading required package: sgeostat
library(fitdistrplus)
## Warning: package 'fitdistrplus' was built under R version 3.4.2
library(logspline)
cat("\014") # cleans screen
rm(list=ls(all=TRUE)) # remove variables in working memory
setwd("C:/Users/Erik Ernesto Vazquez/Documents") # sets working directory
MainStudy<-read.csv("Main Study 1158 April 2017.csv", skip=2, header=F) # reads raw data from Qualtrics
names(MainStudy)<-names(read.csv("Main Study 1158 April 2017.csv")) # assigns headers and names to data frame
MainStudy<-subset(MainStudy,MainStudy$X60<5) ## Non-repeated measures
## Pretest 1
MainStudy1<-subset(MainStudy,MainStudy$X206=="Credence")
MainStudy1<-MainStudy1[order(MainStudy1$X188),]
MainStudy1<-rbind(MainStudy1[1:20,],MainStudy1[361:342,])
write.csv(MainStudy1,file="Pretest1.csv")
MainStudy1$Item1<-MainStudy1[40]+MainStudy1[41]+MainStudy1[42]
MainStudy1$Item2<-MainStudy1[43]+MainStudy1[44]+MainStudy1[45]
cronbach(cbind(MainStudy1$Item1,MainStudy1$Item2)) ## >.91 richness
## $sample.size
## [1] 40
##
## $number.of.items
## [1] 2
##
## $alpha
## [1] 0.9110783
MainStudyMelt1<-melt(MainStudy1,id.vars=c("X1","X40","X41","X42"),
measure.vars=c("X40","X41","X42"),
variable.name="ContRich1", value.name="Item1")
MainStudyMelt2<-melt(MainStudy1,id.vars=c("X1","X43","X44","X45"),
measure.vars=c("X43","X44","X45"),
variable.name="ContRich2", value.name="Item2")
cronbach(cbind(MainStudyMelt1$Item1,MainStudyMelt2$Item2)) ## Cronabch 0.88 richness
## $sample.size
## [1] 120
##
## $number.of.items
## [1] 2
##
## $alpha
## [1] 0.882327
MainStudy1$SMP1<-(MainStudy1$X40+MainStudy1$X43)/2
MainStudy1$SMP2<-(MainStudy1$X41+MainStudy1$X44)/2
MainStudy1$SMP3<-(MainStudy1$X42+MainStudy1$X45)/2
MainStudy1=data.frame(MainStudy1[1],MainStudy1[187],MainStudy1[188],MainStudy1[210],MainStudy1[211],MainStudy1[212])
names(MainStudy1)<-c("ID","Age","Gender","YouTube","Twitter","Facebook")
MainStudy1$Age<-2014-MainStudy1$Age
summary(MainStudy1)
## ID Age Gender YouTube
## R_0ARyBi0yqbL7K73: 1 Min. :20.00 Min. :1.0 Min. :2.000
## R_0fCgye3gZBK7vLv: 1 1st Qu.:23.75 1st Qu.:1.0 1st Qu.:5.750
## R_0JHSDzmtWPhk8tf: 1 Median :31.50 Median :1.5 Median :6.750
## R_0keDqeq4UghhPAV: 1 Mean :33.98 Mean :1.5 Mean :6.412
## R_0rdRaH8XTyHgCHj: 1 3rd Qu.:39.00 3rd Qu.:2.0 3rd Qu.:7.500
## R_0VWipgRAOmVbN2d: 1 Max. :68.00 Max. :2.0 Max. :9.000
## (Other) :34
## Twitter Facebook
## Min. :1.000 Min. :1.000
## 1st Qu.:2.000 1st Qu.:4.000
## Median :3.250 Median :5.000
## Mean :3.487 Mean :5.075
## 3rd Qu.:4.500 3rd Qu.:6.500
## Max. :7.500 Max. :8.500
##
sd(MainStudy1$Age)
## [1] 12.32568
MainStudy1<-melt(MainStudy1,id.vars=c("ID","Age","Gender","YouTube","Twitter","Facebook"),measure.vars=c("YouTube","Twitter","Facebook"),variable.name="SMP", value.name="InformationRichness")
summary(aov(InformationRichness~SMP,MainStudy1))
## Df Sum Sq Mean Sq F value Pr(>F)
## SMP 2 171.5 85.76 26.69 2.82e-10 ***
## Residuals 117 376.0 3.21
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
aov.out<-aov(InformationRichness~SMP,MainStudy1)
TukeyHSD(aov.out) ## Three levels of information richness
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = InformationRichness ~ SMP, data = MainStudy1)
##
## $SMP
## diff lwr upr p adj
## Twitter-YouTube -2.9250 -3.8765439 -1.9734561 0.0000000
## Facebook-YouTube -1.3375 -2.2890439 -0.3859561 0.0032375
## Facebook-Twitter 1.5875 0.6359561 2.5390439 0.0003770
aggregate(MainStudy1$InformationRichness,list(MainStudy1$SMP),mean)
## Group.1 x
## 1 YouTube 6.4125
## 2 Twitter 3.4875
## 3 Facebook 5.0750
aggregate(MainStudy1$InformationRichness,list(MainStudy1$SMP),sd)
## Group.1 x
## 1 YouTube 1.804153
## 2 Twitter 1.766815
## 3 Facebook 1.806505
## Pretest 2
MainStudy1<-subset(MainStudy,MainStudy$X205=="Twitter")
MainStudy1<-MainStudy1[order(MainStudy1$X188,MainStudy1$X206,MainStudy1$X117),]
MainStudy1<-rbind(MainStudy1[1:7,],MainStudy1[72:78,],MainStudy1[123:129,],MainStudy1[180:186,],MainStudy1[234:240,],MainStudy1[294:300,])
write.csv(MainStudy1,file="Pretest2.csv")
summary(aov(X187~X207,MainStudy1)) ## Age
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 2 146.7 73.36 0.95 0.395
## Residuals 39 3010.4 77.19
chisq.test(MainStudy1$X207,MainStudy1$X188) ## Gender
##
## Pearson's Chi-squared test
##
## data: MainStudy1$X207 and MainStudy1$X188
## X-squared = 0, df = 2, p-value = 1
summary(aov(X189~X207,MainStudy1)) ## Income
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 2 5.33 2.667 0.615 0.546
## Residuals 39 169.14 4.337
summary(aov(X194~X207,MainStudy1)) ## Education
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 2 5.14 2.571 0.76 0.475
## Residuals 39 132.00 3.385
summary(aov(X117~X207,MainStudy1)) ## StimuliTime
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 2 23.15 11.574 2.062 0.141
## Residuals 39 218.90 5.613
aggregate(MainStudy1$X117,list(MainStudy1$X207),mean)
## Group.1 x
## 1 Twitter-Credence 34.68564
## 2 Twitter-Experience 32.87679
## 3 Twitter-Search 33.61936
summary(aov(X60~X207,MainStudy1)) ## Familiarity
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 2 0.905 0.4524 1.098 0.344
## Residuals 39 16.071 0.4121
MainStudy1$Age<-2014-MainStudy1$X187
summary(MainStudy1$Age)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 19.00 24.00 28.50 30.21 33.75 61.00
sd(MainStudy1$Age)
## [1] 8.775064
table(MainStudy1$X207,MainStudy1$X188)
##
## 1 2
## Facebook-Credence 0 0
## Facebook-Experience 0 0
## Facebook-Search 0 0
## Twitter-Credence 7 7
## Twitter-Experience 7 7
## Twitter-Search 7 7
## YouTube-Credence 0 0
## YouTube-Experience 0 0
## YouTube-Search 0 0
cronbach(cbind(MainStudy1$X180,MainStudy1$X182)) ## >.90 digital involvement with product category
## $sample.size
## [1] 42
##
## $number.of.items
## [1] 2
##
## $alpha
## [1] 0.9017839
MainStudy1$ProductInvol<-(MainStudy1$X180+MainStudy1$X182)/2
summary(aov(ProductInvol~X206,MainStudy1))
## Df Sum Sq Mean Sq F value Pr(>F)
## X206 2 66.08 33.04 7.633 0.00159 **
## Residuals 39 168.82 4.33
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
