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
## Warning: package 'reshape2' was built under R version 3.4.3
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 3.4.3
## Loading required package: mvtnorm
## Warning: package 'mvtnorm' was built under R version 3.4.3
## 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 3.4.3
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 3.4.3
##
## 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 3.4.3
## 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 3.4.4
library(Hmisc) # correlation matrix
## Warning: package 'Hmisc' was built under R version 3.4.3
## Loading required package: Formula
## Loading required package: ggplot2
## Warning: package 'ggplot2' was built under R version 3.4.3
##
## 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, units
library(plyr) # count
##
## Attaching package: 'plyr'
## The following objects are masked from 'package:Hmisc':
##
## is.discrete, summarize
## The following object is masked from 'package:maps':
##
## ozone
## PRETEST 1
cat("\014") # cleans screen
rm(list=ls(all=TRUE)) # remove variables in working memory
setwd("C:/Users/Erik Ernesto Vazquez/Downloads/Results") # sets working directory
Pretest<-read.csv("Main_Study/Main_study__3x3_United_States.csv", skip=2, header=F) # reads raw data from Qualtrics
NamesandHeaders<-read.csv("Main_Study/Main_study__3x3_United_States.csv") # assigns headers and names to data frame
names(Pretest)<-names(NamesandHeaders)
Pretest$V6<-as.character(Pretest$V6)
Pretest<-Pretest[which(!duplicated(Pretest$V6)&Pretest$t2.frmwrk_3>0&Pretest$t12_3>0),] # This procedure displays a freq. table and a bar plot showing grouping' without IPs duplicates
framework.wide=data.frame(Pretest[1],Pretest[34:36],Pretest[596:598],Pretest[603:604])
names(framework.wide)<-c("Subject","Credence","Experience","Search","Age","Gender","Income","Education","RE")
table(framework.wide$Gender)
##
## 1 2
## 539 506
women<-subset(framework.wide,framework.wide$Gender==1)
men<-subset(framework.wide,framework.wide$Gender==2)
Pretest<-rbind(women[1:60,],men[1:60,])
table(Pretest$Gender)
##
## 1 2
## 60 60
mean(Pretest$Age)-2014
## [1] -31.25
sd(Pretest$Age)
## [1] 8.758429
aggregate(Pretest$Age,list(Pretest$Gender),mean)
## Group.1 x
## 1 1 1983.067
## 2 2 1982.433
aggregate(Pretest$Age,list(Pretest$Gender),sd)
## Group.1 x
## 1 1 8.938332
## 2 2 8.638437
women<-subset(Pretest,Pretest$Gender==1)
men<-subset(Pretest,Pretest$Gender==2)
t.test(women$Age,men$Age) ## groups are equivalent in Age
##
## Welch Two Sample t-test
##
## data: women$Age and men$Age
## t = 0.39466, df = 117.86, p-value = 0.6938
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -2.544578 3.811245
## sample estimates:
## mean of x mean of y
## 1983.067 1982.433
summary(Pretest[,2:4])
## Credence Experience Search
## Min. :1.000 Min. :1.000 Min. :1.000
## 1st Qu.:4.750 1st Qu.:3.000 1st Qu.:3.000
## Median :7.000 Median :4.000 Median :3.000
## Mean :6.125 Mean :4.675 Mean :4.117
## 3rd Qu.:8.000 3rd Qu.:6.000 3rd Qu.:6.000
## Max. :9.000 Max. :9.000 Max. :9.000
t.test(Pretest$Credence,Pretest$Experience) ## Validation of SEC levels of ease to evaluate quality
##
## Welch Two Sample t-test
##
## data: Pretest$Credence and Pretest$Experience
## t = 5.0975, df = 237.96, p-value = 7.019e-07
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.8896311 2.0103689
## sample estimates:
## mean of x mean of y
## 6.125 4.675
t.test(Pretest$Experience,Pretest$Search) ## The chance of error type 2 is very small according to Winter 2013
##
## Welch Two Sample t-test
##
## data: Pretest$Experience and Pretest$Search
## t = 2.0762, df = 235.43, p-value = 0.03896
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.02853859 1.08812808
## sample estimates:
## mean of x mean of y
## 4.675000 4.116667
framework.long<-melt(Pretest,id.vars=c("Subject","Age","Gender","Income","Education","RE"),measure.vars=c("Credence", "Experience", "Search" ),variable.name="Framework", value.name="Measurement")
aggregate(framework.long$Measurement,list(framework.long$Framework),sd)
## Group.1 x
## 1 Credence 2.217435
## 2 Experience 2.189212
## 3 Search 1.971150
## PRETEST 2
cat("\014") # cleans screen
rm(list=ls(all=TRUE)) # remove variables in working memory
setwd("C:/Users/Erik Ernesto Vazquez/Downloads/IJEC Data recollection") # sets working directory
MainStudy<-read.csv("Pretest Analysis Tie Strength and Media Richness.csv", header=T) # reads raw data from Qualtrics
MainStudy<-subset(MainStudy,MainStudy$X3<1991&MainStudy$X1_15>0&MainStudy$X2_15>0&MainStudy$X3_15>0&MainStudy$X4_15>0&MainStudy$X5_15>0)
table(MainStudy$V3)
##
## 9 10
## 25 26
MainStudyF<-subset(MainStudy,MainStudy$V3==9)
MainStudyM<-subset(MainStudy,MainStudy$V3==10)
MainStudy<-rbind(MainStudyF[1:25,],MainStudyM[1:25,])
table(MainStudy$V3)
##
## 9 10
## 25 25
mean(MainStudy$X3)-2014
## [1] -31.88
sd(MainStudy$X3)
## [1] 8.66294
##Reliability content Vividness
MainStudyMelt1<-melt(MainStudy,id.vars=c("ResponseId","X3","V3","X1_1","X1_2","X1_3",
"X1_4","X1_5","X1_6","X1_7",
"X1_15"),
measure.vars=c("X1_1","X1_2","X1_3",
"X1_4","X1_5","X1_6","X1_7",
"X1_15"),
variable.name="MediaRichness1", value.name="MRItem1")
MainStudyMelt2<-melt(MainStudy,id.vars=c("ResponseId","X3","V3","X2_1","X2_2","X2_3",
"X2_4","X2_5","X2_6","X2_7",
"X2_15"),
measure.vars=c("X2_1","X2_2","X2_3",
"X2_4","X2_5","X2_6","X2_7",
"X2_15"),
variable.name="MediaRichness2", value.name="MRItem2")
MainStudyMelt3<-melt(MainStudy,id.vars=c("ResponseId","X3","V3","X3_1","X3_2","X3_3",
"X3_4","X3_5","X3_6","X3_7",
"X3_15"),
measure.vars=c("X3_1","X3_2","X3_3",
"X3_4","X3_5","X3_6","X3_7",
"X3_15"),
variable.name="MediaRichness3", value.name="MRItem3")
cronbach(cbind(MainStudyMelt1$MRItem1,MainStudyMelt2$MRItem2,MainStudyMelt3$MRItem3)) ## Cronabch 0.81
## $sample.size
## [1] 400
##
## $number.of.items
## [1] 3
##
## $alpha
## [1] 0.8172619
## Reliability Tie Strength
MainStudyMelt4<-melt(MainStudy,id.vars=c("ResponseId","X3","V3","X4_1","X4_2","X4_3",
"X4_4","X4_5","X4_6","X4_7",
"X4_15"),
measure.vars=c("X4_1","X4_2","X4_3",
"X4_4","X4_5","X4_6","X4_7",
"X4_15"),
variable.name="TieStr1", value.name="TieStrItem1")
MainStudyMelt5<-melt(MainStudy,id.vars=c("ResponseId","X3","V3","X5_1","X5_2","X5_3",
"X5_4","X5_5","X5_6","X5_7",
"X5_15"),
measure.vars=c("X5_1","X5_2","X5_3",
"X5_4","X5_5","X5_6","X5_7",
"X5_15"),
variable.name="TieStr2", value.name="TieStrItem2")
cronbach(cbind(MainStudyMelt4$TieStrItem1,MainStudyMelt5$TieStrItem2)) ## Cronabch 0.89
## $sample.size
## [1] 400
##
## $number.of.items
## [1] 2
##
## $alpha
## [1] 0.8946946
validity<-data.frame(cbind(MainStudyMelt1$MRItem1,MainStudyMelt2$MRItem2,MainStudyMelt3$MRItem3,MainStudyMelt4$TieStrItem1,MainStudyMelt5$TieStrItem2))
KMO(validity)
## Kaiser-Meyer-Olkin factor adequacy
## Call: KMO(r = validity)
## Overall MSA = 0.79
## MSA for each item =
## X1 X2 X3 X4 X5
## 0.88 0.82 0.83 0.75 0.73
factanal(validity,2,rotation="varimax")
