MKTG 3P98 Project 2 Team 8
2024-06-19
Honda Brand Perspective
Prof. Lianxi Zhou
Dustin Pinch (6754550)
Rob Mantini (6282024)
Jeffery J. Unrau (7307572)
Cleaning Data
# Set working directory and install necessary packages
getwd()
setwd("/Users/dustinpinch/Downloads/Project 2")
install.packages("readr")
install.packages("dplyr")
install.packages("tidyr")
install.packages("ggplot2")
library(readr)
library(dplyr)
library(tidyr)
library(ggplot2)
# Read data from files
Car1<-read_csv("/Users/dustinpinch/Downloads/Project 2/Car_Survey_1.csv")
str(Car1)
head(Car1,n=10)
Car2<-read_csv("/Users/dustinpinch/Downloads/Project 2/Car_Survey_2.csv")
str(Car2)
head(Car2,n=10)
# Merge Files
names(Car2)[1]<-c("Resp")
head(Car2,n=1)
Car_Total<-merge(Car1,Car2,By="Resp")
str(Car_Total)
summary(Car_Total)
# Replace NA values with column mean
Car_Total<-Car_Total %>% mutate(across(where(is.numeric), ~replace(.,is.na(.), mean(.,na.rm = TRUE))))
na_Car_Total<-summarise_all(Car_Total, ~sum(is.na(.)))
print(na_Car_Total)Payment Methods
Data filtering and processing
# Filter data for the region
global_car_data <- Car_Total[,"Pay_Meth", drop = FALSE]
asian_car_data <- subset(Car_Total, Region == "Asian")
american_car_data <- subset(Car_Total, Region == "American")
european_car_data <- subset(Car_Total, Region == "European")
middleEastern_car_data <- subset(Car_Total, Region == "Middle Eastern")
# Filter Data for Honda, Toyota, and Ford
Honda_Data <-Car_Total %>% filter(grepl("Honda", Model))
# View(Honda_Data)
Toyota_Data <-Car_Total %>% filter(grepl("Toyota", Model))
# View(Toyota_Data)
Ford_Data <-Car_Total %>% filter(grepl("Ford", Model))
# View(Ford_Data)
HTF_Data<- bind_rows(Honda_Data, Toyota_Data, Ford_Data)
# View(HTF_Data)
# Change Numerical Data to Categorical
global_car_data <- global_car_data %>% mutate(Pay_Meth = case_when
(Pay_Meth == 1 ~ "Lease", Pay_Meth == 2 ~ "Finance", Pay_Meth == 3 ~ "Cash"))
asian_car_data <- asian_car_data %>% mutate(Pay_Meth = case_when
(Pay_Meth == 1 ~ "Lease", Pay_Meth == 2 ~ "Finance", Pay_Meth == 3 ~ "Cash"))
american_car_data <- american_car_data %>% mutate(Pay_Meth = case_when
(Pay_Meth == 1 ~ "Lease", Pay_Meth == 2 ~ "Finance", Pay_Meth == 3 ~ "Cash"))
european_car_data <- european_car_data %>% mutate(Pay_Meth = case_when
(Pay_Meth == 1 ~ "Lease", Pay_Meth == 2 ~ "Finance", Pay_Meth == 3 ~ "Cash"))
middleEastern_car_data <- middleEastern_car_data %>% mutate(Pay_Meth = case_when
(Pay_Meth == 1 ~ "Lease", Pay_Meth == 2 ~ "Finance", Pay_Meth == 3 ~ "Cash"))
Honda_Data <- Honda_Data %>% mutate(Pay_Meth = case_when
(Pay_Meth == 1 ~ "Lease", Pay_Meth == 2 ~ "Finance", Pay_Meth == 3 ~ "Cash"))
# Filter Honda data by region
honda_asian_car_data <- subset(Honda_Data, Region == "Asian")
honda_american_car_data <- subset(Honda_Data, Region == "American")
honda_european_car_data <- subset(Honda_Data, Region == "European")
honda_middleEastern_car_data <- subset(Honda_Data, Region == "Middle Eastern")Creation of plots for payment methods
# Bar chart for Global Payments
ggplot(global_car_data, aes(x = Pay_Meth)) +
geom_bar(fill = "lightgreen", color = "black") +
geom_text(stat = 'count', aes(label = ..count..),vjust = -.5 ) +
labs(title = "Payment Methods in Global Market",
x = "Payment Method",
y = "Count") +
custom_theme
# Global Honda Payment Methods
ggplot(Honda_Data, aes(x = Pay_Meth)) +
geom_bar(fill = "lightgreen", color = "black") +
geom_text(stat = 'count', aes(label = ..count..),vjust = -.5 ) +
labs(title = "Global Payment Methods for Honda",
x = "Payment Type",