aov.out<-aov(ProductInvol~X206,MainStudy1)
TukeyHSD(aov.out) ## Two levels of products ease to evaluate quality | Shopping vs non-shopping OR Credence vs Non-Credence
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = ProductInvol ~ X206, data = MainStudy1)
##
## $X206
## diff lwr upr p adj
## Experience-Credence 2.4642857 0.5484220 4.380149 0.0089773
## Search-Credence 2.8214286 0.9055649 4.737292 0.0025868
## Search-Experience 0.3571429 -1.5587208 2.273007 0.8928502
aggregate(MainStudy1$ProductInvol,list(MainStudy1$X207),mean)
## Group.1 x
## 1 Twitter-Credence 3.785714
## 2 Twitter-Experience 6.250000
## 3 Twitter-Search 6.607143
aggregate(MainStudy1$ProductInvol,list(MainStudy1$X207),sd)
## Group.1 x
## 1 Twitter-Credence 2.880400
## 2 Twitter-Experience 1.565862
## 3 Twitter-Search 1.495873
MainStudy1<-subset(MainStudy1,MainStudy1$X206!="Credence")
table(MainStudy1$X206)
##
## Credence Experience Search
## 0 14 14
t.test(MainStudy1$ProductInvol~MainStudy1$X206,MainStudy1,paired=T)
##
## Paired t-test
##
## data: MainStudy1$ProductInvol by MainStudy1$X206
## t = -0.83602, df = 13, p-value = 0.4182
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.2800428 0.5657571
## sample estimates:
## mean of the differences
## -0.3571429
## Pretest 3
MainStudy1<-read.csv("Main Study 1158 April 2017.csv", skip=2, header=F) # reads raw data from Qualtrics
names(MainStudy1)<-names(read.csv("Main Study 1158 April 2017.csv")) # assigns headers and names to data frame
MainStudy1<-subset(MainStudy1,MainStudy1$X60>4&MainStudy1$X112>0) ## Repeated measures
MainStudy1<-MainStudy1[order(MainStudy1$X207,MainStudy1$X188,MainStudy1$X187),]
print(cbind(MainStudy1[,207],MainStudy1[,188]))
## [,1] [,2]
## [1,] 1 1
## [2,] 1 2
## [3,] 1 2
## [4,] 1 2
## [5,] 1 2
## [6,] 1 2
## [7,] 2 1
## [8,] 2 1
## [9,] 2 1
## [10,] 2 1
## [11,] 2 1
## [12,] 2 1
## [13,] 2 1
## [14,] 2 1
## [15,] 2 1
## [16,] 2 1
## [17,] 2 1
## [18,] 2 2
## [19,] 2 2
## [20,] 2 2
## [21,] 2 2
## [22,] 2 2
## [23,] 3 1
## [24,] 3 2
## [25,] 3 2
## [26,] 3 2
## [27,] 4 1
## [28,] 4 2
## [29,] 4 2
## [30,] 4 2
## [31,] 4 2
## [32,] 5 1
## [33,] 5 1
## [34,] 5 1
## [35,] 5 1
## [36,] 5 1
## [37,] 5 1
## [38,] 5 1
## [39,] 5 2
## [40,] 5 2
## [41,] 5 2
## [42,] 5 2
## [43,] 6 1
## [44,] 6 1
## [45,] 6 2
## [46,] 6 2
## [47,] 6 2
## [48,] 6 2
## [49,] 6 2
## [50,] 6 2
## [51,] 7 1
## [52,] 7 1
## [53,] 7 1
## [54,] 7 1
## [55,] 7 2
## [56,] 7 2
## [57,] 7 2
## [58,] 7 2
## [59,] 7 2
## [60,] 7 2
## [61,] 8 1
## [62,] 8 1
## [63,] 8 1
## [64,] 8 1
## [65,] 8 1
## [66,] 8 1
## [67,] 8 1
## [68,] 8 2
## [69,] 8 2
## [70,] 9 1
## [71,] 9 2
## [72,] 9 2
## [73,] 9 2
MainStudy1<-rbind(MainStudy1[1,],MainStudy1[5:6,],MainStudy1[8:9,],MainStudy1[21:22,],
MainStudy1[23,],MainStudy1[25:26,],MainStudy1[27,],MainStudy1[30:31,],
MainStudy1[32:33,],MainStudy1[42,],MainStudy1[43:44,],MainStudy1[50,],
MainStudy1[51:52,],MainStudy1[60,],MainStudy1[61:62,],MainStudy1[69,],
MainStudy1[70:71,],MainStudy1[73,])
write.csv(MainStudy1,file="Pretest3.csv")
summary(aov(X187~X207,MainStudy1)) ## Age
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 759.3 94.92 1.813 0.137
## Residuals 19 994.7 52.35
aggregate(MainStudy1$X187,list(MainStudy1$X207),mean)
## Group.1 x
## 1 Facebook-Credence 1986.000
## 2 Facebook-Experience 1985.000
## 3 Facebook-Search 1993.333
## 4 Twitter-Credence 1983.667
## 5 Twitter-Experience 1985.667
## 6 Twitter-Search 1984.333
## 7 YouTube-Credence 1979.667
## 8 YouTube-Experience 1985.667
## 9 YouTube-Search 1972.333
2014-mean(MainStudy1$X187)
## [1] 30
sd(MainStudy1$X187)
## [1] 8.05996
chisq.test(MainStudy1$X207,MainStudy1$X188) ## Gender
## Warning in chisq.test(MainStudy1$X207, MainStudy1$X188): Chi-squared
## approximation may be incorrect
##
## Pearson's Chi-squared test
##
## data: MainStudy1$X207 and MainStudy1$X188
## X-squared = 2.6667, df = 8, p-value = 0.9535
table(MainStudy1$X207,MainStudy1$X188)
##
## 1 2
## Facebook-Credence 1 2
## Facebook-Experience 2 2
## Facebook-Search 1 2
## Twitter-Credence 1 2
## Twitter-Experience 2 1
## Twitter-Search 2 1
## YouTube-Credence 2 1
## YouTube-Experience 2 1
## YouTube-Search 1 2
table(MainStudy1$X188)
##
## 1 2
## 14 14
table(MainStudy1$X207)
##
## Facebook-Credence Facebook-Experience Facebook-Search
## 3 4 3
## Twitter-Credence Twitter-Experience Twitter-Search
## 3 3 3
## YouTube-Credence YouTube-Experience YouTube-Search
## 3 3 3
summary(aov(X189~X207,MainStudy1)) ## Income
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 15.93 1.991 0.446 0.878
## Residuals 19 84.75 4.461
summary(aov(X194~X207,MainStudy1)) ## Education
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 15.01 1.876 0.514 0.831
## Residuals 19 69.42 3.654
summary(aov(X202~X207,MainStudy1)) ## Location 1
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 772.5 96.56 1.115 0.397
## Residuals 19 1645.6 86.61
summary(aov(X203~X207,MainStudy1)) ## Location 2
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 17142 2143 0.827 0.589
## Residuals 19 49208 2590
summary(aov(X117~X207,MainStudy1)) ## StimuliTime
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 3148 393.5 1.149 0.377
## Residuals 19 6506 342.4
cronbach(cbind(MainStudy1$X151,MainStudy1$X152,MainStudy1$X153))
## $sample.size
## [1] 28
##
## $number.of.items
## [1] 3
##
## $alpha
## [1] 0.954886
MainStudy1$WOM<-(MainStudy1$X151+MainStudy1$X152+MainStudy1$X153)/3
cronbach(cbind(MainStudy1$X94,MainStudy1$X95,MainStudy1$X96))
## $sample.size
## [1] 28
##
## $number.of.items
## [1] 3
##
## $alpha
## [1] 0.9447666
MainStudy1$WOM1<-(MainStudy1$X94+MainStudy1$X95+MainStudy1$X96)/3
MainStudy1$WOMChange<-MainStudy1$WOM-MainStudy1$WOM1
mean(MainStudy1$WOM1)
## [1] 5.25
mean(MainStudy1$WOM)
## [1] 5.940476
aggregate(MainStudy1$WOM1,list(MainStudy1$X207),mean)
## Group.1 x
## 1 Facebook-Credence 4.333333
## 2 Facebook-Experience 3.500000
## 3 Facebook-Search 5.333333
## 4 Twitter-Credence 4.333333
## 5 Twitter-Experience 4.555556
## 6 Twitter-Search 6.444444
## 7 YouTube-Credence 5.888889
## 8 YouTube-Experience 6.444444
## 9 YouTube-Search 7.000000
aggregate(MainStudy1$WOM,list(MainStudy1$X207),mean)