##
## Call:
## factanal(x = validity, factors = 2, rotation = "varimax")
##
## Uniquenesses:
## X1 X2 X3 X4 X5
## 0.533 0.295 0.309 0.322 0.005
##
## Loadings:
## Factor1 Factor2
## X1 0.636 0.249
## X2 0.800 0.254
## X3 0.694 0.457
## X4 0.384 0.729
## X5 0.309 0.948
##
## Factor1 Factor2
## SS loadings 1.769 1.766
## Proportion Var 0.354 0.353
## Cumulative Var 0.354 0.707
##
## Test of the hypothesis that 2 factors are sufficient.
## The chi square statistic is 0.23 on 1 degree of freedom.
## The p-value is 0.633
summary(prcomp(validity)) ## Two components explain 69% of the variance
## Importance of components%s:
## PC1 PC2 PC3 PC4 PC5
## Standard deviation 4.7071 2.2192 1.57373 1.41028 1.16880
## Proportion of Variance 0.6732 0.1496 0.07525 0.06043 0.04151
## Cumulative Proportion 0.6732 0.8228 0.89807 0.95849 1.00000
screeplot(prcomp(validity),type="lines")
biplot(prcomp(validity,scale.=T),cex=0.5,xlabs=rep(".",nrow(validity)))
rcorr(as.matrix(validity))
## X1 X2 X3 X4 X5
## X1 1.00 0.57 0.56 0.43 0.43
## X2 0.57 1.00 0.67 0.50 0.49
## X3 0.56 0.67 1.00 0.59 0.65
## X4 0.43 0.50 0.59 1.00 0.81
## X5 0.43 0.49 0.65 0.81 1.00
##
## n= 400
##
##
## P
## X1 X2 X3 X4 X5
## X1 0 0 0 0
## X2 0 0 0 0
## X3 0 0 0 0
## X4 0 0 0 0
## X5 0 0 0 0
MainStudy$MRFacebook<-(MainStudy$X3_1+MainStudy$X1_1+MainStudy$X2_1)/3-38
MainStudy$MRTwitter<-(MainStudy$X3_2+MainStudy$X1_2+MainStudy$X2_2)/3-38
MainStudy$MRYouTube<-(MainStudy$X3_3+MainStudy$X1_3+MainStudy$X2_3)/3-38
MainStudy$MRInstagram<-(MainStudy$X3_4+MainStudy$X1_4+MainStudy$X2_4)/3-38
MainStudy$MRPinterest<-(MainStudy$X3_5+MainStudy$X1_5+MainStudy$X2_5)/3-38
MainStudy$MRSnapChat<-(MainStudy$X3_6+MainStudy$X1_6+MainStudy$X2_6)/3-38
MainStudy$MRLinkedIn<-(MainStudy$X3_7+MainStudy$X1_7+MainStudy$X2_7)/3-38
MainStudy$MRSecondLife<-(MainStudy$X3_15+MainStudy$X1_15+MainStudy$X2_15)/3-38
MainStudy$TSFacebook<-(MainStudy$X4_1+MainStudy$X5_1)/2-38
MainStudy$TSTwitter<-(MainStudy$X4_2+MainStudy$X5_2)/2-38
MainStudy$TSYouTube<-(MainStudy$X4_3+MainStudy$X5_3)/2-38
MainStudy$TSInstagram<-(MainStudy$X4_4+MainStudy$X5_4)/2-38
MainStudy$TSPinterest<-(MainStudy$X4_5+MainStudy$X5_5)/2-38
MainStudy$TSSnapChat<-(MainStudy$X4_6+MainStudy$X5_6)/2-38
MainStudy$TSLinkedIn<-(MainStudy$X4_7+MainStudy$X5_7)/2-38
MainStudy$TSSecondLife<-(MainStudy$X4_15+MainStudy$X5_15)/2-38
summary(MainStudy)
## StartDate EndDate Status IPAddress
## 9/28/2017 5:29: 3 9/28/2017 2:26: 2 Min. :0 37.187.147.158: 2
## 9/28/2017 1:44: 2 9/28/2017 3:43: 2 1st Qu.:0 1.22.132.15 : 1
## 9/28/2017 3:20: 2 9/28/2017 5:02: 2 Median :0 103.204.47.33 : 1
## 9/28/2017 4:51: 2 9/28/2017 5:06: 2 Mean :0 103.25.47.134 : 1
## 9/28/2017 5:52: 2 9/28/2017 5:36: 2 3rd Qu.:0 103.88.77.3 : 1
## 9/28/2017 8:30: 2 9/28/2017 1:21: 1 Max. :0 106.51.152.46 : 1
## (Other) :37 (Other) :39 (Other) :43
## Progress Duration..in.seconds. Finished RecordedDate
## Min. :100 Min. : 374.0 Min. :1 9/28/2017 2:26: 2
## 1st Qu.:100 1st Qu.: 424.2 1st Qu.:1 9/28/2017 3:43: 2
## Median :100 Median : 490.0 Median :1 9/28/2017 5:02: 2
## Mean :100 Mean : 582.4 Mean :1 9/28/2017 5:06: 2
## 3rd Qu.:100 3rd Qu.: 634.2 3rd Qu.:1 9/28/2017 5:36: 2
## Max. :100 Max. :1753.0 Max. :1 9/28/2017 1:21: 1
## (Other) :39
## ResponseId RecipientLastName RecipientFirstName
## R_10T8rIxyUdDqUvY: 1 Mode:logical Mode:logical
## R_1BoX1ncuNMXu7TD: 1 NA's:50 NA's:50
## R_1CazBZ3AMwO2Xwb: 1
## R_1f2xJMMytF6btHR: 1
## R_1hALgNb68Qa7d9i: 1
## R_1i2LOTcQ6vqrYPd: 1
## (Other) :44
## RecipientEmail ExternalReference LocationLatitude LocationLongitude
## Mode:logical Mode:logical Min. : 8.00 Min. :-122.68
## NA's:50 NA's:50 1st Qu.:13.08 1st Qu.: -71.70
## Median :15.92 Median : 77.09
## Mean :23.26 Mean : 29.41
## 3rd Qu.:39.12 3rd Qu.: 80.28
## Max. :53.75 Max. : 121.02
##
## DistributionChannel UserLanguage t0_First.Click t0_Last.Click
## anonymous:50 : 2 Min. : 0.000 Min. : 0.000
## EN:48 1st Qu.: 0.000 1st Qu.: 0.000
## Median : 0.000 Median : 0.000
## Mean : 2.289 Mean : 4.524
## 3rd Qu.: 0.000 3rd Qu.: 0.000
## Max. :36.752 Max. :59.745
##
## t0_Page.Submit t0_Click.Count X0B_Browser X0B_Version
## Min. : 15.97 Min. :0.00 Chrome :35 61.0.3163.100:20
## 1st Qu.: 17.50 1st Qu.:0.00 Firefox :12 55 : 8
## Median : 20.49 Median :0.00 Safari : 2 60.0.3112.113: 3
## Mean : 43.52 Mean :0.66 Safari iPad: 1 49.0.2623.112: 2
## 3rd Qu.: 27.98 3rd Qu.:0.00 Chrome iPad: 0 5.1.7 : 2
## Max. :416.32 Max. :9.00 Edge : 0 60.0.3112.90 : 2
## (Other) : 0 (Other) :13
## X0B_Operating.System X0B_Resolution V1 V2
## Windows NT 6.1 :19 1366x768 :19 Min. :10 Min. :9
## Windows NT 10.0:11 1280x800 : 6 1st Qu.:10 1st Qu.:9
## Android 6.0.1 : 3 1280x1024: 5 Median :10 Median :9
## Macintosh : 3 1024x768 : 3 Mean :10 Mean :9
## Windows NT 5.1 : 3 360x640 : 3 3rd Qu.:10 3rd Qu.:9
## Windows NT 6.3 : 3 1440x900 : 2 Max. :10 Max. :9
## (Other) : 8 (Other) :12
## V3 V4 tV_First.Click tV_Last.Click
## Min. : 9.0 4,5,6,7,8,9,10 :13 Min. : 1.250 Min. : 9.257
## 1st Qu.: 9.0 4,5,6,7,8,10 : 6 1st Qu.: 2.507 1st Qu.:17.028
## Median : 9.5 4,5,6 : 4 Median : 3.454 Median :21.335
## Mean : 9.5 4,5,6,7,8,9,10,11: 3 Mean : 4.995 Mean :23.431
## 3rd Qu.:10.0 4 : 2 3rd Qu.: 4.787 3rd Qu.:27.987
## Max. :10.0 4,5,6,10 : 2 Max. :52.247 Max. :69.337
## (Other) :20
## tV_Page.Submit tV_Click.Count X1_1 X1_2
## Min. :10.17 Min. : 4.00 Min. :39.00 Min. :39.00
## 1st Qu.:18.41 1st Qu.: 8.00 1st Qu.:44.00 1st Qu.:43.00
## Median :23.34 Median :10.00 Median :45.00 Median :45.00
## Mean :25.40 Mean :11.18 Mean :44.76 Mean :44.12
## 3rd Qu.:30.00 3rd Qu.:11.00 3rd Qu.:46.00 3rd Qu.:45.75
## Max. :75.09 Max. :32.00 Max. :47.00 Max. :47.00
##
## X1_3 X1_4 X1_5 X1_6
## Min. :40.00 Min. :39.00 Min. :39.0 Min. :39.00
## 1st Qu.:44.00 1st Qu.:43.00 1st Qu.:43.0 1st Qu.:41.25
## Median :45.00 Median :45.00 Median :44.5 Median :43.00
## Mean :45.08 Mean :44.42 Mean :44.1 Mean :43.16
## 3rd Qu.:47.00 3rd Qu.:46.00 3rd Qu.:46.0 3rd Qu.:44.75
## Max. :47.00 Max. :47.00 Max. :47.0 Max. :47.00
##
## X1_7 X1_15 t1_First.Click t1_Last.Click
## Min. :39.00 Min. :39.00 Min. : 0.836 Min. : 4.256
## 1st Qu.:41.00 1st Qu.:41.00 1st Qu.: 4.304 1st Qu.: 21.102
## Median :43.00 Median :43.00 Median : 6.212 Median : 39.944
## Mean :43.14 Mean :42.74 Mean : 7.830 Mean : 43.273
## 3rd Qu.:45.00 3rd Qu.:44.75 3rd Qu.: 8.868 3rd Qu.: 60.008
## Max. :47.00 Max. :47.00 Max. :33.204 Max. :122.232
##
## t1_Page.Submit t1_Click.Count X2_1 X2_2
## Min. : 61.11 Min. : 8.00 Min. :41.00 Min. :39.00
## 1st Qu.: 62.59 1st Qu.: 9.00 1st Qu.:44.00 1st Qu.:42.25
## Median : 65.43 Median :11.50 Median :45.00 Median :44.50
## Mean :103.60 Mean :13.38 Mean :44.82 Mean :43.82
## 3rd Qu.: 84.50 3rd Qu.:17.00 3rd Qu.:46.00 3rd Qu.:45.00
## Max. :981.92 Max. :31.00 Max. :47.00 Max. :47.00
##
## X2_3 X2_4 X2_5 X2_6
## Min. :41.0 Min. :39.00 Min. :39.00 Min. :39.00
## 1st Qu.:45.0 1st Qu.:42.25 1st Qu.:43.00 1st Qu.:40.25
## Median :46.0 Median :44.00 Median :44.50 Median :43.00
## Mean :45.5 Mean :43.80 Mean :43.88 Mean :42.30
## 3rd Qu.:47.0 3rd Qu.:45.00 3rd Qu.:45.75 3rd Qu.:44.00
## Max. :47.0 Max. :47.00 Max. :47.00 Max. :47.00
##
## X2_7 X2_15 Q774_First.Click Q774_Last.Click
## Min. :39.00 Min. :39.00 Min. : 0.692 Min. : 3.404
## 1st Qu.:41.25 1st Qu.:40.25 1st Qu.: 4.062 1st Qu.: 18.695
## Median :43.00 Median :43.00 Median : 7.470 Median : 33.911
## Mean :42.96 Mean :42.34 Mean : 9.889 Mean : 50.221
## 3rd Qu.:45.00 3rd Qu.:44.00 3rd Qu.:12.105 3rd Qu.: 60.494
## Max. :47.00 Max. :47.00 Max. :64.105 Max. :555.504
##
## Q774_Page.Submit Q774_Click.Count X3_1 X3_2
## Min. : 61.16 Min. : 8.00 Min. :39.00 Min. :39.00
## 1st Qu.: 62.66 1st Qu.: 8.00 1st Qu.:43.00 1st Qu.:40.25
## Median : 69.47 Median :11.00 Median :45.00 Median :43.00
## Mean :102.26 Mean :13.22 Mean :44.34 Mean :42.70
## 3rd Qu.:102.92 3rd Qu.:14.75 3rd Qu.:46.00 3rd Qu.:45.00
## Max. :555.58 Max. :50.00 Max. :47.00 Max. :47.00
##
## X3_3 X3_4 X3_5 X3_6
## Min. :39.00 Min. :39.00 Min. :39.00 Min. :39.00
## 1st Qu.:43.00 1st Qu.:41.00 1st Qu.:42.00 1st Qu.:39.00
## Median :45.00 Median :43.00 Median :43.00 Median :41.00
## Mean :44.46 Mean :42.66 Mean :43.18 Mean :41.64
## 3rd Qu.:46.00 3rd Qu.:44.75 3rd Qu.:45.00 3rd Qu.:43.75
## Max. :47.00 Max. :47.00 Max. :47.00 Max. :47.00
##
## X3_7 X3_15 Q776_First.Click Q776_Last.Click
## Min. :39.00 Min. :39.00 Min. : 0.849 Min. : 3.421
## 1st Qu.:40.25 1st Qu.:39.00 1st Qu.: 2.640 1st Qu.: 19.424
## Median :43.00 Median :39.00 Median : 5.031 Median : 30.837
## Mean :42.46 Mean :40.98 