y = "Count") +
custom_theme
# Bar Chart Asian Market
ggplot(asian_car_data, aes(x = Pay_Meth)) +
geom_bar(fill = "lightgreen", color = "black") +
geom_text(stat = 'count', aes(label = ..count..),vjust = -.5 ) +
labs(title = "Payment Methods in Asian Market",
x = "Payment Type",
y = "Count") +
custom_theme
# Honda Payment Methods in Asian Market
ggplot(honda_asian_car_data, aes(x = Pay_Meth)) +
geom_bar(fill = "lightgreen", color = "black") +
geom_text(stat = 'count', aes(label = ..count..),vjust = -.5 ) +
labs(title = "Payment Methods in Asian Market for Honda",
x = "Payment Type",
y = "Count") +
custom_theme
# Bar Chart American Market
ggplot(american_car_data, aes(x = Pay_Meth)) +
geom_bar(fill = "lightgreen", color = "black") +
geom_text(stat = 'count', aes(label = ..count..),vjust = -.5 ) +
labs(title = "Payment Methods in American Market",
x = "Payment Type",
y = "Count") +
custom_theme
# Honda Payment Methods in American Market
ggplot(honda_american_car_data, aes(x = Pay_Meth)) +
geom_bar(fill = "lightgreen", color = "black") +
geom_text(stat = 'count', aes(label = ..count..),vjust = -.5 ) +
labs(title = "Payment Methods in American Market for Honda",
x = "Payment Type",
y = "Count") +
custom_theme
# Bar Chart European Market
ggplot(european_car_data, aes(x = Pay_Meth)) +
geom_bar(fill = "lightgreen", color = "black") +
geom_text(stat = 'count', aes(label = ..count..),vjust = -.5 ) +
labs(title = "Payment Methods in European Market",
x = "Payment Type",
y = "Count") +
custom_theme
# Honda Payment Methods in European Market
ggplot(honda_european_car_data, aes(x = Pay_Meth)) +
geom_bar(fill = "lightgreen", color = "black") +
geom_text(stat = 'count', aes(label = ..count..),vjust = -.5 ) +
labs(title = "Payment Methods in European Market for Honda",
x = "Payment Type",
y = "Count") +
custom_theme
# Bar Chart Middle Eastern Market
ggplot(middleEastern_car_data, aes(x = Pay_Meth)) +
geom_bar(fill = "lightgreen", color = "black") +
geom_text(stat = 'count', aes(label = ..count..),vjust = -.5 ) +
labs(title = "Payment Methods in Middle Eastern Market",
x = "Payment Type",
y = "Count") +
custom_theme
# Honda Payment Methods in Middle Eastern Market
ggplot(honda_middleEastern_car_data, aes(x = Pay_Meth)) +
geom_bar(fill = "lightgreen", color = "black") +
geom_text(stat = 'count', aes(label = ..count..),vjust = -.5 ) +
labs(title = "Payment Methods in Middle Eastern Market for Honda",
x = "Payment Type",
y = "Count") +
custom_theme
Difference in Middle Eastern Market for Honda
All other markets but Middle Eastern did not have significant observable differences in payment methods used between Honda customers and rest of market
Statistical test to see if observed difference in Middle Eastern Market is statistically significant using chi-squared test
# Create a new column to categorize Honda and Other cars
middleEastern_car_data <- middleEastern_car_data %>%
mutate(isHonda = case_when(
Model == "Honda Pilot" ~ "Honda",
Model == "Honda CRV" ~ "Honda",
TRUE ~ "Other"
))
# Create a contingency table
contingency_table <- table(middleEastern_car_data$isHonda, middleEastern_car_data$Pay_Meth)
print(contingency_table)
# Perform the chi-squared test of independence
# Is there a significant difference in payment methods for Honda in the Middle Eastern market
# H0: There is no difference in payment methods for Honda in the Middle Eastern market
chi_squared_test <- chisq.test(contingency_table)
print(chi_squared_test)##