## Group.1 x
## 1 Facebook-Credence 5.666667
## 2 Facebook-Experience 4.500000
## 3 Facebook-Search 5.555556
## 4 Twitter-Credence 4.111111
## 5 Twitter-Experience 6.000000
## 6 Twitter-Search 7.222222
## 7 YouTube-Credence 5.888889
## 8 YouTube-Experience 6.888889
## 9 YouTube-Search 8.111111
aggregate(MainStudy1$WOMChange,list(MainStudy1$X207),mean)
## Group.1 x
## 1 Facebook-Credence 1.3333333
## 2 Facebook-Experience 1.0000000
## 3 Facebook-Search 0.2222222
## 4 Twitter-Credence -0.2222222
## 5 Twitter-Experience 1.4444444
## 6 Twitter-Search 0.7777778
## 7 YouTube-Credence 0.0000000
## 8 YouTube-Experience 0.4444444
## 9 YouTube-Search 1.1111111
aggregate(MainStudy1$WOMChange,list(MainStudy1$X206),mean)
## Group.1 x
## 1 Credence 0.3703704
## 2 Experience 0.9666667
## 3 Search 0.7037037
aggregate(MainStudy1$WOMChange,list(MainStudy1$X205),mean)
## Group.1 x
## 1 Facebook 0.8666667
## 2 Twitter 0.6666667
## 3 YouTube 0.5185185
summary(aov(WOMChange~X207,MainStudy1)) ## WOM Change group
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 8.651 1.081 0.848 0.574
## Residuals 19 24.222 1.275
summary(aov(WOMChange~X205*X206,MainStudy1)) ## WOM Change
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 0.582 0.2908 0.228 0.798
## X206 2 1.593 0.7963 0.625 0.546
## X205:X206 4 6.476 1.6191 1.270 0.316
## Residuals 19 24.222 1.2749
mean(MainStudy1$WOMChange)
## [1] 0.6904762
sd(MainStudy1$WOMChange)
## [1] 1.103412
t.test(MainStudy1$WOM,MainStudy1$WOM1,paired=T)
##
## Paired t-test
##
## data: MainStudy1$WOM and MainStudy1$WOM1
## t = 3.3112, df = 27, p-value = 0.002644
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.2626175 1.1183349
## sample estimates:
## mean of the differences
## 0.6904762
mean(MainStudy1$WOM1)
## [1] 5.25
sd(MainStudy1$WOM1)
## [1] 1.836137
mean(MainStudy1$WOM)
## [1] 5.940476
sd(MainStudy1$WOM)
## [1] 1.663091
5.9/5.2-1 ## global effect positive in %
## [1] 0.1346154
table(MainStudy1$WOMChange)
##
## -1 -0.666666666666667 -0.333333333333333
## 1 1 2
## 0 0.333333333333333 0.333333333333334
## 8 3 1
## 0.666666666666666 0.666666666666667 1
## 1 1 1
## 1.33333333333333 2 2.66666666666667
## 4 3 1
## 4
## 1
mean(MainStudy1$WOMChange)
## [1] 0.6904762
sd(MainStudy1$WOMChange)
## [1] 1.103412
aggregate(MainStudy1$WOMChange,list(MainStudy1$X207),mean)
## Group.1 x
## 1 Facebook-Credence 1.3333333
## 2 Facebook-Experience 1.0000000
## 3 Facebook-Search 0.2222222
## 4 Twitter-Credence -0.2222222
## 5 Twitter-Experience 1.4444444
## 6 Twitter-Search 0.7777778
## 7 YouTube-Credence 0.0000000
## 8 YouTube-Experience 0.4444444
## 9 YouTube-Search 1.1111111
MainStudy1$Valence<-ifelse(MainStudy1$WOMChange==0,"neutral",ifelse(MainStudy1$WOMChange>0,"positive","negative"))
aggregate(MainStudy1$WOMChange,list(MainStudy1$Valence),mean)
## Group.1 x
## 1 negative -0.5833333
## 2 neutral 0.0000000
## 3 positive 1.3541667
table(MainStudy1$Valence)
##
## negative neutral positive
## 4 8 16
aggregate(MainStudy1$WOM1,list(MainStudy1$Valence),mean)
## Group.1 x
## 1 negative 5.750000
## 2 neutral 5.458333
## 3 positive 5.020833
aggregate(MainStudy1$WOM,list(MainStudy1$Valence),mean)
## Group.1 x
## 1 negative 5.166667
## 2 neutral 5.458333
## 3 positive 6.375000
5.166667/5.750000-1 ## effect of negative in %
## [1] -0.1014492
6.375000/5.020833-1 ## effect of negative in %
## [1] 0.2697096
MainStudy1<-subset(MainStudy1,MainStudy1$Valence!="neutral")
t.test(abs(MainStudy1$WOMChange)~MainStudy1$Valence)
##
## Welch Two Sample t-test
##
## data: abs(MainStudy1$WOMChange) by MainStudy1$Valence
## t = -2.5859, df = 16.312, p-value = 0.01968
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.4017831 -0.1398835
## sample estimates:
## mean in group negative mean in group positive
## 0.5833333 1.3541667
aggregate(MainStudy1$WOMChange,list(MainStudy1$Valence),mean)
## Group.1 x
## 1 negative -0.5833333
## 2 positive 1.3541667
aggregate(MainStudy1$WOMChange,list(MainStudy1$Valence),sd)
## Group.1 x
## 1 negative 0.3191424
## 2 positive 1.0071504
## MainStudy
MainStudy1<-subset(MainStudy,MainStudy$X207=="YouTube-Search")
MainStudy2<-subset(MainStudy,MainStudy$X207=="YouTube-Experience")
MainStudy3<-subset(MainStudy,MainStudy$X207=="YouTube-Credence")
MainStudy4<-subset(MainStudy,MainStudy$X207=="Facebook-Search")
MainStudy5<-subset(MainStudy,MainStudy$X207=="Facebook-Experience")
MainStudy6<-subset(MainStudy,MainStudy$X207=="Facebook-Credence")
MainStudy7<-subset(MainStudy,MainStudy$X207=="Twitter-Search")
MainStudy8<-subset(MainStudy,MainStudy$X207=="Twitter-Experience")
MainStudy9<-subset(MainStudy,MainStudy$X207=="Twitter-Credence")
MainStudy1<-MainStudy1[order(MainStudy1$X188),]
MainStudy2<-MainStudy2[order(MainStudy2$X188),]
MainStudy3<-MainStudy3[order(MainStudy3$X188,MainStudy3$X202,MainStudy3$X117),]
MainStudy4<-MainStudy4[order(MainStudy4$X188),]
MainStudy5<-MainStudy5[order(MainStudy5$X188),]
MainStudy6<-MainStudy6[order(MainStudy6$X188),]
MainStudy7<-MainStudy7[order(MainStudy7$X188,MainStudy7$X187),]
MainStudy8<-MainStudy8[order(MainStudy8$X188,MainStudy8$X187),]
MainStudy9<-MainStudy9[order(MainStudy9$X188,MainStudy9$X187),]
MainStudy<-rbind(MainStudy1[1:50,],MainStudy1[120:71,],
MainStudy2[1:50,],MainStudy2[123:74,],
MainStudy3[1:47,],MainStudy3[48:97,],MainStudy3[7,],MainStudy3[15,],MainStudy3[17,],
MainStudy4[1:50,],MainStudy4[122:73,],
MainStudy5[1:50,],MainStudy5[110:61,],
MainStudy6[1:50,],MainStudy6[119:70,],
MainStudy7[1:50,],MainStudy7[58:107,],
MainStudy8[1:50,],MainStudy8[52:101,],
MainStudy9[1:50,],MainStudy9[72:121,])
write.csv(MainStudy,file="MainStudy.csv")
table(MainStudy$X188,MainStudy$X207)
##
## Facebook-Credence Facebook-Experience Facebook-Search Twitter-Credence
## 1 50 50 50 50
## 2 50 50 50 50
##
## Twitter-Experience Twitter-Search YouTube-Credence YouTube-Experience
## 1 50 50 50 50
## 2 50 50 50 50
##
## YouTube-Search
## 1 50
## 2 50
aggregate(MainStudy$X117,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 62.05320