Mean : 8.607 Mean : 38.878
## 3rd Qu.:44.00 3rd Qu.:42.75 3rd Qu.: 8.897 3rd Qu.: 59.546
## Max. :47.00 Max. :46.00 Max. :72.620 Max. :116.506
##
## Q776_Page.Submit Q776_Click.Count X4_1 X4_2
## Min. : 61.13 Min. : 8.00 Min. :40.00 Min. :39.00
## 1st Qu.: 62.72 1st Qu.: 8.00 1st Qu.:44.25 1st Qu.:43.00
## Median : 71.14 Median :10.00 Median :46.00 Median :44.00
## Mean : 96.13 Mean :13.22 Mean :45.40 Mean :43.96
## 3rd Qu.: 88.52 3rd Qu.:15.75 3rd Qu.:47.00 3rd Qu.:46.00
## Max. :683.66 Max. :42.00 Max. :47.00 Max. :47.00
##
## X4_3 X4_4 X4_5 X4_6
## Min. :39.00 Min. :39.00 Min. :39.00 Min. :39.00
## 1st Qu.:41.25 1st Qu.:42.25 1st Qu.:40.00 1st Qu.:39.00
## Median :44.00 Median :44.00 Median :42.00 Median :42.00
## Mean :43.46 Mean :43.66 Mean :42.32 Mean :42.12
## 3rd Qu.:45.75 3rd Qu.:46.00 3rd Qu.:44.00 3rd Qu.:44.75
## Max. :47.00 Max. :47.00 Max. :47.00 Max. :47.00
##
## X4_7 X4_15 Q778_First.Click Q778_Last.Click
## Min. :39.00 Min. :39.0 Min. : 0.885 Min. : 4.476
## 1st Qu.:42.25 1st Qu.:39.0 1st Qu.: 4.983 1st Qu.: 27.854
## Median :44.00 Median :40.0 Median : 7.396 Median : 44.758
## Mean :43.80 Mean :41.4 Mean : 9.577 Mean : 46.593
## 3rd Qu.:45.75 3rd Qu.:44.0 3rd Qu.:10.875 3rd Qu.: 60.206
## Max. :47.00 Max. :47.0 Max. :65.250 Max. :121.689
##
## Q778_Page.Submit Q778_Click.Count X5_1 X5_2
## Min. : 61.08 Min. : 8.00 Min. :40.00 Min. :39.00
## 1st Qu.: 61.93 1st Qu.: 9.00 1st Qu.:45.00 1st Qu.:41.25
## Median : 62.95 Median :10.50 Median :46.00 Median :44.00
## Mean : 73.21 Mean :15.24 Mean :45.36 Mean :43.46
## 3rd Qu.: 76.19 3rd Qu.:16.75 3rd Qu.:47.00 3rd Qu.:46.00
## Max. :157.36 Max. :66.00 Max. :47.00 Max. :47.00
##
## X5_3 X5_4 X5_5 X5_6
## Min. :39.00 Min. :39.00 Min. :39.00 Min. :39.00
## 1st Qu.:41.00 1st Qu.:41.00 1st Qu.:39.00 1st Qu.:39.00
## Median :44.00 Median :43.00 Median :42.00 Median :41.50
## Mean :43.18 Mean :42.96 Mean :42.12 Mean :41.96
## 3rd Qu.:45.00 3rd Qu.:45.00 3rd Qu.:44.00 3rd Qu.:44.00
## Max. :47.00 Max. :47.00 Max. :47.00 Max. :47.00
##
## X5_7 X5_15 Q780_First.Click Q780_Last.Click
## Min. :39.00 Min. :39.00 Min. : 0.882 Min. : 5.832
## 1st Qu.:42.00 1st Qu.:39.00 1st Qu.: 4.543 1st Qu.: 32.826
## Median :43.00 Median :40.00 Median : 8.931 Median : 49.100
## Mean :43.34 Mean :40.94 Mean : 9.971 Mean : 52.914
## 3rd Qu.:45.75 3rd Qu.:42.75 3rd Qu.:12.747 3rd Qu.: 62.276
## Max. :47.00 Max. :47.00 Max. :40.273 Max. :196.035
##
## Q780_Page.Submit Q780_Click.Count X3 X4
## Min. : 60.93 Min. : 8.00 Min. :1949 Min. :1.00
## 1st Qu.: 62.53 1st Qu.: 9.00 1st Qu.:1979 1st Qu.:1.00
## Median : 64.97 Median :11.00 Median :1986 Median :3.50
## Mean : 86.12 Mean :15.42 Mean :1982 Mean :3.18
## 3rd Qu.: 75.39 3rd Qu.:16.50 3rd Qu.:1988 3rd Qu.:5.00
## Max. :505.77 Max. :70.00 Max. :1990 Max. :7.00
##
## t3_First.Click t3_Last.Click t3_Page.Submit t3_Click.Count
## Min. : 1.404 Min. : 4.553 Min. : 5.587 Min. : 2.00
## 1st Qu.: 2.312 1st Qu.: 7.345 1st Qu.: 8.748 1st Qu.: 2.00
## Median : 3.377 Median : 8.812 Median : 11.464 Median : 2.00
## Mean : 9.071 Mean : 16.427 Mean : 18.759 Mean : 2.94
## 3rd Qu.: 5.055 3rd Qu.: 16.309 3rd Qu.: 18.027 3rd Qu.: 3.00
## Max. :203.935 Max. :205.335 Max. :208.775 Max. :13.00
##
## X5 X6 X6_6_TEXT t4_First.Click
## Min. : 9.00 Min. :1.00 :38 Min. : 1.254
## 1st Qu.:12.00 1st Qu.:2.00 Indian : 4 1st Qu.: 2.792
## Median :12.00 Median :2.00 asian : 2 Median : 3.466
## Mean :12.06 Mean :3.22 Asian : 2 Mean : 5.989
## 3rd Qu.:13.00 3rd Qu.:4.75 Asian/Indian: 1 3rd Qu.: 4.179
## Max. :15.00 Max. :6.00 Mixed : 1 Max. :71.134
## (Other) : 2
## t4_Last.Click t4_Page.Submit t4_Click.Count mTurkCode
## Min. : 3.004 Min. : 3.834 Min. : 2.00 Min. : 0.00
## 1st Qu.: 5.530 1st Qu.: 7.016 1st Qu.: 2.00 1st Qu.:21.25
## Median : 7.705 Median : 9.520 Median : 2.50 Median :44.00
## Mean :11.190 Mean :16.241 Mean : 3.78 Mean :47.90
## 3rd Qu.:12.505 3rd Qu.:16.907 3rd Qu.: 4.00 3rd Qu.:78.50
## Max. :75.028 Max. :85.504 Max. :16.00 Max. :99.00
##
## MRFacebook MRTwitter MRYouTube MRInstagram
## Min. :2.333 Min. :1.000 Min. :3.667 Min. :2.000
## 1st Qu.:6.000 1st Qu.:4.333 1st Qu.:6.333 1st Qu.:4.333
## Median :6.667 Median :6.000 Median :7.333 Median :6.000
## Mean :6.640 Mean :5.547 Mean :7.013 Mean :5.627
## 3rd Qu.:7.667 3rd Qu.:6.667 3rd Qu.:8.000 3rd Qu.:7.000
## Max. :9.000 Max. :8.333 Max. :9.000 Max. :9.000
##
## MRPinterest MRSnapChat MRLinkedIn MRSecondLife
## Min. :1.667 Min. :1.000 Min. :1.000 Min. :1.000
## 1st Qu.:4.000 1st Qu.:2.417 1st Qu.:3.667 1st Qu.:2.417
## Median :6.000 Median :4.333 Median :5.000 Median :3.667
## Mean :5.720 Mean :4.367 Mean :4.853 Mean :4.020
## 3rd Qu.:7.333 3rd Qu.:5.833 3rd Qu.:6.333 3rd Qu.:5.500
## Max. :8.333 Max. :8.667 Max. :8.667 Max. :8.667
##
## TSFacebook TSTwitter TSYouTube TSInstagram
## Min. :2.000 Min. :1.000 Min. :1.000 Min. :1.000
## 1st Qu.:6.625 1st Qu.:4.125 1st Qu.:3.125 1st Qu.:3.625
## Median :7.750 Median :6.000 Median :5.500 Median :5.500
## Mean :7.380 Mean :5.710 Mean :5.320 Mean :5.310
## 3rd Qu.:9.000 3rd Qu.:7.500 3rd Qu.:7.500 3rd Qu.:7.000
## Max. :9.000 Max. :9.000 Max. :9.000 Max. :9.000
##
## TSPinterest TSSnapChat TSLinkedIn TSSecondLife
## Min. :1.00 Min. :1.000 Min. :1.00 Min. :1.000
## 1st Qu.:2.00 1st Qu.:1.625 1st Qu.:4.50 1st Qu.:1.000
## Median :4.00 Median :3.500 Median :5.50 Median :2.250
## Mean :4.22 Mean :4.040 Mean :5.57 Mean :3.170
## 3rd Qu.:6.00 3rd Qu.:6.500 3rd Qu.:7.50 3rd Qu.:4.875
## Max. :9.00 Max. :9.000 Max. :9.00 Max. :8.500
##
t.test(MainStudy$MRYouTube,MainStudy$MRFacebook)
##
## Welch Two Sample t-test
##
## data: MainStudy$MRYouTube and MainStudy$MRFacebook
## t = 1.2293, df = 96.807, p-value = 0.2219
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.2294196 0.9760863
## sample estimates:
## mean of x mean of y
## 7.013333 6.640000
t.test(MainStudy$MRYouTube,MainStudy$MRTwitter)
##
## Welch Two Sample t-test
##
## data: MainStudy$MRYouTube and MainStudy$MRTwitter
## t = 4.5105, df = 93.294, p-value = 1.88e-05
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.8209743 2.1123590
## sample estimates:
## mean of x mean of y
## 7.013333 5.546667
t.test(MainStudy$MRFacebook,MainStudy$MRTwitter) ## Two levels of Content Vividness
##
## Welch Two Sample t-test
##
## data: MainStudy$MRFacebook and MainStudy$MRTwitter
## t = 3.2105, df = 96.688, p-value = 0.001799
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.4174128 1.7692539
## sample estimates:
## mean of x mean of y
## 6.640000 5.546667
mean(MainStudy$MRYouTube)
## [1] 7.013333
sd(MainStudy$MRYouTube)
## [1] 1.431679
mean(MainStudy$MRFacebook)
## [1] 6.64
sd(MainStudy$MRFacebook)
## [1] 1.60051
mean(MainStudy$MRTwitter)
## [1] 5.546667
sd(MainStudy$MRTwitter)
## [1] 1.799168
t.test(MainStudy$TSYouTube,MainStudy$TSFacebook)
##
## Welch Two Sample t-test
##
## data: MainStudy$TSYouTube and MainStudy$TSFacebook
## t = -4.743, df = 83.499, p-value = 8.585e-06
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -2.923775 -1.196225
## sample estimates:
## mean of x mean of y
## 5.32 7.38
t.test(MainStudy$TSYouTube,MainStudy$TSTwitter)
##
## Welch Two Sample t-test
##
## data: MainStudy$TSYouTube and MainStudy$TSTwitter
## t = -0.78715, df = 97.235, p-value = 0.4331
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.3733189 0.5933189
## sample estimates:
## mean of x mean of y
## 5.32 5.71
t.test(MainStudy$TSFacebook,MainStudy$TSTwitter) ## Two levels of Tie Str
##
## Welch Two Sample t-test
##
## data: MainStudy$TSFacebook and MainStudy$TSTwitter
## t = 4.0882, df = 87.81, p-value = 9.599e-05
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.8581796 2.4818204
## sample estimates:
## mean of x mean of y
## 7.38 5.71
mean(MainStudy$TSYouTube)
## [1] 5.32
sd(MainStudy$TSYouTube)
## [1] 2.584806
mean(MainStudy$TSFacebook)
## [1] 7.38
sd(MainStudy$TSFacebook)
## [1] 1.658497
mean(MainStudy$TSTwitter)
## [1] 5.71
sd(MainStudy$TSTwitter)
## [1] 2.364901
MainStudyMelt1<-melt(MainStudy,id.vars=c("ResponseId","X3","V3","TSFacebook","TSTwitter","TSYouTube","TSInstagram","TSPinterest","TSSnapChat","TSLinkedIn","TSSecondLife"),measure.vars=c("TSFacebook","TSTwitter","TSYouTube","TSInstagram","TSPinterest","TSSnapChat","TSLinkedIn","TSSecondLife"),variable.name="SMP", value.name="TieStrength")
MainStudyMelt2<-melt(MainStudy,id.vars=c("ResponseId","X3","V3","MRFacebook","MRTwitter","MRYouTube","MRInstagram","MRPinterest","MRSnapChat","MRLinkedIn","MRSecondLife"),measure.vars=c("MRFacebook","MRTwitter","MRYouTube","MRInstagram","MRPinterest","MRSnapChat","MRLinkedIn","MRSecondLife"),variable.name="SMP", value.name="MediaRichness")