## Pearson's Chi-squared test
##
## data: contingency_table
## X-squared = 75.669, df = 2, p-value < 2.2e-16Result was found to be statistically significant meaning there is a difference in the payment methods used by Honda customers and the rest of the Middle Eastern market
Age
Filter and plot age distributions for Honda, Toyota, and Ford
# Filter data for Honda models and plot Age Distribution by Region
Honda_Data <- Car_Total %>% filter(grepl("Honda", Model))
ggplot(Honda_Data, aes(x = Age, fill = Region)) +
geom_histogram(binwidth = 5, position = "dodge") +
labs(title = "Age Distribution by Region (Honda)", x = "Age", y = "Count") +
theme_minimal()
# Filter data for Toyota models
Toyota_Data <- Car_Total %>% filter(grepl("Toyota", Model))
# Plot Age Distribution by Region for Toyota models
ggplot(Toyota_Data, aes(x = Age, fill = Region)) +
geom_histogram(binwidth = 5, position = "dodge") +
labs(title = "Age Distribution by Region (Toyota)", x = "Age", y = "Count") +
theme_minimal()
# Filter data for Ford models
Ford_Data <- Car_Total %>% filter(grepl("Ford", Model))
# Plot Age Distribution by Region for Ford models
ggplot(Ford_Data, aes(x = Age, fill = Region)) +
geom_histogram(binwidth = 5, position = "dodge") +
labs(title = "Age Distribution by Region (Ford)", x = "Age", y = "Count") +
theme_minimal()
Insurance Type
Filter and plot insurance types for Honda, Toyota, and Ford
# Visual interpretation of insurance type within in each region based on Honda, Toyota and Ford models
library(ggplot2)
Insur_Honda <-ggplot(Honda_Data, aes(x=Region, fill=Insur_Type)) +geom_bar(position = "fill")+
labs(title = "Insurance Types for Honda Owners", x="Region", fill="Insurance Type") +
scale_fill_manual(values = c("Collision" = "steelblue", "Comprehensive" = "powderblue", "Liability" = "navy", "N/A"="Black" ))
Insur_Honda
Insur_Toyota <-ggplot(Toyota_Data, aes(x=Region, fill=Insur_Type)) +geom_bar(position = "fill")+
labs(title = "Insurance Types for Toyota Owners", x="Region", fill="Insurance Type") +
scale_fill_manual(values = c("Collision" = "steelblue", "Comprehensive" = "powderblue", "Liability" = "navy", "N/A"="Black" ))
Insur_Toyota
Insur_Ford <-ggplot(Ford_Data, aes(x=Region, fill=Insur_Type)) +geom_bar(position = "fill")+
labs(title = "Insurance Types for Ford Owners", x="Region", fill="Insurance Type") +
scale_fill_manual(values = c("Collision" = "steelblue", "Comprehensive" = "powderblue", "Liability" = "navy", "N/A"="Black" ))
Insur_Ford
Relationships from Survey Data
Post-Purchase Satisfaction and Word of Mouth 1 & 2
Testing for relationship between Post-Purchase Satisfaction and Word of Mouth 1 & 2 using Pearson correlation test with 95% confidence interval
Global and Regional Data used for comparison
# Plot Post-Purchase Satisfaction vs. Word of Mouth 1 & 2 for whole car market and run Pearson correlation test
plot(Car_Total$`Post-Satis`,Car_Total$WOM_1)
plot(Car_Total$`Post-Satis`,Car_Total$WOM_2)
cor.test(Car_Total$`Post-Satis`,Car_Total$WOM_1, alternative = "two.sided", conf.level = .95) # insignificant result
cor.test(Car_Total$`Post-Satis`,Car_Total$WOM_2, alternative = "two.sided", conf.level = .95) # insignificant result
# Plot Post-Purchase Satisfaction vs. Word of Mouth 1 & 2 for Honda globally and run Pearson correlation test
plot(Honda_Data$`Post-Satis`,Honda_Data$WOM_1)
plot(Honda_Data$`Post-Satis`,Honda_Data$WOM_2)
cor.test(Honda_Data$`Post-Satis`,Honda_Data$WOM_1, alternative = "two.sided", conf.level = .95) # significant result: -0.249722
cor.test(Honda_Data$`Post-Satis`,Honda_Data$WOM_2, alternative = "two.sided", conf.level = .95) # significant result: -0.164312