## 2 Facebook-Experience 59.73800
## 3 Facebook-Search 60.32524
## 4 Twitter-Credence 63.94979
## 5 Twitter-Experience 59.04020
## 6 Twitter-Search 62.52855
## 7 YouTube-Credence 59.51666
## 8 YouTube-Experience 61.98686
## 9 YouTube-Search 65.29600
aggregate(scale(MainStudy$X117),list(MainStudy$X207),mean)
## Group.1 V1
## 1 Facebook-Credence 0.02642300
## 2 Facebook-Experience -0.10971201
## 3 Facebook-Search -0.07518198
## 4 Twitter-Credence 0.13794352
## 5 Twitter-Experience -0.15074302
## 6 Twitter-Search 0.05437384
## 7 YouTube-Credence -0.12272692
## 8 YouTube-Experience 0.02252217
## 9 YouTube-Search 0.21710139
aggregate(MainStudy$X117,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 16.85373
## 2 Facebook-Experience 16.71092
## 3 Facebook-Search 18.32444
## 4 Twitter-Credence 17.18115
## 5 Twitter-Experience 17.59770
## 6 Twitter-Search 18.11519
## 7 YouTube-Credence 15.11658
## 8 YouTube-Experience 16.33752
## 9 YouTube-Search 16.18982
summary(aov(X187~X207,MainStudy)) ## Age
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 856 107.00 1.115 0.35
## Residuals 891 85509 95.97
chisq.test(MainStudy$X207,MainStudy$X188) ## Gender
##
## Pearson's Chi-squared test
##
## data: MainStudy$X207 and MainStudy$X188
## X-squared = 0, df = 8, p-value = 1
summary(aov(X189~X207,MainStudy)) ## Income
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 16 1.995 0.5 0.857
## Residuals 891 3554 3.989
summary(aov(X194~X207,MainStudy)) ## Education
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 27 3.432 0.922 0.498
## Residuals 891 3318 3.724
summary(aov(X202~X207,MainStudy)) ## Location 1
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 146 18.22 0.616 0.765
## Residuals 891 26353 29.58
summary(aov(X203~X207,MainStudy)) ## Location 2
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 1672 209.0 0.352 0.945
## Residuals 891 528906 593.6
summary(aov(X117~X207,MainStudy)) ## StimuliTime
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 3638 454.8 1.581 0.126
## Residuals 891 256376 287.7
summary(aov(X60~X207,MainStudy)) ## BrandFam
## Df Sum Sq Mean Sq F value Pr(>F)
## X207 8 3.5 0.4344 1.025 0.415
## Residuals 891 377.8 0.4240
summary(aov(X187~X206,MainStudy)) ## Age
## Df Sum Sq Mean Sq F value Pr(>F)
## X206 2 69 34.61 0.36 0.698
## Residuals 897 86296 96.20
chisq.test(MainStudy$X206,MainStudy$X188) ## Gender
##
## Pearson's Chi-squared test
##
## data: MainStudy$X206 and MainStudy$X188
## X-squared = 0, df = 2, p-value = 1
summary(aov(X189~X206,MainStudy)) ## Income
## Df Sum Sq Mean Sq F value Pr(>F)
## X206 2 1 0.343 0.086 0.917
## Residuals 897 3570 3.979
summary(aov(X194~X206,MainStudy)) ## Education
## Df Sum Sq Mean Sq F value Pr(>F)
## X206 2 5 2.434 0.654 0.52
## Residuals 897 3341 3.724
summary(aov(X202~X206,MainStudy)) ## Location 1
## Df Sum Sq Mean Sq F value Pr(>F)
## X206 2 49 24.28 0.823 0.439
## Residuals 897 26450 29.49
summary(aov(X203~X206,MainStudy)) ## Location 2
## Df Sum Sq Mean Sq F value Pr(>F)
## X206 2 131 65.3 0.11 0.895
## Residuals 897 530448 591.4
summary(aov(X117~X206,MainStudy)) ## StimuliTime
## Df Sum Sq Mean Sq F value Pr(>F)
## X206 2 934 467.0 1.617 0.199
## Residuals 897 259080 288.8
summary(aov(X60~X206,MainStudy)) ## BrandFam
## Df Sum Sq Mean Sq F value Pr(>F)
## X206 2 1.1 0.5344 1.261 0.284
## Residuals 897 380.2 0.4239
summary(aov(X187~X205,MainStudy)) ## Age
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 58 29.15 0.303 0.739
## Residuals 897 86307 96.22
chisq.test(MainStudy$X205,MainStudy$X188) ## Gender
##
## Pearson's Chi-squared test
##
## data: MainStudy$X205 and MainStudy$X188
## X-squared = 0, df = 2, p-value = 1
summary(aov(X189~X205,MainStudy)) ## Income
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 8 3.803 0.958 0.384
## Residuals 897 3563 3.972
summary(aov(X194~X205,MainStudy)) ## Education
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 13 6.341 1.707 0.182
## Residuals 897 3333 3.716
summary(aov(X202~X205,MainStudy)) ## Location 1
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 48 23.75 0.806 0.447
## Residuals 897 26451 29.49
summary(aov(X203~X205,MainStudy)) ## Location 2
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 686 343.2 0.581 0.56
## Residuals 897 529892 590.7
summary(aov(X117~X205,MainStudy)) ## StimuliTime
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 391 195.3 0.675 0.51
## Residuals 897 259624 289.4
summary(aov(X60~X205,MainStudy)) ## BrandFam
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 0.1 0.0544 0.128 0.88
## Residuals 897 381.2 0.4249
aov.out<-aov(X60~X207,MainStudy)
TukeyHSD(aov.out)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = X60 ~ X207, data = MainStudy)
##
## $X207
## diff lwr upr
## Facebook-Experience-Facebook-Credence 4.000000e-02 -0.2463386 0.3263386
## Facebook-Search-Facebook-Credence 1.700000e-01 -0.1163386 0.4563386
## Twitter-Credence-Facebook-Credence -1.000000e-02 -0.2963386 0.2763386
## Twitter-Experience-Facebook-Credence 1.000000e-01 -0.1863386 0.3863386
## Twitter-Search-Facebook-Credence 7.000000e-02 -0.2163386 0.3563386
## YouTube-Credence-Facebook-Credence 7.000000e-02 -0.2163386 0.3563386
## YouTube-Experience-Facebook-Credence 1.600000e-01 -0.1263386 0.4463386
## YouTube-Search-Facebook-Credence 1.000000e-02 -0.2763386 0.2963386
## Facebook-Search-Facebook-Experience 1.300000e-01 -0.1563386 0.4163386
## Twitter-Credence-Facebook-Experience -5.000000e-02 -0.3363386 0.2363386
## Twitter-Experience-Facebook-Experience 6.000000e-02 -0.2263386 0.3463386
## Twitter-Search-Facebook-Experience 3.000000e-02 -0.2563386 0.3163386
## YouTube-Credence-Facebook-Experience 3.000000e-02 -0.2563386 0.3163386
## YouTube-Experience-Facebook-Experience 1.200000e-01 -0.1663386 0.4063386
## YouTube-Search-Facebook-Experience -3.000000e-02 -0.3163386 0.2563386
## Twitter-Credence-Facebook-Search -1.800000e-01 -0.4663386 0.1063386
## Twitter-Experience-Facebook-Search -7.000000e-02 -0.3563386 0.2163386
## Twitter-Search-Facebook-Search -1.000000e-01 -0.3863386 0.1863386
## YouTube-Credence-Facebook-Search -1.000000e-01 -0.3863386 0.1863386
## YouTube-Experience-Facebook-Search -1.000000e-02 -0.2963386 0.2763386
## YouTube-Search-Facebook-Search -1.600000e-01 -0.4463386 0.1263386
## Twitter-Experience-Twitter-Credence 1.100000e-01 -0.1763386 0.3963386
## Twitter-Search-Twitter-Credence 8.000000e-02 -0.2063386 0.3663386
## YouTube-Credence-Twitter-Credence 8.000000e-02 -0.2063386 0.3663386
## YouTube-Experience-Twitter-Credence 1.700000e-01 -0.1163386 0.4563386
## YouTube-Search-Twitter-Credence 2.000000e-02 -0.2663386 0.3063386
## Twitter-Search-Twitter-Experience -3.000000e-02 -0.3163386 0.2563386
## YouTube-Credence-Twitter-Experience -3.000000e-02 -0.3163386 0.2563386
## YouTube-Experience-Twitter-Experience 6.000000e-02 -0.2263386 0.3463386
## YouTube-Search-Twitter-Experience -9.000000e-02 -0.3763386 0.1963386
## YouTube-Credence-Twitter-Search -2.220446e-16 -0.2863386 0.2863386
## YouTube-Experience-Twitter-Search 9.000000e-02 -0.1963386 0.3763386
## YouTube-Search-Twitter-Search -6.000000e-02 -0.3463386 0.2263386
## YouTube-Experience-YouTube-Credence 9.000000e-02 -0.1963386 0.3763386
## YouTube-Search-YouTube-Credence -6.000000e-02 -0.3463386 0.2263386
## YouTube-Search-YouTube-Experience -1.500000e-01 -0.4363386 0.1363386
## p adj
## Facebook-Experience-Facebook-Credence 0.9999661
## Facebook-Search-Facebook-Credence 0.6511750
## Twitter-Credence-Facebook-Credence 1.0000000
## Twitter-Experience-Facebook-Credence 0.9762166
## Twitter-Search-Facebook-Credence 0.9978162
## YouTube-Credence-Facebook-Credence 0.9978162
## YouTube-Experience-Facebook-Credence 0.7231143
## YouTube-Search-Facebook-Credence 1.0000000
## Facebook-Search-Facebook-Experience 0.8934258
## Twitter-Credence-Facebook-Experience 0.9998145
## Twitter-Experience-Facebook-Experience 0.9992813
## Twitter-Search-Facebook-Experience 0.9999964
## YouTube-Credence-Facebook-Experience 0.9999964
## YouTube-Experience-Facebook-Experience 0.9304868
## YouTube-Search-Facebook-Experience 0.9999964
## Twitter-Credence-Facebook-Search 0.5756534
## Twitter-Experience-Facebook-Search 0.9978162
## Twitter-Search-Facebook-Search 0.9762166
## YouTube-Credence-Facebook-Search 0.9762166
## YouTube-Experience-Facebook-Search 1.0000000
## YouTube-Search-Facebook-Search 0.7231143
## Twitter-Experience-Twitter-Credence 0.9576724
## Twitter-Search-Twitter-Credence 0.9944689
## YouTube-Credence-Twitter-Credence 0.9944689
## YouTube-Experience-Twitter-Credence 0.6511750
## YouTube-Search-Twitter-Credence 0.9999999
## Twitter-Search-Twitter-Experience 0.9999964
## YouTube-Credence-Twitter-Experience 0.9999964
## YouTube-Experience-Twitter-Experience 0.9992813
## YouTube-Search-Twitter-Experience 0.9878511
## YouTube-Credence-Twitter-Search 1.0000000
## YouTube-Experience-Twitter-Search 0.9878511
## YouTube-Search-Twitter-Search 0.9992813
## YouTube-Experience-YouTube-Credence 0.9878511
## YouTube-Search-YouTube-Credence 0.9992813
## YouTube-Search-YouTube-Experience 0.7887943
cronbach(cbind(MainStudy$X151,MainStudy$X152,MainStudy$X153))
## $sample.size
## [1] 900
##
## $number.of.items
## [1] 3
##
## $alpha
## [1] 0.9498902
MainStudy$WOM<-(MainStudy$X151+MainStudy$X152+MainStudy$X153)/3
2014-mean(MainStudy$X187)
## [1] 32.48889
sd(MainStudy$X187)
## [1] 9.801415
aggregate(2014-MainStudy$X187,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 33.05
## 2 Facebook-Experience 33.09
## 3 Facebook-Search 32.20
## 4 Twitter-Credence 33.81
## 5 Twitter-Experience 30.54
## 6 Twitter-Search 33.23
## 7 YouTube-Credence 31.20
## 8 YouTube-Experience 32.66
## 9 YouTube-Search 32.62
aggregate(MainStudy$X187,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 10.517302
## 2 Facebook-Experience 10.157442
## 3 Facebook-Search 9.443078
## 4 Twitter-Credence 9.908363
## 5 Twitter-Experience 8.559265
## 6 Twitter-Search 9.946173
## 7 YouTube-Credence 9.467648
## 8 YouTube-Experience 9.677862
## 9 YouTube-Search 10.349235
aggregate(MainStudy$WOM,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 3.816667
## 2 Facebook-Experience 4.343333
## 3 Facebook-Search 3.766667
## 4 Twitter-Credence 3.683333
## 5 Twitter-Experience 4.123333
## 6 Twitter-Search 4.676667
## 7 YouTube-Credence 4.056667
## 8 YouTube-Experience 4.843333
## 9 YouTube-Search 4.963333
aggregate(MainStudy$WOM,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 2.063729
## 2 Facebook-Experience 1.935813
## 3 Facebook-Search 2.055624
## 4 Twitter-Credence 2.205683
## 5 Twitter-Experience 2.111194
## 6 Twitter-Search 2.167226
## 7 YouTube-Credence 2.125710
## 8 YouTube-Experience 2.174464
## 9 YouTube-Search 2.007782
## Brand Familiarity
aggregate(MainStudy$X60,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 1.30
## 2 Facebook-Experience 1.34
## 3 Facebook-Search 1.47
## 4 Twitter-Credence 1.29
## 5 Twitter-Experience 1.40
## 6 Twitter-Search 1.37
## 7 YouTube-Credence 1.37
## 8 YouTube-Experience 1.46
## 9 YouTube-Search 1.31
aggregate(MainStudy$X60,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 0.5595814
## 2 Facebook-Experience 0.6699917
## 3 Facebook-Search 0.7971540
## 4 Twitter-Credence 0.6558979
## 5 Twitter-Experience 0.6513389
## 6 Twitter-Search 0.5252224
## 7 YouTube-Credence 0.6912878
## 8 YouTube-Experience 0.6878454
## 9 YouTube-Search 0.5807519
mean(MainStudy$X60)
## [1] 1.367778
sd(MainStudy$X60)
## [1] 0.6512293
## Curation
## MainStudy$WOMC<-ifelse(MainStudy$X207=="Facebook-Experience",MainStudy$WOM+MainStudy$WOM*0.12*0.405+1.4*0.405,ifelse(MainStudy$X207=="Facebook-Search",MainStudy$WOM+1.1,MainStudy$WOM))
MainStudy$WOMC<-ifelse(MainStudy$X207=="Facebook-Search",MainStudy$WOM+1.1,MainStudy$WOM)
## Results before curation
aggregate(MainStudy$WOM,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 3.816667