MainStudyMelt<-cbind(MainStudyMelt1,MainStudyMelt2)
hist(MainStudyMelt$TieStrength)
plot(density(MainStudyMelt$TieStrength))
screeplot(prcomp(cbind(MainStudyMelt$SMP,MainStudyMelt$TieStrength),type="lines"))
## Warning: In prcomp.default(cbind(MainStudyMelt$SMP, MainStudyMelt$TieStrength),
## type = "lines") :
## extra argument 'type' will be disregarded
summary(prcomp(cbind(MainStudyMelt$SMP,MainStudyMelt$TieStrength)))
## Importance of components%s:
## PC1 PC2
## Standard deviation 2.9096 1.9372
## Proportion of Variance 0.6928 0.3071
## Cumulative Proportion 0.6928 1.0000
## At least two components
mydata<-data.frame(MainStudyMelt$TieStrength)
wss <- (nrow(mydata)-1)*sum(apply(mydata,2,var))
for (i in 2:15) wss[i] <- sum(kmeans(mydata,
centers=i)$withinss)
plot(1:15, wss, type="b", xlab="Number of Clusters",
ylab="Within groups sum of squares")
wss<-wss/sum(wss)*100
for (i in 2:15)
wss[i]<-wss[i]+wss[i-1]
plot(1:15, wss, type="b", xlab="Number of Clusters",
ylab="% Var explained")
wss
## [1] 63.92178 79.98327 86.29516 89.96238 92.14484 93.93989 95.10378
## [8] 95.98143 98.11696 98.59247 98.95988 99.44536 99.67426 99.87121
## [15] 100.00000
## 3 clusters explain more than 80% of the variance
abline(v=3,lty=2)
d <- dist(mydata,method="euclidean") # distance matrix
fit <- hclust(d, method="ward")
## The "ward" method has been renamed to "ward.D"; note new "ward.D2"
plot(fit) # display dendogram all raw data
groups <- cutree(fit,k=3) # cut tree into 3 clusters
rect.hclust(fit,k=3,border="red") # draw dendogram with red borders around the 4 clusters
mydata2<-aggregate(MainStudyMelt$TieStrength,list(MainStudyMelt$SMP),mean)
rownames(mydata2)<-c("Facebook","Twitter","YouTube","Instagram","Pinterest",
"SnapChat","LinkedIn","SecondLife")
d<-dist(mydata2,method="euclidean") # distance matrix
## Warning in dist(mydata2, method = "euclidean"): NAs introduced by coercion
fit <- hclust(d, method="ward")
## The "ward" method has been renamed to "ward.D"; note new "ward.D2"
plot(fit,ylab="Tie strength") # display dendogram mean by SMP
groups <- cutree(fit,k=3) # cut tree into 4 clusters
rect.hclust(fit,k=3,border="red") # draw dendogram with red borders around the 4 clusters
hist(MainStudyMelt$MediaRichness)
plot(density(MainStudyMelt$MediaRichness))
screeplot(prcomp(cbind(MainStudyMelt$SMP,MainStudyMelt$MediaRichness),type="lines"))
## Warning: In prcomp.default(cbind(MainStudyMelt$SMP, MainStudyMelt$MediaRichness),
## type = "lines") :
## extra argument 'type' will be disregarded
summary(prcomp(cbind(MainStudyMelt$SMP,MainStudyMelt$MediaRichness)))
## Importance of components%s:
## PC1 PC2
## Standard deviation 2.5873 1.6883
## Proportion of Variance 0.7014 0.2987
## Cumulative Proportion 0.7014 1.0000
## At least two components
mydata<-data.frame(MainStudyMelt$MediaRichness)
wss <- (nrow(mydata)-1)*sum(apply(mydata,2,var))
for (i in 2:15) wss[i] <- sum(kmeans(mydata,
centers=i)$withinss)
plot(1:15, wss, type="b", xlab="Number of Clusters",
ylab="Within groups sum of squares")
wss<-wss/sum(wss)*100
for (i in 2:15)
wss[i]<-wss[i]+wss[i-1]
plot(1:15, wss, type="b", xlab="Number of Clusters",
ylab="% Var explained")
wss
## [1] 56.97477 72.13680 80.73386 85.51758 89.35718 91.52489 93.28846
## [8] 94.54336 95.41749 97.07264 97.71886 98.53486 99.20590 99.66907
## [15] 100.00000
## 3 clusters explain more than 80% of the variance
abline(v=3,lty=2)
d <- dist(mydata,method="euclidean") # distance matrix
fit <- hclust(d, method="ward")
## The "ward" method has been renamed to "ward.D"; note new "ward.D2"
plot(fit) # display dendogram all raw data
groups <- cutree(fit,k=3) # cut tree into 3 clusters
rect.hclust(fit,k=3,border="red") # draw dendogram with red borders around the 4 clusters
mydata2<-aggregate(MainStudyMelt$MediaRichness,list(MainStudyMelt$SMP),mean)
rownames(mydata2)<-c("Facebook","Twitter","YouTube","Instagram","Pinterest",
"SnapChat","LinkedIn","SecondLife")
d<-dist(mydata2,method="euclidean") # distance matrix
## Warning in dist(mydata2, method = "euclidean"): NAs introduced by coercion
fit <- hclust(d, method="ward")
## The "ward" method has been renamed to "ward.D"; note new "ward.D2"
plot(fit,ylab="Content vividness") # display dendogram mean by SMP
groups <- cutree(fit,k=3) # cut tree into 4 clusters
rect.hclust(fit,k=3,border="red") # draw dendogram with red borders around the 4 clusters
## STUDY 1 (SECONDARY DATA)
cat("\014") # cleans screen
rm(list=ls(all=TRUE)) # remove variables in working memory
setwd("C:/Users/Erik Ernesto Vazquez/Documents") # sets working directory
X<-read.csv("E-Retailer.csv", skip=0, header=T) # reads raw data from Qualtrics
X<-subset(X,X$Merchandise.Category=="Apparel/Accessories"|
X$Merchandise.Category=="Computers/Electronics"|
X$Merchandise.Category=="Health/Beauty")
X$ProdCat<-ifelse(X$Merchandise.Category=="Computers/Electronics","Search",
ifelse(X$Merchandise.Category=="Apparel/Accessories","Experience",
"Credence"))
X$ProdCatLvl<-ifelse(X$Merchandise.Category=="Computers/Electronics",3,
ifelse(X$Merchandise.Category=="Apparel/Accessories",2,
1))
X$WebOnly<-ifelse(X$Merchant.Type=="Web Only",1,0)
X$RetailChain<-ifelse(X$Merchant.Type=="Retail Chain",1,0)
X$ConsumerBM<-ifelse(X$Merchant.Type=="Consumer Brand Manufacturer",1,0)
X$CatalogCallCenter<-ifelse(X$Merchant.Type=="Catallog/Call Center",1,0)
X$ConsistencyLvl<-ifelse(X$Consistency=="Poor",1,
ifelse(X$Consistency=="Fair",2,
ifelse(X$Consistency=="Good",3,
4)))
X$PersonalizationBinary<-ifelse(X$Personalization=="",0,1)
X$ConsistencyLvl<-ifelse(X$Consistency=="Poor",1,
ifelse(X$Consistency=="Fair",2,
ifelse(X$Consistency=="Good",3,
4)))
X$Mobile.Commerce.PlatformBinary<-ifelse(X$Mobile.Commerce.Platform=="",0,1)
X$X2011.Annual.Visits<-X$X2011.Monthly.Visits*12
table(X$Merchandise.Category)
##
## Apparel/Accessories Automotive Parts/Accessories
## 139 0
## Books/Music/Video Computers/Electronics
## 0 48
## Flowers/Gifts Food/Drug
## 0 0
## Hardware/Home Improvement Health/Beauty
## 0 29
## Housewares/Home Furnishings Jewelry
## 0 0
## Mass Merchant Office Supplies
## 0 0
## Specialty/Non-Apparel Sporting Goods
## 0 0
## Toys/Hobbies
## 0
count(X,c("Merchandise.Category","Merchant.Type"))
## Merchandise.Category Merchant.Type freq
## 1 Apparel/Accessories Catalog/Call Center 17
## 2 Apparel/Accessories Consumer Brand Manufacturer 34
## 3 Apparel/Accessories Retail Chain 60
## 4 Apparel/Accessories Web Only 28
## 5 Computers/Electronics Catalog/Call Center 7
## 6 Computers/Electronics Consumer Brand Manufacturer 12
## 7 Computers/Electronics Retail Chain 9
## 8 Computers/Electronics Web Only 20
## 9 Health/Beauty Catalog/Call Center 3
## 10 Health/Beauty Consumer Brand Manufacturer 3
## 11 Health/Beauty Retail Chain 7
## 12 Health/Beauty Web Only 16
count(X,c("Merchandise.Category","Mobile.Commerce.PlatformBinary"))
## Merchandise.Category Mobile.Commerce.PlatformBinary freq
## 1 Apparel/Accessories 0 95
## 2 Apparel/Accessories 1 44
## 3 Computers/Electronics 0 39
## 4 Computers/Electronics 1 9
## 5 Health/Beauty 0 15
## 6 Health/Beauty 1 14
aggregate(X$X2011.Annual.Visits,list(X$Merchandise.Category),mean)
## Group.1 x
## 1 Apparel/Accessories 43070925
## 2 Computers/Electronics 183258367
## 3 Health/Beauty 27888090
aggregate(X$X2011.Annual.Visits,list(X$Merchandise.Category),sd)
## Group.1 x
## 1 Apparel/Accessories 59335237
## 2 Computers/Electronics 868808208
## 3 Health/Beauty 64033836
leveneTest(X$X2011.Annual.Visits,X$Merchandise.Category)
## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 2 2.4274 0.09071 .
## 213
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
fit<-(lm(X2011.Annual.Visits~ProdCatLvl+WebOnly+RetailChain+
ConsumerBM+ConsistencyLvl+PersonalizationBinary+
Mobile.Commerce.PlatformBinary,data=X,x=T))
vif(fit)
## ProdCatLvl WebOnly
## 1.050945 2.422315
## RetailChain ConsumerBM
## 2.483937 2.192398
## ConsistencyLvl PersonalizationBinary
## 1.043076 1.057326
## Mobile.Commerce.PlatformBinary
## 1.070315
summary(fit) ## H1a Not supported
##
## Call:
## lm(formula = X2011.Annual.Visits ~ ProdCatLvl + WebOnly + RetailChain +
## ConsumerBM + ConsistencyLvl + PersonalizationBinary + Mobile.Commerce.PlatformBinary,
## data = X, x = T)
##
## Residuals:
## Min 1Q Median 3Q Max
## -283817308 -89400048 -31368030 38690661 5711382692
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -203206992 155593642 -1.306 0.1930