# Pearson correlation test for Post-Purchase Satisfaction vs. Word of Mouth 1 & 2 for American Market
cor.test(american_car_data$`Post-Satis`,american_car_data$WOM_1, alternative = "two.sided", conf.level = .95) # insignificant result
cor.test(american_car_data$`Post-Satis`,american_car_data$WOM_2, alternative = "two.sided", conf.level = .95) # insignificant result
# Plot Post-Purchase Satisfaction vs. Word of Mouth 1 & 2 for Honda in American market and run Pearson correlation test
plot(honda_american_car_data$`Post-Satis`,honda_american_car_data$WOM_1)
plot(honda_american_car_data$`Post-Satis`,honda_american_car_data$WOM_2)
cor.test(honda_american_car_data$`Post-Satis`,honda_american_car_data$WOM_1, alternative = "two.sided", conf.level = .95) # insignificant result
cor.test(honda_american_car_data$`Post-Satis`,honda_american_car_data$WOM_2, alternative = "two.sided", conf.level = .95) # insignificant result
# Pearson correlation test for Post-Purchase Satisfaction vs. Word of Mouth 1 & 2 for Asian Market
cor.test(asian_car_data$`Post-Satis`,asian_car_data$WOM_1, alternative = "two.sided", conf.level = .95) # significant result: -0.1588893
cor.test(asian_car_data$`Post-Satis`,asian_car_data$WOM_2, alternative = "two.sided", conf.level = .95) # significant result: -0.1780611
# Plot Post-Purchase Satisfaction vs. Word of Mouth 1 & 2 for Honda in Asian market and run Pearson correlation test
plot(honda_asian_car_data$`Post-Satis`,honda_asian_car_data$WOM_1)
plot(honda_asian_car_data$`Post-Satis`,honda_asian_car_data$WOM_2)
cor.test(honda_asian_car_data$`Post-Satis`,honda_asian_car_data$WOM_1, alternative = "two.sided", conf.level = .95) # significant result: -0.4003095
cor.test(honda_asian_car_data$`Post-Satis`,honda_asian_car_data$WOM_2, alternative = "two.sided", conf.level = .95) # significant result: -0.3489047
# Pearson correlation test for Post-Purchase Satisfaction vs. Word of Mouth 1 & 2 for European Market
cor.test(european_car_data$`Post-Satis`,european_car_data$WOM_1, alternative = "two.sided", conf.level = .95) # insignificant result
cor.test(european_car_data$`Post-Satis`,european_car_data$WOM_2, alternative = "two.sided", conf.level = .95) # insignificant result
# Plot Post-Purchase Satisfaction vs. Word of Mouth 1 & 2 for Honda in European market and run Pearson correlation test
plot(honda_european_car_data$`Post-Satis`,honda_european_car_data$WOM_1)
plot(honda_european_car_data$`Post-Satis`,honda_european_car_data$WOM_2)
cor.test(honda_european_car_data$`Post-Satis`,honda_european_car_data$WOM_1, alternative = "two.sided", conf.level = .95) # insignificant result
cor.test(honda_european_car_data$`Post-Satis`,honda_european_car_data$WOM_2, alternative = "two.sided", conf.level = .95) # insignificant result
# Pearson correlation test for Post-Purchase Satisfaction vs. Word of Mouth 1 & 2 for Middle Eastern Market
cor.test(middleEastern_car_data$`Post-Satis`,middleEastern_car_data$WOM_1, alternative = "two.sided", conf.level = .95) # insignificant result
cor.test(middleEastern_car_data$`Post-Satis`,middleEastern_car_data$WOM_2, alternative = "two.sided", conf.level = .95) # insignificant result
# Plot Post-Purchase Satisfaction vs. Word of Mouth 1 & 2 for Honda in Middle Eastern market and run Pearson correlation test
plot(honda_middleEastern_car_data$`Post-Satis`,honda_middleEastern_car_data$WOM_1)
plot(honda_middleEastern_car_data$`Post-Satis`,honda_middleEastern_car_data$WOM_2)