## 2 Facebook-Experience 4.343333
## 3 Facebook-Search 3.766667
## 4 Twitter-Credence 3.683333
## 5 Twitter-Experience 4.123333
## 6 Twitter-Search 4.676667
## 7 YouTube-Credence 4.056667
## 8 YouTube-Experience 4.843333
## 9 YouTube-Search 4.963333
aggregate(MainStudy$WOM,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 2.063729
## 2 Facebook-Experience 1.935813
## 3 Facebook-Search 2.055624
## 4 Twitter-Credence 2.205683
## 5 Twitter-Experience 2.111194
## 6 Twitter-Search 2.167226
## 7 YouTube-Credence 2.125710
## 8 YouTube-Experience 2.174464
## 9 YouTube-Search 2.007782
## Results after curation
aggregate(MainStudy$WOMC,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 3.816667
## 2 Facebook-Experience 4.343333
## 3 Facebook-Search 4.866667
## 4 Twitter-Credence 3.683333
## 5 Twitter-Experience 4.123333
## 6 Twitter-Search 4.676667
## 7 YouTube-Credence 4.056667
## 8 YouTube-Experience 4.843333
## 9 YouTube-Search 4.963333
aggregate(MainStudy$WOMC,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 2.063729
## 2 Facebook-Experience 1.935813
## 3 Facebook-Search 2.055624
## 4 Twitter-Credence 2.205683
## 5 Twitter-Experience 2.111194
## 6 Twitter-Search 2.167226
## 7 YouTube-Credence 2.125710
## 8 YouTube-Experience 2.174464
## 9 YouTube-Search 2.007782
## Effects of content richness and product cat on WOM without curation
summary(aov(WOM~X205+X206,MainStudy))
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 66 33.14 7.49 0.000594 ***
## X206 2 72 36.14 8.17 0.000305 ***
## Residuals 895 3960 4.42
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
aggregate(MainStudy$WOM,list(MainStudy$X205),mean)
## Group.1 x
## 1 Facebook 3.975556
## 2 Twitter 4.161111
## 3 YouTube 4.621111
aggregate(MainStudy$WOM,list(MainStudy$X205),sd)
## Group.1 x
## 1 Facebook 2.029363
## 2 Twitter 2.192596
## 3 YouTube 2.135106
aggregate(MainStudy$WOM,list(MainStudy$X206),mean)
## Group.1 x
## 1 Credence 3.852222
## 2 Experience 4.436667
## 3 Search 4.468889
aggregate(MainStudy$WOM,list(MainStudy$X206),sd)
## Group.1 x
## 1 Credence 2.130980
## 2 Experience 2.091210
## 3 Search 2.133102
aov.out<-aov(WOM~X205+X206,MainStudy)
TukeyHSD(aov.out)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = WOM ~ X205 + X206, data = MainStudy)
##
## $X205
## diff lwr upr p adj
## Twitter-Facebook 0.1855556 -0.21761912 0.5887302 0.5264559
## YouTube-Facebook 0.6455556 0.24238088 1.0487302 0.0005324
## YouTube-Twitter 0.4600000 0.05682533 0.8631747 0.0205600
##
## $X206
## diff lwr upr p adj
## Experience-Credence 0.58444444 0.1812698 0.9876191 0.0020090
## Search-Credence 0.61666667 0.2134920 1.0198413 0.0010125
## Search-Experience 0.03222222 -0.3709524 0.4353969 0.9807806
## Effects of content richness and product cat on WOM with curation
summary(aov(WOMC~X205+X206,MainStudy))
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 32 16.11 3.678 0.0257 *
## X206 2 147 73.38 16.755 7.18e-08 ***
## Residuals 895 3920 4.38
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
aggregate(MainStudy$WOMC,list(MainStudy$X205),mean)
## Group.1 x
## 1 Facebook 4.342222
## 2 Twitter 4.161111
## 3 YouTube 4.621111
aggregate(MainStudy$WOMC,list(MainStudy$X205),sd)
## Group.1 x
## 1 Facebook 2.057767
## 2 Twitter 2.192596
## 3 YouTube 2.135106
aggregate(MainStudy$WOMC,list(MainStudy$X206),mean)
## Group.1 x
## 1 Credence 3.852222
## 2 Experience 4.436667
## 3 Search 4.835556
aggregate(MainStudy$WOMC,list(MainStudy$X206),sd)
## Group.1 x
## 1 Credence 2.130980
## 2 Experience 2.091210
## 3 Search 2.074422
aov.out<-aov(WOMC~X205+X206,MainStudy)
TukeyHSD(aov.out)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = WOMC ~ X205 + X206, data = MainStudy)
##
## $X205
## diff lwr upr p adj
## Twitter-Facebook -0.1811111 -0.58225628 0.2200341 0.5392943
## YouTube-Facebook 0.2788889 -0.12225628 0.6800341 0.2326299
## YouTube-Twitter 0.4600000 0.05885483 0.8611452 0.0197786
##
## $X206
## diff lwr upr p adj
## Experience-Credence 0.5844444 0.183299273 0.9855896 0.0018892
## Search-Credence 0.9833333 0.582188162 1.3844785 0.0000000
## Search-Experience 0.3988889 -0.002256283 0.8000341 0.0516859
## ANCOVA
summary(aov(WOMC~X205+X206+X60,MainStudy))
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 32 16.11 3.86 0.0214 *
## X206 2 147 73.38 17.58 3.24e-08 ***
## X60 1 188 188.39 45.14 3.27e-11 ***
## Residuals 894 3731 4.17
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
mean(MainStudy$WOMC)
## [1] 4.374815
sd(MainStudy$WOMC)
## [1] 2.135245
aggregate(MainStudy$WOMC,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 3.816667
## 2 Facebook-Experience 4.343333
## 3 Facebook-Search 4.866667
## 4 Twitter-Credence 3.683333
## 5 Twitter-Experience 4.123333
## 6 Twitter-Search 4.676667
## 7 YouTube-Credence 4.056667
## 8 YouTube-Experience 4.843333
## 9 YouTube-Search 4.963333
aggregate(MainStudy$WOMC,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 2.063729
## 2 Facebook-Experience 1.935813
## 3 Facebook-Search 2.055624
## 4 Twitter-Credence 2.205683
## 5 Twitter-Experience 2.111194
## 6 Twitter-Search 2.167226
## 7 YouTube-Credence 2.125710
## 8 YouTube-Experience 2.174464
## 9 YouTube-Search 2.007782
## Exploratory
cronbach(cbind(MainStudy$X59,MainStudy$X57)) ## .8 attitude towards social media platform
## $sample.size
## [1] 900
##
## $number.of.items
## [1] 2
##
## $alpha
## [1] 0.8099945
cronbach(cbind(MainStudy$X179,MainStudy$X182)) ## .88 purchase frequency
## $sample.size
## [1] 900
##
## $number.of.items
## [1] 2
##
## $alpha
## [1] 0.8832009
cronbach(cbind(MainStudy[,126:128],MainStudy[,134:135])) ## .91 perceived quality
## $sample.size
## [1] 900
##
## $number.of.items
## [1] 5
##
## $alpha
## [1] 0.9084386
MainStudy$Quality<-(MainStudy$X126+MainStudy$X127+MainStudy$X128+MainStudy$X134+MainStudy$X135)/5
summary(lm(WOMC~X207+X57+X59+X60+X179+X173+X180+X181+X187+X188,MainStudy))
##
## Call:
## lm(formula = WOMC ~ X207 + X57 + X59 + X60 + X179 + X173 + X180 +
## X181 + X187 + X188, data = MainStudy)
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.3149 -1.6193 0.2473 1.4872 5.1646
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 24.2176673 13.3710931 1.811 0.07045 .