## ProdCatLvl 85059986 48640424 1.749 0.0818 .
## WebOnly -10475055 95503197 -0.110 0.9128
## RetailChain 20398481 92472769 0.221 0.8256
## ConsumerBM 120294333 99064444 1.214 0.2260
## ConsistencyLvl 29087502 23238394 1.252 0.2121
## PersonalizationBinary -32131496 57712032 -0.557 0.5783
## Mobile.Commerce.PlatformBinary 15176372 62667114 0.242 0.8089
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 411800000 on 208 degrees of freedom
## Multiple R-squared: 0.04266, Adjusted R-squared: 0.01044
## F-statistic: 1.324 on 7 and 208 DF, p-value: 0.2402
aggregate(X$X2011.Conversion.Rate,list(X$Merchandise.Category),mean)
## Group.1 x
## 1 Apparel/Accessories 0.02956835
## 2 Computers/Electronics 0.02250000
## 3 Health/Beauty 0.05655172
aggregate(X$X2011.Conversion.Rate,list(X$Merchandise.Category),sd)
## Group.1 x
## 1 Apparel/Accessories 0.01735689
## 2 Computers/Electronics 0.01344809
## 3 Health/Beauty 0.03819834
leveneTest(X$X2011.Conversion.Rate,X$Merchandise.Category)
## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 2 8.454 0.000293 ***
## 213
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
fit<-(lm(X2011.Conversion.Rate~ProdCatLvl+WebOnly+RetailChain+
ConsumerBM+ConsistencyLvl+PersonalizationBinary+
Mobile.Commerce.PlatformBinary,X))
vif(fit)
## ProdCatLvl WebOnly
## 1.050945 2.422315
## RetailChain ConsumerBM
## 2.483937 2.192398
## ConsistencyLvl PersonalizationBinary
## 1.043076 1.057326
## Mobile.Commerce.PlatformBinary
## 1.070315
summary(fit) ## H1b Approved
##
## Call:
## lm(formula = X2011.Conversion.Rate ~ ProdCatLvl + WebOnly + RetailChain +
## ConsumerBM + ConsistencyLvl + PersonalizationBinary + Mobile.Commerce.PlatformBinary,
## data = X)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.045984 -0.009485 -0.003161 0.007115 0.103892
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.089544 0.007047 12.708 < 2e-16 ***
## ProdCatLvl -0.015505 0.002203 -7.039 2.77e-11 ***
## WebOnly -0.017721 0.004325 -4.097 5.99e-05 ***
## RetailChain -0.029782 0.004188 -7.111 1.81e-11 ***
## ConsumerBM -0.028788 0.004486 -6.417 9.24e-10 ***
## ConsistencyLvl -0.001202 0.001052 -1.142 0.255
## PersonalizationBinary -0.003177 0.002614 -1.215 0.226
## Mobile.Commerce.PlatformBinary 0.004473 0.002838 1.576 0.117
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.01865 on 208 degrees of freedom
## Multiple R-squared: 0.3628, Adjusted R-squared: 0.3413
## F-statistic: 16.92 on 7 and 208 DF, p-value: < 2.2e-16
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$X60<5) ## Non-repeated measures
MainStudy1<-subset(MainStudy,MainStudy$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
Demographics<-cbind(MainStudy[187:189],MainStudy[194:195],MainStudy[202:207])
Demographics$Age<-2014-MainStudy$X187
Demographics$Income<-MainStudy$X189
Demographics$Education<-MainStudy$X194
Demographics$Location1<-Demographics$X202
Demographics$Location2<-Demographics$X203
Demographics$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")
cronbach(cbind(MainStudy$X126,MainStudy$X127,MainStudy$X128,MainStudy$X134,MainStudy$X135)) ## Quality Alpha 0.82 Good
## $sample.size
## [1] 900
##
## $number.of.items
## [1] 5
##
## $alpha
## [1] 0.9084386
MainStudy$Quality<-(MainStudy$X126+MainStudy$X127+MainStudy$X128+MainStudy$X134+MainStudy$X135)/5
2014-mean(MainStudy$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$X207),sd)
## Group.1 x
## 1 Facebook-Credence 10.517302
## 2 Facebook-Experience 10.157442
## 3 Facebook-Search 9.443078
## 4 Twitter-Credence 9.908363
## 5 Twitter-Experience 8.559265
## 6 Twitter-Search 9.946173
## 7 YouTube-Credence 9.467648
## 8 YouTube-Experience 9.677862
## 9 YouTube-Search 10.349235
aggregate(MainStudy$Quality,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 5.402
## 2 Facebook-Experience 5.792
## 3 Facebook-Search 5.336
## 4 Twitter-Credence 5.254
## 5 Twitter-Experience 5.732
## 6 Twitter-Search 6.126
## 7 YouTube-Credence 5.514
## 8 YouTube-Experience 5.970
## 9 YouTube-Search 6.276
aggregate(MainStudy$Quality,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 1.2363174
## 2 Facebook-Experience 1.0909805
## 3 Facebook-Search 1.6572778
## 4 Twitter-Credence 1.4269711
## 5 Twitter-Experience 1.2398175
## 6 Twitter-Search 1.2819351
## 7 YouTube-Credence 1.2408062
## 8 YouTube-Experience 1.0852082
## 9 YouTube-Search 0.9972812
## Brand Familiarity
aggregate(MainStudy$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$X60)
## [1] 0.6512293
## Effects of content richness and product cat on quality
aggregate(MainStudy$Quality,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 5.402
## 2 Facebook-Experience 5.792
## 3 Facebook-Search 5.336
## 4 Twitter-Credence 5.254
## 5 Twitter-Experience 5.732
## 6 Twitter-Search 6.126
## 7 YouTube-Credence 5.514
## 8 YouTube-Experience 5.970
## 9 YouTube-Search 6.276
aggregate(MainStudy$Quality,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 1.2363174
## 2 Facebook-Experience 1.0909805
## 3 Facebook-Search 1.6572778
## 4 Twitter-Credence 1.4269711
## 5 Twitter-Experience 1.2398175
## 6 Twitter-Search 1.2819351
## 7 YouTube-Credence 1.2408062
## 8 YouTube-Experience 1.0852082
## 9 YouTube-Search 0.9972812
aov.out<-aov(Quality~X207,MainStudy)
TukeyHSD(aov.out)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Quality ~ X207, data = MainStudy)
##
## $X207
## diff lwr upr
## Facebook-Experience-Facebook-Credence 0.390 -0.16611075 0.94611075
## Facebook-Search-Facebook-Credence -0.066 -0.62211075 0.49011075
## Twitter-Credence-Facebook-Credence -0.148 -0.70411075 0.40811075
## Twitter-Experience-Facebook-Credence 0.330 -0.22611075 0.88611075
## Twitter-Search-Facebook-Credence 0.724 0.16788925 1.28011075
## YouTube-Credence-Facebook-Credence 0.112 -0.44411075 0.66811075
## YouTube-Experience-Facebook-Credence 0.568 0.01188925 1.12411075
## YouTube-Search-Facebook-Credence 0.874 0.31788925 1.43011075
## Facebook-Search-Facebook-Experience -0.456 -1.01211075 0.10011075
## Twitter-Credence-Facebook-Experience -0.538 -1.09411075 0.01811075
## Twitter-Experience-Facebook-Experience -0.060 -0.61611075 0.49611075
## Twitter-Search-Facebook-Experience 0.334 -0.22211075 0.89011075
## YouTube-Credence-Facebook-Experience -0.278 -0.83411075 0.27811075
## YouTube-Experience-Facebook-Experience 0.178 -0.37811075 0.73411075
## YouTube-Search-Facebook-Experience 0.484 -0.07211075 1.04011075
## Twitter-Credence-Facebook-Search -0.082 -0.63811075 0.47411075
## Twitter-Experience-Facebook-Search 0.396 -0.16011075 0.95211075
## Twitter-Search-Facebook-Search 0.790 0.23388925 1.34611075
## YouTube-Credence-Facebook-Search 0.178 -0.37811075 0.73411075
## YouTube-Experience-Facebook-Search 0.634 0.07788925 1.19011075
## YouTube-Search-Facebook-Search 0.940 0.38388925 1.49611075
## Twitter-Experience-Twitter-Credence 0.478 -0.07811075 1.03411075
## Twitter-Search-Twitter-Credence 0.872 0.31588925 1.42811075
## YouTube-Credence-Twitter-Credence 0.260 -0.29611075 0.81611075
## YouTube-Experience-Twitter-Credence 0.716 0.15988925 1.27211075
## YouTube-Search-Twitter-Credence 1.022 0.46588925 1.57811075
## Twitter-Search-Twitter-Experience 0.394 -0.16211075 0.95011075
## YouTube-Credence-Twitter-Experience -0.218 -0.77411075 0.33811075
## YouTube-Experience-Twitter-Experience 0.238 -0.31811075 0.79411075
## YouTube-Search-Twitter-Experience 0.544 -0.01211075 1.10011075
## YouTube-Credence-Twitter-Search -0.612 -1.16811075 -0.05588925
## YouTube-Experience-Twitter-Search -0.156 -0.71211075 0.40011075
## YouTube-Search-Twitter-Search 0.150 -0.40611075 0.70611075
## YouTube-Experience-YouTube-Credence 0.456 -0.10011075 1.01211075
## YouTube-Search-YouTube-Credence 0.762 0.20588925 1.31811075
## YouTube-Search-YouTube-Experience 0.306 -0.25011075 0.86211075
## p adj
## Facebook-Experience-Facebook-Credence 0.4192655
## Facebook-Search-Facebook-Credence 0.9999904
## Twitter-Credence-Facebook-Credence 0.9960400
## Twitter-Experience-Facebook-Credence 0.6518033
## Twitter-Search-Facebook-Credence 0.0018343
## YouTube-Credence-Facebook-Credence 0.9994629
## YouTube-Experience-Facebook-Credence 0.0409829
## YouTube-Search-Facebook-Credence 0.0000426
## Facebook-Search-Facebook-Experience 0.2099053
## Twitter-Credence-Facebook-Experience 0.0669387
## Twitter-Experience-Facebook-Experience 0.9999954
## Twitter-Search-Facebook-Experience 0.6364544
## YouTube-Credence-Facebook-Experience 0.8290689
## YouTube-Experience-Facebook-Experience 0.9863260
## YouTube-Search-Facebook-Experience 0.1469837
## Twitter-Credence-Facebook-Search 0.9999487
## Twitter-Experience-Facebook-Search 0.3972076
## Twitter-Search-Facebook-Search 0.0003822
## YouTube-Credence-Facebook-Search 0.9863260
## YouTube-Experience-Facebook-Search 0.0122796
## YouTube-Search-Facebook-Search 0.0000065
## Twitter-Experience-Twitter-Credence 0.1591300
## Twitter-Search-Twitter-Credence 0.0000450
## YouTube-Credence-Twitter-Credence 0.8762935
## YouTube-Experience-Twitter-Credence 0.0021969
## YouTube-Search-Twitter-Credence 0.0000005
## Twitter-Search-Twitter-Experience 0.4045090
## YouTube-Credence-Twitter-Experience 0.9523703
## YouTube-Experience-Twitter-Experience 0.9220249
## YouTube-Search-Twitter-Experience 0.0608634
## YouTube-Credence-Twitter-Search 0.0186954
## YouTube-Experience-Twitter-Search 0.9943145
## YouTube-Search-Twitter-Search 0.9956550
## YouTube-Experience-YouTube-Credence 0.2099053
## YouTube-Search-YouTube-Credence 0.0007565
## YouTube-Search-YouTube-Experience 0.7398261
summary(aov(Quality~X205+X206,MainStudy))
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 25.2 12.620 7.751 0.00046 ***
## X206 2 47.5 23.729 14.574 5.91e-07 ***
## Residuals 895 1457.2 1.628
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
aggregate(MainStudy$Quality,list(MainStudy$X205),mean)
## Group.1 x
## 1 Facebook 5.510
## 2 Twitter 5.704
## 3 YouTube 5.920
aggregate(MainStudy$Quality,list(MainStudy$X205),sd)
## Group.1 x
## 1 Facebook 1.360221
## 2 Twitter 1.361923
## 3 YouTube 1.152110
aggregate(MainStudy$Quality,list(MainStudy$X206),mean)
## Group.1 x
## 1 Credence 5.390000
## 2 Experience 5.831333
## 3 Search 5.912667
aggregate(MainStudy$Quality,list(MainStudy$X206),sd)
## Group.1 x
## 1 Credence 1.304392
## 2 Experience 1.141591
## 3 Search 1.397647
aov.out<-aov(Quality~X205+X206,MainStudy)