cor.test(honda_middleEastern_car_data$`Post-Satis`,honda_middleEastern_car_data$WOM_1, alternative = "two.sided", conf.level = .95) # insignificant result
cor.test(honda_middleEastern_car_data$`Post-Satis`,honda_middleEastern_car_data$WOM_2, alternative = "two.sided", conf.level = .95) # insignificant resultSignificant results were found between:
Post-Purchase satisfaction vs. Word-of-Mouth 1 & 2 for Honda globally
cor.test(Honda_Data$`Post-Satis`,Honda_Data$WOM_1, alternative = "two.sided", conf.level = .95)##
## Pearson's product-moment correlation
##
## data: Honda_Data$`Post-Satis` and Honda_Data$WOM_1
## t = -3.2314, df = 157, p-value = 0.001501
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.39020263 -0.09787924
## sample estimates:
## cor
## -0.249722cor.test(Honda_Data$`Post-Satis`,Honda_Data$WOM_2, alternative = "two.sided", conf.level = .95)##
## Pearson's product-moment correlation
##
## data: Honda_Data$`Post-Satis` and Honda_Data$WOM_2
## t = -2.0872, df = 157, p-value = 0.03849
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.31198037 -0.00889217
## sample estimates:
## cor
## -0.164312Post-Purchase satisfaction vs. Word-of-Mouth 1 & 2 in Asian Market
cor.test(asian_car_data$`Post-Satis`,asian_car_data$WOM_1, alternative = "two.sided", conf.level = .95)##
## Pearson's product-moment correlation
##
## data: asian_car_data$`Post-Satis` and asian_car_data$WOM_1
## t = -2.3098, df = 206, p-value = 0.02189
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.28869023 -0.02335293
## sample estimates:
## cor
## -0.1588893cor.test(asian_car_data$`Post-Satis`,asian_car_data$WOM_2, alternative = "two.sided", conf.level = .95)##
## Pearson's product-moment correlation
##
## data: asian_car_data$`Post-Satis` and asian_car_data$WOM_2
## t = -2.5972, df = 206, p-value = 0.01008
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.30667347 -0.04306309
## sample estimates:
## cor
## -0.1780611Post-Purchase satisfaction vs. Word-of-Mouth 1 & 2 for Honda in Asian Market
cor.test(honda_asian_car_data$`Post-Satis`,honda_asian_car_data$WOM_1, alternative = "two.sided", conf.level = .95)##
## Pearson's product-moment correlation
##
## data: honda_asian_car_data$`Post-Satis` and honda_asian_car_data$WOM_1
## t = -3.269, df = 56, p-value = 0.001848
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.5968881 -0.1583911
## sample estimates:
## cor
## -0.4003095cor.test(honda_asian_car_data$`Post-Satis`,honda_asian_car_data$WOM_2, alternative = "two.sided", conf.level = .95)##
## Pearson's product-moment correlation
##
## data: honda_asian_car_data$`Post-Satis` and honda_asian_car_data$WOM_2
## t = -2.786, df = 56, p-value = 0.007269
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.55700303 -0.09958346
## sample estimates:
## cor
## -0.3489047Value Perception, Future Purchase, and Insurance
The relationship between Value Perception, Future Purchase, and Insurance
#Honda
futurep_value_relationH <- aov(Futu_Pur_2~as.factor(Valu_Percp_2), var.equal=FALSE, data = Honda_Data)
summary(futurep_value_relationH)## Df Sum Sq Mean Sq F value Pr(>F)
## as.factor(Valu_Percp_2) 6 81.89 13.649 9.727 4.66e-09 ***
## Residuals 152 213.29 1.403
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1TukeyHSD(futurep_value_relationH)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Futu_Pur_2 ~ as.factor(Valu_Percp_2), data = Honda_Data, var.equal = FALSE)