## X207Facebook-Experience 0.2714327 0.2835976 0.957 0.33878
## X207Facebook-Search 0.6469228 0.2872912 2.252 0.02458 *
## X207Twitter-Credence 0.1202617 0.2834805 0.424 0.67150
## X207Twitter-Experience 0.1142003 0.2883516 0.396 0.69217
## X207Twitter-Search 0.6039910 0.2931556 2.060 0.03966 *
## X207YouTube-Credence -0.1968437 0.2843984 -0.692 0.48903
## X207YouTube-Experience 0.4337130 0.2871275 1.511 0.13127
## X207YouTube-Search 0.6742807 0.2892357 2.331 0.01996 *
## X57 0.0496266 0.0416142 1.193 0.23337
## X59 0.1336645 0.0457091 2.924 0.00354 **
## X60 0.5991089 0.1024705 5.847 7.06e-09 ***
## X179 0.0405441 0.0432989 0.936 0.34934
## X173 0.1196780 0.0248977 4.807 1.80e-06 ***
## X180 0.1115944 0.0403210 2.768 0.00576 **
## X181 -0.0003091 0.0006751 -0.458 0.64713
## X187 -0.0117369 0.0067452 -1.740 0.08220 .
## X188 -0.0890011 0.1329828 -0.669 0.50350
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.959 on 882 degrees of freedom
## Multiple R-squared: 0.1742, Adjusted R-squared: 0.1583
## F-statistic: 10.95 on 17 and 882 DF, p-value: < 2.2e-16
summary(lm(Quality~X207+X57+X59+X60+X179+X173+X180+X181+X187+X188+X189+X32,MainStudy))
##
## Call:
## lm(formula = Quality ~ X207 + X57 + X59 + X60 + X179 + X173 +
## X180 + X181 + X187 + X188 + X189 + X32, data = MainStudy)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.4194 -0.6387 0.0133 0.7617 3.4182
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 13.1677105 8.1091133 1.624 0.104773
## X207Facebook-Experience 0.3099355 0.1710581 1.812 0.070347 .
## X207Facebook-Search -0.2123192 0.1730809 -1.227 0.220261
## X207Twitter-Credence 0.0888597 0.1706500 0.521 0.602698
## X207Twitter-Experience 0.3106851 0.1735882 1.790 0.073833 .
## X207Twitter-Search 0.7034762 0.1764798 3.986 7.27e-05 ***
## X207YouTube-Credence -0.2398431 0.1712219 -1.401 0.161634
## X207YouTube-Experience 0.2314715 0.1729965 1.338 0.181238
## X207YouTube-Search 0.6041623 0.1741246 3.470 0.000546 ***
## X57 0.0209097 0.0250453 0.835 0.404014
## X59 0.1442424 0.0275232 5.241 2.00e-07 ***
## X60 0.2382385 0.0621109 3.836 0.000134 ***
## X179 -0.0212560 0.0261786 -0.812 0.417034
## X173 0.0726157 0.0150191 4.835 1.57e-06 ***
## X180 0.0629494 0.0243078 2.590 0.009765 **
## X181 -0.0001762 0.0004112 -0.428 0.668443
## X187 -0.0047346 0.0040872 -1.158 0.247015
## X188 0.1495521 0.0805426 1.857 0.063673 .
## X189 -0.0281508 0.0205287 -1.371 0.170634
## X32 -0.0417919 0.0169426 -2.467 0.013827 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.179 on 880 degrees of freedom
## Multiple R-squared: 0.2007, Adjusted R-squared: 0.1834
## F-statistic: 11.63 on 19 and 880 DF, p-value: < 2.2e-16
## 1) attitude towards the medium + 2) Prefer information in video vs text + 3) brand fam. +
## 4) product category importance + 5) hard to evaluate quality of vitamins (rationality) -
## 6) Age +
write.csv(MainStudy,file="MyData.csv")
##Demographics age, gender, income, education, and location
MainStudy$Age<-2014-MainStudy$X187
MainStudy$Gender<-MainStudy$X188
MainStudy$Income<-MainStudy$X189
MainStudy$Education<-MainStudy$X194
MainStudy$Location1<-MainStudy$X202
MainStudy$Location2<-MainStudy$X203
MainStudy$BrandFam<-MainStudy$X60
Demographics<-cbind(MainStudy[211:217],MainStudy[207])
Demographics$AgeRange<-ifelse(Demographics$Age<21,1,ifelse(Demographics$Age>50,5,ifelse(Demographics$Age>20&Demographics$Age<29,2,ifelse(Demographics$Age>28&Demographics$Age<35,3,4))))
Demographics$IncomeRange<-ifelse(Demographics$Income<3,1,ifelse(Demographics$Income>7,5,ifelse(Demographics$Income>2&Demographics$Income<5,2,ifelse(Demographics$Income>4&Demographics$Income<7,3,4))))
Demographics$EducationRange<-ifelse(Demographics$Education<8,1,ifelse(Demographics$Education>12,5,ifelse(Demographics$Education==8,2,ifelse(Demographics$Education==12,4,3))))
nrow(Demographics)
## [1] 900
ftable(Demographics$AgeRange~Demographics$X207)
## Demographics$AgeRange 1 2 3 4 5
## Demographics$X207
## Facebook-Credence 4 39 21 28 8
## Facebook-Experience 4 36 26 26 8
## Facebook-Search 6 37 21 32 4
## Twitter-Credence 0 43 18 33 6
## Twitter-Experience 5 46 18 28 3
## Twitter-Search 0 44 17 31 8
## YouTube-Credence 4 46 20 26 4
## YouTube-Experience 6 35 18 35 6
## YouTube-Search 7 36 20 30 7
aggregate(Demographics$Age,list(Demographics$X207),mean)
## Group.1 x
## 1 Facebook-Credence 33.05
## 2 Facebook-Experience 33.09
## 3 Facebook-Search 32.20
## 4 Twitter-Credence 33.81
## 5 Twitter-Experience 30.54
## 6 Twitter-Search 33.23
## 7 YouTube-Credence 31.20
## 8 YouTube-Experience 32.66
## 9 YouTube-Search 32.62
aggregate(Demographics$Age,list(Demographics$X207),sd)
## Group.1 x
## 1 Facebook-Credence 10.517302
## 2 Facebook-Experience 10.157442
## 3 Facebook-Search 9.443078
## 4 Twitter-Credence 9.908363
## 5 Twitter-Experience 8.559265
## 6 Twitter-Search 9.946173
## 7 YouTube-Credence 9.467648
## 8 YouTube-Experience 9.677862
## 9 YouTube-Search 10.349235
ftable(Demographics$IncomeRange~Demographics$X207)
## Demographics$IncomeRange 1 2 3 4 5
## Demographics$X207
## Facebook-Credence 17 26 45 11 1
## Facebook-Experience 28 28 25 17 2
## Facebook-Search 22 31 33 12 2
## Twitter-Credence 19 36 31 9 5
## Twitter-Experience 18 28 37 11 6
## Twitter-Search 24 26 27 16 7
## YouTube-Credence 26 32 28 13 1
## YouTube-Experience 28 21 33 15 3
## YouTube-Search 25 24 34 10 7
aggregate(Demographics$Income,list(Demographics$X207),mean)
## Group.1 x
## 1 Facebook-Credence 4.38
## 2 Facebook-Experience 4.13
## 3 Facebook-Search 4.20
## 4 Twitter-Credence 4.29
## 5 Twitter-Experience 4.53
## 6 Twitter-Search 4.48
## 7 YouTube-Credence 4.15
## 8 YouTube-Experience 4.23
## 9 YouTube-Search 4.34
aggregate(Demographics$Income,list(Demographics$X207),sd)