TukeyHSD(aov.out)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Quality ~ X205 + X206, data = MainStudy)
##
## $X205
## diff lwr upr p adj
## Twitter-Facebook 0.194 -0.0505869 0.4385869 0.1504328
## YouTube-Facebook 0.410 0.1654131 0.6545869 0.0002637
## YouTube-Twitter 0.216 -0.0285869 0.4605869 0.0960073
##
## $X206
## diff lwr upr p adj
## Experience-Credence 0.44133333 0.1967464 0.6859202 0.0000744
## Search-Credence 0.52266667 0.2780798 0.7672536 0.0000019
## Search-Experience 0.08133333 -0.1632536 0.3259202 0.7150230
MainStudy$TieStr<-ifelse(MainStudy$X205=="Facebook","Strong","Weak")
MainStudy$TieStrLvl<-ifelse(MainStudy$X205=="Facebook",3,
ifelse(MainStudy$X205=="Twitter",2,1))
MainStudy$MR<-ifelse(MainStudy$X205=="Twitter","Poor","Rich")
MainStudy$MRLvl<-ifelse(MainStudy$X205=="Twitter",1,
ifelse(MainStudy$X205=="Facebook",2,3))
MainStudy$Pure<-ifelse(MainStudy$X205=="YouTube","Mix","Pure")
MainStudyX<-subset(MainStudy,MainStudy$X206=="Search")
aggregate(MainStudyX$Quality,list(MainStudyX$MR),mean)
## Group.1 x
## 1 Poor 6.126
## 2 Rich 5.806
aggregate(MainStudyX$Quality,list(MainStudyX$MR),sd)
## Group.1 x
## 1 Poor 1.281935
## 2 Rich 1.443323
## Cohen d 0.23442944671866184 and effect size r 0.11641770326988482
## above small effect size
## using https://www.uccs.edu/lbecker/
aggregate(MainStudyX$Quality,list(MainStudyX$TieStr),mean)
## Group.1 x
## 1 Strong 5.336
## 2 Weak 6.201
aggregate(MainStudyX$Quality,list(MainStudyX$TieStr),sd)
## Group.1 x
## 1 Strong 1.657278
## 2 Weak 1.148037
## Cohen d 0.6067703798001213 and effect size r 0.29031841991493496
## above medium effect size
## Cohen suggests that d values of 0.2, 0.5, and 0.8
## represent small, medium, and large effect sizes respectively
## Hence, above small effect from vividness vs above medium effect from tie str
## H1b approved
t.test(Quality~TieStr,MainStudyX) ## H1a approved
##
## Welch Two Sample t-test
##
## data: Quality by TieStr
## t = -4.6873, df = 147.97, p-value = 6.229e-06
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.2296777 -0.5003223
## sample estimates:
## mean in group Strong mean in group Weak
## 5.336 6.201
summary(lm(Quality~TieStrLvl,MainStudyX))
##
## Call:
## lm(formula = Quality ~ TieStrLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.4427 -0.7977 0.0173 1.0173 3.0873
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.85267 0.20560 33.329 < 2e-16 ***
## TieStrLvl -0.47000 0.09518 -4.938 1.32e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.346 on 298 degrees of freedom
## Multiple R-squared: 0.07564, Adjusted R-squared: 0.07254
## F-statistic: 24.39 on 1 and 298 DF, p-value: 1.316e-06
t.test(Quality~MR,MainStudyX)
##
## Welch Two Sample t-test
##
## data: Quality by MR
## t = 1.9529, df = 220.25, p-value = 0.0521
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.002930428 0.642930428
## sample estimates:
## mean in group Poor mean in group Rich
## 6.126 5.806
summary(lm(Quality~MRLvl,MainStudyX))
##
## Call:
## lm(formula = Quality ~ MRLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.9127 -0.8377 0.0123 1.0123 3.1623
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.7627 0.2137 26.973 <2e-16 ***
## MRLvl 0.0750 0.0989 0.758 0.449
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.399 on 298 degrees of freedom
## Multiple R-squared: 0.001926, Adjusted R-squared: -0.001423
## F-statistic: 0.5751 on 1 and 298 DF, p-value: 0.4488
MainStudyX<-subset(MainStudy,MainStudy$X206=="Experience")
aggregate(MainStudyX$Quality,list(MainStudyX$MR),mean)
## Group.1 x
## 1 Poor 5.732
## 2 Rich 5.881
aggregate(MainStudyX$Quality,list(MainStudyX$MR),sd)
## Group.1 x
## 1 Poor 1.239818
## 2 Rich 1.089022
## Cohen d 0.12769328761125776 and effect size r 0.06371690827553132
## below small effect-size
aggregate(MainStudyX$Quality,list(MainStudyX$TieStr),mean)
## Group.1 x
## 1 Strong 5.792
## 2 Weak 5.851
aggregate(MainStudyX$Quality,list(MainStudyX$TieStr),sd)
## Group.1 x
## 1 Strong 1.090980
## 2 Weak 1.168256
## Cohen d 0.052199519508090444 and effect size r 0.02609087474810489
## below small effect-size
## Cohen suggests that d values of 0.2, 0.5, and 0.8
## represent small, medium, and large effect sizes respectively
## Hence, small effect from vividness vs small effect from tie str
## the effect of content vividness is only 2.4x greater than vividness
## in this context the effect are considered similar
## it should be +5x the difference to be considered different
## according to the Cohen's d comparisons
## H3b approved
12/5
## [1] 2.4
t.test(Quality~TieStr,MainStudyX) ## H3a rejected
##
## Welch Two Sample t-test
##
## data: Quality by TieStr
## t = -0.43115, df = 210.62, p-value = 0.6668
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.3287609 0.2107609
## sample estimates:
## mean in group Strong mean in group Weak
## 5.792 5.851
summary(lm(Quality~TieStrLvl,MainStudyX))
##
## Call:
## lm(formula = Quality ~ TieStrLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.8313 -0.7423 -0.1313 0.7909 3.2577
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.00933 0.17432 34.473 <2e-16 ***
## TieStrLvl -0.08900 0.08069 -1.103 0.271
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.141 on 298 degrees of freedom
## Multiple R-squared: 0.004066, Adjusted R-squared: 0.0007235
## F-statistic: 1.216 on 1 and 298 DF, p-value: 0.2709
t.test(Quality~MR,MainStudyX)
##
## Welch Two Sample t-test
##
## data: Quality by MR
## t = -1.0209, df = 177.01, p-value = 0.3087
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.4370254 0.1390254
## sample estimates:
## mean in group Poor mean in group Rich
## 5.732 5.881
summary(lm(Quality~MRLvl,MainStudyX))
##
## Call:
## lm(formula = Quality ~ MRLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.8313 -0.7123 -0.1123 0.8497 3.1687
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.59333 0.17404 32.139 <2e-16 ***
## MRLvl 0.11900 0.08056 1.477 0.141
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.139 on 298 degrees of freedom
## Multiple R-squared: 0.007268, Adjusted R-squared: 0.003937
## F-statistic: 2.182 on 1 and 298 DF, p-value: 0.1407
MainStudyX<-subset(MainStudy,MainStudy$X206=="Credence")
aggregate(MainStudyX$Quality,list(MainStudyX$MR),mean)
## Group.1 x
## 1 Poor 5.254
## 2 Rich 5.458
aggregate(MainStudyX$Quality,list(MainStudyX$MR),sd)
## Group.1 x
## 1 Poor 1.426971
## 2 Rich 1.236723
## Cohen d 0.15278155887796593 and effect size r 0.07616885908703643
## below small effect size
aggregate(MainStudyX$Quality,list(MainStudyX$TieStr),mean)
## Group.1 x
## 1 Strong 5.402
## 2 Weak 5.384
aggregate(MainStudyX$Quality,list(MainStudyX$TieStr),sd)
## Group.1 x
## 1 Strong 1.236317
## 2 Weak 1.340121
## Cohen d 0.013961452515901696 and effect size r 0.00698055617689004
## below small effect size
## Cohen suggests that d values of 0.2, 0.5, and 0.8
## represent small, medium, and large effect sizes respectively
## Hence, both vividness and tie strength have below small effect size but
## the effect of vividness is +11x greater than tie strength
15/1.3
## [1] 11.53846
## H3a accepted
t.test(Quality~MR,MainStudyX)
##
## Welch Two Sample t-test
##
## data: Quality by MR
## t = -1.2189, df = 175.04, p-value = 0.2245
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.5343062 0.1263062
## sample estimates:
## mean in group Poor mean in group Rich
## 5.254 5.458
summary(lm(Quality~MRLvl,MainStudyX))
##
## Call:
## lm(formula = Quality ~ MRLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.3900 -0.5200 -0.1900 0.7575 3.7400
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.13000 0.19892 25.789 <2e-16 ***
## MRLvl 0.13000 0.09208 1.412 0.159
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.302 on 298 degrees of freedom
## Multiple R-squared: 0.006644, Adjusted R-squared: 0.003311
## F-statistic: 1.993 on 1 and 298 DF, p-value: 0.1591
t.test(Quality~TieStr,MainStudyX) ## H4b rejected but with trends
##
## Welch Two Sample t-test
##
## data: Quality by TieStr
## t = 0.11555, df = 212.93, p-value = 0.9081
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.2890496 0.3250496
## sample estimates:
## mean in group Strong mean in group Weak
## 5.402 5.384
summary(lm(Quality~TieStrLvl,MainStudyX))
##
## Call:
## lm(formula = Quality ~ TieStrLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.390 -0.446 -0.134 0.768 3.610
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.50200 0.19946 27.584 <2e-16 ***
## TieStrLvl -0.05600 0.09233 -0.607 0.545
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.306 on 298 degrees of freedom
## Multiple R-squared: 0.001233, Adjusted R-squared: -0.002119
## F-statistic: 0.3678 on 1 and 298 DF, p-value: 0.5446
### ALTERNATIVE STUDY WITH VARIABLE PURCH INT
cronbach(cbind(MainStudy$X142,MainStudy$X143,MainStudy$X144)) ## PurchInt Alpha 0.95 Good
## $sample.size
## [1] 900
##
## $number.of.items
## [1] 3
##
## $alpha
## [1] 0.9463625
MainStudy$PurchInt<-(MainStudy$X142+MainStudy$X143+MainStudy$X144)/3
## Effects of content richness and product cat on quality
aggregate(MainStudy$PurchInt,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 2.663333
## 2 Facebook-Experience 2.856667
## 3 Facebook-Search 3.063333
## 4 Twitter-Credence 2.563333
## 5 Twitter-Experience 2.956667
## 6 Twitter-Search 3.476667
## 7 YouTube-Credence 3.300000
## 8 YouTube-Experience 3.433333
## 9 YouTube-Search 3.693333
aggregate(MainStudy$PurchInt,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 1.811235
## 2 Facebook-Experience 1.803635
## 3 Facebook-Search 1.847554
## 4 Twitter-Credence 1.872295
## 5 Twitter-Experience 2.000368
## 6 Twitter-Search 2.081534
## 7 YouTube-Credence 1.867977
## 8 YouTube-Experience 2.179643
## 9 YouTube-Search 2.011013
aov.out<-aov(PurchInt~X207,MainStudy)
TukeyHSD(aov.out)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PurchInt ~ X207, data = MainStudy)
##
## $X207
## diff lwr upr
## Facebook-Experience-Facebook-Credence 0.19333333 -0.66225748 1.0489242
## Facebook-Search-Facebook-Credence 0.40000000 -0.45559082 1.2555908
## Twitter-Credence-Facebook-Credence -0.10000000 -0.95559082 0.7555908
## Twitter-Experience-Facebook-Credence 0.29333333 -0.56225748 1.1489242
## Twitter-Search-Facebook-Credence 0.81333333 -0.04225748 1.6689242
## YouTube-Credence-Facebook-Credence 0.63666667 -0.21892415 1.4922575
## YouTube-Experience-Facebook-Credence 0.77000000 -0.08559082 1.6255908
## YouTube-Search-Facebook-Credence 1.03000000 0.17440918 1.8855908
## Facebook-Search-Facebook-Experience 0.20666667 -0.64892415 1.0622575
## Twitter-Credence-Facebook-Experience -0.29333333 -1.14892415 0.5622575
## Twitter-Experience-Facebook-Experience 0.10000000 -0.75559082 0.9555908
## Twitter-Search-Facebook-Experience 0.62000000 -0.23559082 1.4755908
## YouTube-Credence-Facebook-Experience 0.44333333 -0.41225748 1.2989242
## YouTube-Experience-Facebook-Experience 0.57666667 -0.27892415 1.4322575
## YouTube-Search-Facebook-Experience 0.83666667 -0.01892415 1.6922575
## Twitter-Credence-Facebook-Search -0.50000000 -1.35559082 0.3555908
## Twitter-Experience-Facebook-Search -0.10666667 -0.96225748 0.7489242
## Twitter-Search-Facebook-Search 0.41333333 -0.44225748 1.2689242
## YouTube-Credence-Facebook-Search 0.23666667 -0.61892415 1.0922575
## YouTube-Experience-Facebook-Search 0.37000000 -0.48559082 1.2255908
## YouTube-Search-Facebook-Search 0.63000000 -0.22559082 1.4855908
## Twitter-Experience-Twitter-Credence 0.39333333 -0.46225748 1.2489242
## Twitter-Search-Twitter-Credence 0.91333333 0.05774252 1.7689242
## YouTube-Credence-Twitter-Credence 0.73666667 -0.11892415 1.5922575
## YouTube-Experience-Twitter-Credence 0.87000000 0.01440918 1.7255908
## YouTube-Search-Twitter-Credence 1.13000000 0.27440918 1.9855908
## Twitter-Search-Twitter-Experience 0.52000000 -0.33559082 1.3755908
## YouTube-Credence-Twitter-Experience 0.34333333 -0.51225748 1.1989242
## YouTube-Experience-Twitter-Experience 0.47666667 -0.37892415 1.3322575
## YouTube-Search-Twitter-Experience 0.73666667 -0.11892415 1.5922575
## YouTube-Credence-Twitter-Search -0.17666667 -1.03225748 0.6789242
## YouTube-Experience-Twitter-Search -0.04333333 -0.89892415 0.8122575
## YouTube-Search-Twitter-Search 0.21666667 -0.63892415 1.0722575
## YouTube-Experience-YouTube-Credence 0.13333333 -0.72225748 0.9889242
## YouTube-Search-YouTube-Credence 0.39333333 -0.46225748 1.2489242
## YouTube-Search-YouTube-Experience 0.26000000 -0.59559082 1.1155908
## p adj
## Facebook-Experience-Facebook-Credence 0.9987561
## Facebook-Search-Facebook-Credence 0.8763194
## Twitter-Credence-Facebook-Credence 0.9999915
## Twitter-Experience-Facebook-Credence 0.9788078
## Twitter-Search-Facebook-Credence 0.0774089
## YouTube-Credence-Facebook-Credence 0.3347652
## YouTube-Experience-Facebook-Credence 0.1171999
## YouTube-Search-Facebook-Credence 0.0060055
## Facebook-Search-Facebook-Experience 0.9979945
## Twitter-Credence-Facebook-Experience 0.9788078
## Twitter-Experience-Facebook-Experience 0.9999915
## Twitter-Search-Facebook-Experience 0.3721506
## YouTube-Credence-Facebook-Experience 0.7987520
## YouTube-Experience-Facebook-Experience 0.4768534
## YouTube-Search-Facebook-Experience 0.0610479
## Twitter-Credence-Facebook-Search 0.6708217
## Twitter-Experience-Facebook-Search 0.9999859
## Twitter-Search-Facebook-Search 0.8546627
## YouTube-Credence-Facebook-Search 0.9948338
## YouTube-Experience-Facebook-Search 0.9175141
## YouTube-Search-Facebook-Search 0.3494894
## Twitter-Experience-Twitter-Credence 0.8863786
## Twitter-Search-Twitter-Credence 0.0261790
## YouTube-Credence-Twitter-Credence 0.1574417
## YouTube-Experience-Twitter-Credence 0.0427705
## YouTube-Search-Twitter-Credence 0.0014440
## Twitter-Search-Twitter-Experience 0.6210320
## YouTube-Credence-Twitter-Experience 0.9455162
## YouTube-Experience-Twitter-Experience 0.7263958
## YouTube-Search-Twitter-Experience 0.1574417
## YouTube-Credence-Twitter-Search 0.9993547
## YouTube-Experience-Twitter-Search 1.0000000
## YouTube-Search-Twitter-Search 0.9971996
## YouTube-Experience-YouTube-Credence 0.9999217
## YouTube-Search-YouTube-Credence 0.8863786
## YouTube-Search-YouTube-Experience 0.9902684
summary(aov(PurchInt~X205+X206,MainStudy))
## Df Sum Sq Mean Sq F value Pr(>F)
## X205 2 62 31.187 8.253 0.000281 ***
## X206 2 49 24.470 6.476 0.001614 **
## Residuals 895 3382 3.779
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
aggregate(MainStudy$PurchInt,list(MainStudy$X205),mean)
## Group.1 x
## 1 Facebook 2.861111
## 2 Twitter 2.998889
## 3 YouTube 3.475556
aggregate(MainStudy$PurchInt,list(MainStudy$X205),sd)
## Group.1 x
## 1 Facebook 1.822168
## 2 Twitter 2.015086
## 3 YouTube 2.023403
aggregate(MainStudy$PurchInt,list(MainStudy$X206),mean)
## Group.1 x
## 1 Credence 2.842222
## 2 Experience 3.082222
## 3 Search 3.411111
aggregate(MainStudy$PurchInt,list(MainStudy$X206),sd)
## Group.1 x
## 1 Credence 1.873239
## 2 Experience 2.009615
## 3 Search 1.993082
aov.out<-aov(PurchInt~X205+X206,MainStudy)
TukeyHSD(aov.out)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PurchInt ~ X205 + X206, data = MainStudy)
##
## $X205
## diff lwr upr p adj
## Twitter-Facebook 0.1377778 -0.2348388 0.5103944 0.6606165
## YouTube-Facebook 0.6144444 0.2418278 0.9870610 0.0003416
## YouTube-Twitter 0.4766667 0.1040501 0.8492833 0.0077287
##
## $X206
## diff lwr upr p adj
## Experience-Credence 0.2400000 -0.13261660 0.6126166 0.2856785
## Search-Credence 0.5688889 0.19627229 0.9415055 0.0010376
## Search-Experience 0.3288889 -0.04372771 0.7015055 0.0962469
MainStudy$TieStr<-ifelse(MainStudy$X205=="Facebook","Strong","Weak")
MainStudy$TieStrLvl<-ifelse(MainStudy$X205=="Facebook",3,
ifelse(MainStudy$X205=="Twitter",2,1))
MainStudy$MR<-ifelse(MainStudy$X205=="Twitter","Poor","Rich")
MainStudy$MRLvl<-ifelse(MainStudy$X205=="Twitter",1,
ifelse(MainStudy$X205=="Facebook",2,3))
MainStudy$Pure<-ifelse(MainStudy$X205=="YouTube","Mix","Pure")
## Monthly Expenses on Category
aggregate(MainStudy$X181,list(MainStudy$X206),mean)
## Group.1 x
## 1 Credence 23.25663
## 2 Experience 70.60667
## 3 Search 69.70667
aggregate(MainStudy$X181,list(MainStudy$X206),sd)
## Group.1 x
## 1 Credence 30.98185
## 2 Experience 71.13555
## 3 Search 157.61978
aggregate(MainStudy$X181,list(MainStudy$X207),mean)
## Group.1 x
## 1 Facebook-Credence 19.8200
## 2 Facebook-Experience 69.0800
## 3 Facebook-Search 53.0000
## 4 Twitter-Credence 22.1599
## 5 Twitter-Experience 72.2000
## 6 Twitter-Search 75.8600
## 7 YouTube-Credence 27.7900
## 8 YouTube-Experience 70.5400
## 9 YouTube-Search 80.2600
aggregate(MainStudy$X181,list(MainStudy$X207),sd)
## Group.1 x
## 1 Facebook-Credence 29.13687
## 2 Facebook-Experience 65.17648
## 3 Facebook-Search 108.92023
## 4 Twitter-Credence 29.85062
## 5 Twitter-Experience 73.43767
## 6 Twitter-Search 105.03496
## 7 YouTube-Credence 33.54246
## 8 YouTube-Experience 75.08022
## 9 YouTube-Search 227.38787
MainStudyX<-subset(MainStudy,MainStudy$X207=="Facebook-Search"|MainStudy$X207=="YouTube-Search")
t.test(X181~X207,MainStudyX)
##
## Welch Two Sample t-test
##
## data: X181 by X207
## t = -1.0812, df = 142.16, p-value = 0.2814
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -77.10059 22.58059
## sample estimates:
## mean in group Facebook-Search mean in group YouTube-Search
## 53.00 80.26
MainStudyX<-subset(MainStudy,MainStudy$X207=="Facebook-Credence"|MainStudy$X207=="YouTube-Credence")
t.test(X181~X207,MainStudyX) ## there is a difference in prices for credence groups
##
## Welch Two Sample t-test
##
## data: X181 by X207
## t = -1.7938, df = 194.2, p-value = 0.0744
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -16.7327927 0.7927927
## sample estimates:
## mean in group Facebook-Credence mean in group YouTube-Credence
## 19.82 27.79
MainStudyX<-subset(MainStudy,MainStudy$X206=="Search")
aggregate(MainStudyX$PurchInt,list(MainStudyX$MR),mean)
## Group.1 x
## 1 Poor 3.476667
## 2 Rich 3.378333
aggregate(MainStudyX$PurchInt,list(MainStudyX$MR),sd)
## Group.1 x
## 1 Poor 2.081534
## 2 Rich 1.951871
## Cohen d 0.0496 and effect-size r 0.0247
## below small effect size
## from https://www.uccs.edu/lbecker/
aggregate(MainStudyX$PurchInt,list(MainStudyX$TieStr),mean)
## Group.1 x
## 1 Strong 3.063333
## 2 Weak 3.585000
aggregate(MainStudyX$PurchInt,list(MainStudyX$TieStr),sd)
## Group.1 x
## 1 Strong 1.847554
## 2 Weak 2.044316
## Cohen d 0.2677 and effect-size r 0.1326
## above small effect size
## Cohen suggests that d values of 0.2, 0.5, and 0.8
## represent small, medium, and large effect sizes respectively
## Hence, 0.0496 is below small effect from vividness vs 0.2677 above small effect from tie str
## H1b approved
t.test(PurchInt~TieStr,MainStudyX) ## H1a approved WITH PRUCH INT
##
## Welch Two Sample t-test
##
## data: PurchInt by TieStr
## t = -2.2238, df = 216.88, p-value = 0.02719
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.98402736 -0.05930598
## sample estimates:
## mean in group Strong mean in group Weak
## 3.063333 3.585000
summary(lm(PurchInt~TieStrLvl,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ TieStrLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.7261 -1.7444 -0.4111 1.5889 5.2739
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.0411 0.3024 13.36 <2e-16 ***
## TieStrLvl -0.3150 0.1400 -2.25 0.0252 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.98 on 298 degrees of freedom
## Multiple R-squared: 0.01671, Adjusted R-squared: 0.01341
## F-statistic: 5.064 on 1 and 298 DF, p-value: 0.02516
t.test(PurchInt~MR,MainStudyX)
##
## Welch Two Sample t-test
##
## data: PurchInt by MR
## t = 0.39372, df = 187.19, p-value = 0.6942
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.3943601 0.5910267
## sample estimates:
## mean in group Poor mean in group Rich
## 3.476667 3.378333
summary(lm(PurchInt~MRLvl,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ MRLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.5194 -1.6361 -0.4111 1.5889 5.4806
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.1944 0.3047 10.485 <2e-16 ***
## MRLvl 0.1083 0.1410 0.768 0.443
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.994 on 298 degrees of freedom
## Multiple R-squared: 0.001976, Adjusted R-squared: -0.001373
## F-statistic: 0.5901 on 1 and 298 DF, p-value: 0.443
##controlling for monthly expenses on the category
summary(lm(PurchInt~TieStrLvl*X181,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ TieStrLvl * X181, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.7596 -1.7596 -0.4181 1.5803 5.2996
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.076e+00 3.235e-01 12.598 <2e-16 ***