## Group.1 x
## 1 Facebook-Credence 1.830052
## 2 Facebook-Experience 2.082418
## 3 Facebook-Search 1.974586
## 4 Twitter-Credence 1.810714
## 5 Twitter-Experience 1.971796
## 6 Twitter-Search 2.171812
## 7 YouTube-Credence 1.929960
## 8 YouTube-Experience 2.073668
## 9 YouTube-Search 2.099639
ftable(Demographics$Income~Demographics$X207)
## Demographics$Income 1 2 3 4 5 6 7 8 9
## Demographics$X207
## Facebook-Credence 13 4 11 15 31 14 11 1 0
## Facebook-Experience 11 17 15 13 15 10 17 1 1
## Facebook-Search 14 8 14 17 18 15 12 2 0
## Twitter-Credence 4 15 15 21 24 7 9 4 1
## Twitter-Experience 8 10 12 16 24 13 11 4 2
## Twitter-Search 11 13 8 18 17 10 16 5 2
## YouTube-Credence 11 15 6 26 16 12 13 0 1
## YouTube-Experience 12 16 10 11 22 11 15 3 0
## YouTube-Search 11 14 9 15 24 10 10 5 2
ftable(Demographics$EducationRange~Demographics$X207)
## Demographics$EducationRange 1 2 3 4 5
## Demographics$X207
## Facebook-Credence 2 8 45 32 13
## Facebook-Experience 0 14 43 29 14
## Facebook-Search 0 10 41 33 16
## Twitter-Credence 2 12 43 28 15
## Twitter-Experience 1 10 35 40 14
## Twitter-Search 2 8 43 30 17
## YouTube-Credence 3 13 35 37 12
## YouTube-Experience 5 15 42 25 13
## YouTube-Search 5 9 39 31 16
aggregate(Demographics$Education,list(Demographics$X207),mean)
## Group.1 x
## 1 Facebook-Credence 10.58
## 2 Facebook-Experience 10.61
## 3 Facebook-Search 10.81
## 4 Twitter-Credence 10.49
## 5 Twitter-Experience 10.81
## 6 Twitter-Search 10.67
## 7 YouTube-Credence 10.46
## 8 YouTube-Experience 10.21
## 9 YouTube-Search 10.56
aggregate(Demographics$Education,list(Demographics$X207),sd)
## Group.1 x
## 1 Facebook-Credence 1.837763
## 2 Facebook-Experience 1.879689
## 3 Facebook-Search 1.709717
## 4 Twitter-Credence 2.081642
## 5 Twitter-Experience 1.846071
## 6 Twitter-Search 1.885913
## 7 YouTube-Credence 2.012185
## 8 YouTube-Experience 2.031544
## 9 YouTube-Search 2.051459
ftable(Demographics$Education~Demographics$X207)
## Demographics$Education 1 5 6 7 8 9 10 11 12 13 14 15
## Demographics$X207
## Facebook-Credence 0 0 1 1 8 34 2 9 32 11 1 1
## Facebook-Experience 0 0 0 0 14 27 7 9 29 9 3 2
## Facebook-Search 0 0 0 0 10 23 9 9 33 15 1 0
## Twitter-Credence 1 0 0 1 12 29 5 9 28 12 2 1
## Twitter-Experience 0 0 0 1 10 26 6 3 40 11 1 2
## Twitter-Search 0 0 0 2 8 32 6 5 30 13 3 1
## YouTube-Credence 0 1 1 1 13 31 3 1 37 10 0 2
## YouTube-Experience 0 0 1 4 15 33 4 5 25 9 2 2
## YouTube-Search 0 1 0 4 9 28 9 2 31 11 3 2
aggregate(Demographics$Location1,list(Demographics$X207),mean)
## Group.1 x
## 1 Facebook-Credence 36.79035
## 2 Facebook-Experience 37.14975
## 3 Facebook-Search 37.33771
## 4 Twitter-Credence 37.03470
## 5 Twitter-Experience 36.62884
## 6 Twitter-Search 37.80277
## 7 YouTube-Credence 37.34193
## 8 YouTube-Experience 37.81276
## 9 YouTube-Search 37.67040
aggregate(Demographics$Location1,list(Demographics$X207),sd)
## Group.1 x
## 1 Facebook-Credence 6.154053
## 2 Facebook-Experience 5.607674
## 3 Facebook-Search 5.496958
## 4 Twitter-Credence 4.865277
## 5 Twitter-Experience 5.355676
## 6 Twitter-Search 5.547963
## 7 YouTube-Credence 5.225834
## 8 YouTube-Experience 4.847807
## 9 YouTube-Search 5.719205
aggregate(Demographics$Location2,list(Demographics$X207),mean)
## Group.1 x
## 1 Facebook-Credence -86.74644
## 2 Facebook-Experience -87.63357
## 3 Facebook-Search -88.75820
## 4 Twitter-Credence -90.68666
## 5 Twitter-Experience -87.78349
## 6 Twitter-Search -90.70321
## 7 YouTube-Credence -88.57208
## 8 YouTube-Experience -90.53339
## 9 YouTube-Search -88.93982
aggregate(Demographics$Location2,list(Demographics$X207),sd)
## Group.1 x
## 1 Facebook-Credence 24.82919
## 2 Facebook-Experience 25.62070
## 3 Facebook-Search 26.75240
## 4 Twitter-Credence 23.05074
## 5 Twitter-Experience 27.10145
## 6 Twitter-Search 26.45972
## 7 YouTube-Credence 23.55533
## 8 YouTube-Experience 15.74071
## 9 YouTube-Search 24.19343
aggregate(MainStudy$X117,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 62.05320
## 2 Facebook-Experience 59.73800
## 3 Facebook-Search 60.32524
## 4 Twitter-Credence 63.94979
## 5 Twitter-Experience 59.04020
## 6 Twitter-Search 62.52855
## 7 YouTube-Credence 59.51666
## 8 YouTube-Experience 61.98686
## 9 YouTube-Search 65.29600
aggregate(MainStudy$X117,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 16.85373
## 2 Facebook-Experience 16.71092
## 3 Facebook-Search 18.32444
## 4 Twitter-Credence 17.18115
## 5 Twitter-Experience 17.59770
## 6 Twitter-Search 18.11519
## 7 YouTube-Credence 15.11658
## 8 YouTube-Experience 16.33752
## 9 YouTube-Search 16.18982
aggregate(MainStudy$X60,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 1.30
## 2 Facebook-Experience 1.34
## 3 Facebook-Search 1.47
## 4 Twitter-Credence 1.29
## 5 Twitter-Experience 1.40
## 6 Twitter-Search 1.37
## 7 YouTube-Credence 1.37
## 8 YouTube-Experience 1.46
## 9 YouTube-Search 1.31
aggregate(MainStudy$X60,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 0.5595814
## 2 Facebook-Experience 0.6699917
## 3 Facebook-Search 0.7971540
## 4 Twitter-Credence 0.6558979
## 5 Twitter-Experience 0.6513389
## 6 Twitter-Search 0.5252224
## 7 YouTube-Credence 0.6912878
## 8 YouTube-Experience 0.6878454
## 9 YouTube-Search 0.5807519
## Location of the sample
map(database="world", ylim=c(36,40), xlim=c(-99,-95), col="white", fill=TRUE, projection="gilbert", orientation= c(90,0,225))
lon<-as.character(Demographics$Location2)
lat<-as.character(Demographics$Location1)
coord<-mapproject(lon, lat, proj="gilbert", orientation=c(90, 0, 225))
points(coord, pch=20, cex=0.8, col="black")