## TieStrLvl -3.163e-01 1.520e-01 -2.081 0.0383 *
## X181 -3.170e-04 1.597e-03 -0.199 0.8428
## TieStrLvl:X181 -7.797e-05 9.675e-04 -0.081 0.9358
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.985 on 296 degrees of freedom
## Multiple R-squared: 0.01789, Adjusted R-squared: 0.007934
## F-statistic: 1.797 on 3 and 296 DF, p-value: 0.1478
summary(lm(PurchInt~MRLvl*X181,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ MRLvl * X181, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.5247 -1.6651 -0.4354 1.5672 5.4932
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.2989019 0.3626698 9.096 <2e-16 ***
## MRLvl 0.0752515 0.1608478 0.468 0.640
## X181 -0.0014377 0.0026388 -0.545 0.586
## MRLvl:X181 0.0004395 0.0009969 0.441 0.660
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2 on 296 degrees of freedom
## Multiple R-squared: 0.003272, Adjusted R-squared: -0.00683
## F-statistic: 0.3239 on 3 and 296 DF, p-value: 0.8081
MainStudyX<-subset(MainStudy,MainStudy$X206=="Experience")
aggregate(MainStudyX$PurchInt,list(MainStudyX$MR),mean)
## Group.1 x
## 1 Poor 2.956667
## 2 Rich 3.145000
aggregate(MainStudyX$PurchInt,list(MainStudyX$MR),sd)
## Group.1 x
## 1 Poor 2.000368
## 2 Rich 2.016287
## Cohen d 0.0937 and effect size r 0.0468
## below small effect size
aggregate(MainStudyX$PurchInt,list(MainStudyX$TieStr),mean)
## Group.1 x
## 1 Strong 2.856667
## 2 Weak 3.195000
aggregate(MainStudyX$PurchInt,list(MainStudyX$TieStr),sd)
## Group.1 x
## 1 Strong 1.803635
## 2 Weak 2.100298
## Cohen d 0.1728 and effect size r 0.0860
## below small effect size
## Cohen suggests that d values of 0.2, 0.5, and 0.8
## represent small, medium, and large effect sizes respectively
## Hence, small effect from vividness vs small effect from tie str
0.1728/0.0937
## [1] 1.844184
## H3b approved
t.test(PurchInt~TieStr,MainStudyX) ## H3a rejected
##
## Welch Two Sample t-test
##
## data: PurchInt by TieStr
## t = -1.4481, df = 226.87, p-value = 0.149
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.7987135 0.1220468
## sample estimates:
## mean in group Strong mean in group Weak
## 2.856667 3.195000
summary(lm(PurchInt~TieStrLvl,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ TieStrLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.3706 -1.7489 -0.7039 1.5394 6.2061
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6589 0.3054 11.98 <2e-16 ***
## TieStrLvl -0.2883 0.1414 -2.04 0.0423 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.999 on 298 degrees of freedom
## Multiple R-squared: 0.01377, Adjusted R-squared: 0.01046
## F-statistic: 4.161 on 1 and 298 DF, p-value: 0.04226
t.test(PurchInt~MR,MainStudyX)
##
## Welch Two Sample t-test
##
## data: PurchInt by MR
## t = -0.76669, df = 199.52, p-value = 0.4442
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.6727282 0.2960615
## sample estimates:
## mean in group Poor mean in group Rich
## 2.956667 3.145000
summary(lm(PurchInt~MRLvl,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ MRLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.3206 -1.7489 -0.6539 1.4894 6.1561
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.6056 0.3060 8.514 8.4e-16 ***
## MRLvl 0.2383 0.1417 1.682 0.0936 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.003 on 298 degrees of freedom
## Multiple R-squared: 0.009408, Adjusted R-squared: 0.006084
## F-statistic: 2.83 on 1 and 298 DF, p-value: 0.09355
##controlling for monthly expenses on the category
summary(lm(PurchInt~TieStrLvl*X181,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ TieStrLvl * X181, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.4881 -1.5550 -0.6043 1.2739 6.0880
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.138493 0.417602 7.516 6.78e-13 ***
## TieStrLvl -0.226502 0.196701 -1.152 0.2505
## X181 0.007267 0.004135 1.758 0.0798 .
## TieStrLvl:X181 -0.000827 0.001998 -0.414 0.6792
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.964 on 296 degrees of freedom
## Multiple R-squared: 0.05486, Adjusted R-squared: 0.04528
## F-statistic: 5.727 on 3 and 296 DF, p-value: 0.0008038
summary(lm(PurchInt~MRLvl*X181,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ MRLvl * X181, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.6790 -1.4697 -0.5501 1.3714 5.9229
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.864775 0.420568 4.434 1.31e-05 ***
## MRLvl 0.404109 0.193027 2.094 0.0372 *
## X181 0.010289 0.004110 2.503 0.0128 *
## MRLvl:X181 -0.002256 0.001879 -1.200 0.2309
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.963 on 296 degrees of freedom
## Multiple R-squared: 0.05532, Adjusted R-squared: 0.04575
## F-statistic: 5.778 on 3 and 296 DF, p-value: 0.0007503
MainStudyX<-subset(MainStudy,MainStudy$X206=="Credence")
aggregate(MainStudyX$PurchInt,list(MainStudyX$MR),mean)
## Group.1 x
## 1 Poor 2.563333
## 2 Rich 2.981667
aggregate(MainStudyX$PurchInt,list(MainStudyX$MR),sd)
## Group.1 x
## 1 Poor 1.872295
## 2 Rich 1.862737
## Cohen d 0.22 and effect size r 0.11
aggregate(MainStudyX$PurchInt,list(MainStudyX$TieStr),mean)
## Group.1 x
## 1 Strong 2.663333
## 2 Weak 2.931667
aggregate(MainStudyX$PurchInt,list(MainStudyX$TieStr),sd)
## Group.1 x
## 1 Strong 1.811235
## 2 Weak 1.901628
## Cohen d 0.14 and effect size r 0.07
## Cohen suggests that d values of 0.2, 0.5, and 0.8
## represent small, medium, and large effect sizes respectively
## Hence, small effect from vividness vs below small effect from tie str
## they have almos the same effect
22/14
## [1] 1.571429
## H4a rejected
t.test(PurchInt~MR,MainStudyX)
##
## Welch Two Sample t-test
##
## data: PurchInt by MR
## t = -1.8274, df = 197.21, p-value = 0.06915
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.86977641 0.03310974
## sample estimates:
## mean in group Poor mean in group Rich
## 2.563333 2.981667
summary(lm(PurchInt~MRLvl,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ MRLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.2106 -1.4739 -0.5089 1.1928 5.8594
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.1056 0.2829 7.443 1.06e-12 ***
## MRLvl 0.3683 0.1310 2.813 0.00524 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.852 on 298 degrees of freedom
## Multiple R-squared: 0.02586, Adjusted R-squared: 0.02259
## F-statistic: 7.911 on 1 and 298 DF, p-value: 0.005239
t.test(PurchInt~TieStr,MainStudyX) ## H4b rejected
##
## Welch Two Sample t-test
##
## data: PurchInt by TieStr
## t = -1.1895, df = 206.93, p-value = 0.2356
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.7130645 0.1763978
## sample estimates:
## mean in group Strong mean in group Weak
## 2.663333 2.931667
summary(lm(PurchInt~TieStrLvl,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ TieStrLvl, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.1606 -1.5239 -0.5239 1.4761 5.8094
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.4789 0.2838 12.257 <2e-16 ***
## TieStrLvl -0.3183 0.1314 -2.423 0.016 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.858 on 298 degrees of freedom
## Multiple R-squared: 0.01932, Adjusted R-squared: 0.01603
## F-statistic: 5.87 on 1 and 298 DF, p-value: 0.016
MainStudy$WOM<-(MainStudy$X151+MainStudy$X152+MainStudy$X153)/3
##controlling for monthly expenses on the category
summary(lm(PurchInt~TieStrLvl*X181,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ TieStrLvl * X181, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.3214 -1.5072 -0.5072 1.2053 5.1975
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.175548 0.353719 8.978 <2e-16 ***
## TieStrLvl -0.334196 0.160558 -2.081 0.0383 *
## X181 0.008489 0.008518 0.997 0.3198
## TieStrLvl:X181 0.003138 0.004107 0.764 0.4455
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.808 on 296 degrees of freedom
## Multiple R-squared: 0.07769, Adjusted R-squared: 0.06834
## F-statistic: 8.311 on 3 and 296 DF, p-value: 2.524e-05
summary(lm(PurchInt~MRLvl*X181,MainStudyX))
##
## Call:
## lm(formula = PurchInt ~ MRLvl * X181, data = MainStudyX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.2238 -1.4753 -0.4898 1.2471 5.5247
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.973124 0.345669 5.708 2.77e-08 ***
## MRLvl 0.265581 0.162291 1.636 0.103
## X181 0.009331 0.009086 1.027 0.305
## MRLvl:X181 0.002499 0.004047 0.617 0.537
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.801 on 296 degrees of freedom
## Multiple R-squared: 0.08456, Adjusted R-squared: 0.07528
## F-statistic: 9.113 on 3 and 296 DF, p-value: 8.685e-06
## Full model
summary(lm(PurchInt~X181+MRLvl+TieStrLvl,MainStudy))
##
## Call:
## lm(formula = PurchInt ~ X181 + MRLvl + TieStrLvl, data = MainStudy)
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.1739 -1.7257 -0.5936 1.4685 6.0656
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.2853341 0.3285095 10.001 < 2e-16 ***
## X181 0.0013906 0.0006284 2.213 0.02716 *
## MRLvl 0.1160454 0.0920065 1.261 0.20754
## TieStrLvl -0.2406961 0.0921079 -2.613 0.00912 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.952 on 896 degrees of freedom
## Multiple R-squared: 0.02319, Adjusted R-squared: 0.01992
## F-statistic: 7.091 on 3 and 896 DF, p-value: 0.0001034
summary(lm(PurchInt~X181+TieStrLvl,MainStudy))
##
## Call:
## lm(formula = PurchInt ~ X181 + TieStrLvl, data = MainStudy)
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.0926 -1.7275 -0.5596 1.5265 6.1242
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6342713 0.1772044 20.509 < 2e-16 ***
## X181 0.0013786 0.0006286 2.193 0.028545 *
## TieStrLvl -0.2987922 0.0797897 -3.745 0.000192 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.952 on 897 degrees of freedom
## Multiple R-squared: 0.02146, Adjusted R-squared: 0.01928
## F-statistic: 9.835 on 2 and 897 DF, p-value: 5.954e-05
summary(lm(PurchInt~X181+MRLvl,MainStudy))
##
## Call:
## lm(formula = PurchInt ~ X181 + MRLvl, data = MainStudy)
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.2109 -1.7439 -0.6012 1.4644 5.8212
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5590589 0.1757131 14.564 <2e-16 ***
## X181 0.0014715 0.0006297 2.337 0.0197 *
## MRLvl 0.2362805 0.0799341 2.956 0.0032 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.958 on 897 degrees of freedom
## Multiple R-squared: 0.01575, Adjusted R-squared: 0.01355
## F-statistic: 7.176 on 2 and 897 DF, p-value: 0.0008094