##
## $`as.factor(Valu_Percp_2)`
## diff lwr upr p adj
## 2-1 0.6666667 -2.22356411 3.556897 0.9930323
## 3-1 1.3636364 -0.94197020 3.669243 0.5720513
## 4-1 1.6521739 -0.52072879 3.825077 0.2644416
## 5-1 2.3333333 0.18988082 4.476786 0.0232853
## 6-1 2.8253968 0.73360136 4.917192 0.0016360
## 7-1 3.2692308 1.11084112 5.427620 0.0002398
## 3-2 0.6969697 -1.60863686 3.002576 0.9715658
## 4-2 0.9855072 -1.18739545 3.158410 0.8243383
## 5-2 1.6666667 -0.47678585 3.810119 0.2396967
## 6-2 2.1587302 0.06693470 4.250526 0.0382637
## 7-2 2.6025641 0.44417446 4.760954 0.0076061
## 4-3 0.2885375 -1.00911179 1.586187 0.9942864
## 5-3 0.9696970 -0.27801146 2.217405 0.2402255
## 6-3 1.4617605 0.30504273 2.618478 0.0041956
## 7-3 1.9055944 0.63239626 3.178793 0.0002985
## 5-4 0.6811594 -0.29989148 1.662210 0.3729822
## 6-4 1.1732229 0.31085382 2.035592 0.0014668
## 7-4 1.6170569 0.60378589 2.630328 0.0000869
## 6-5 0.4920635 -0.29315175 1.277279 0.5016308
## 7-5 0.9358974 -0.01257533 1.884370 0.0556988
## 7-6 0.4438339 -0.38128433 1.268952 0.6778066#Toyota
futurep_value_relationT <- aov(Futu_Pur_2~as.factor(Valu_Percp_2), var.equal=FALSE, data = Toyota_Data)
summary(futurep_value_relationT)## Df Sum Sq Mean Sq F value Pr(>F)
## as.factor(Valu_Percp_2) 7 40.9 5.845 3.592 0.00101 **
## Residuals 284 462.1 1.627
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1TukeyHSD(futurep_value_relationT)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Futu_Pur_2 ~ as.factor(Valu_Percp_2), data = Toyota_Data, var.equal = FALSE)
##
## $`as.factor(Valu_Percp_2)`
## diff lwr upr p adj
## 2-1 0.65000000 -1.2975780 2.5975780 0.9712878
## 3-1 1.00000000 -0.9475780 2.9475780 0.7690474
## 4-1 0.55510204 -1.2735821 2.3837862 0.9832487
## 5-1 1.37500000 -0.4205782 3.1705782 0.2765967
## 5.11354961832061-1 1.17058262 -3.0963470 5.4375122 0.9907821
## 6-1 1.44044944 -0.3497803 3.2306792 0.2185226
## 7-1 1.15714286 -0.7339716 3.0482573 0.5737307
## 3-2 0.35000000 -0.8817565 1.5817565 0.9886310
## 4-2 -0.09489796 -1.1284595 0.9386636 0.9999932
## 5-2 0.72500000 -0.2487890 1.6987890 0.3119724
## 5.11354961832061-2 0.52058262 -3.4707645 4.5119298 0.9999250
## 6-2 0.79044944 -0.1734420 1.7543408 0.1978779
## 7-2 0.50714286 -0.6332420 1.6475278 0.8752696
## 4-3 -0.44489796 -1.4784595 0.5886636 0.8928696
## 5-3 0.37500000 -0.5987890 1.3487890 0.9381923
## 5.11354961832061-3 0.17058262 -3.8207645 4.1619298 1.0000000
## 6-3 0.44044944 -0.5234420 1.4043408 0.8588937
## 7-3 0.15714286 -0.9832420 1.2975278 0.9998913
## 5-4 0.81989796 0.1132930 1.5265029 0.0107726
## 5.11354961832061-4 0.61548058 -3.3192212 4.5501823 0.9997457
## 6-4 0.88534740 0.1924461 1.5782487 0.0029730
## 7-4 0.60204082 -0.3207285 1.5248102 0.4889046
## 5.11354961832061-5 -0.20441738 -4.1238425 3.7150077 0.9999999
## 6-5 0.06544944 -0.5346569 0.6655558 0.9999778
## 7-5 -0.21785714 -1.0731458 0.6374315 0.9941255
## 6-5.11354961832061 0.26986682 -3.6471109 4.1868446 0.9999991
## 7-5.11354961832061 -0.01343976 -3.9775419 3.9506624 1.0000000
## 7-6 -0.28330658 -1.1273091 0.5606960 0.9703569#Ford
futurep_value_relationF <- aov(Futu_Pur_2~as.factor(Valu_Percp_2), var.equal=FALSE, data = Ford_Data)
summary(futurep_value_relationF)## Df Sum Sq Mean Sq F value Pr(>F)
## as.factor(Valu_Percp_2) 6 96.2 16.033 11.19 1.02e-10 ***
## Residuals 195 279.5 1.434
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1TukeyHSD(futurep_value_relationF)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Futu_Pur_2 ~ as.factor(Valu_Percp_2), data = Ford_Data, var.equal = FALSE)
##
## $`as.factor(Valu_Percp_2)`
## diff lwr upr p adj
## 2-1 -1.31274541 -3.971566481 1.3460757 0.7616514
## 3-1 1.22222222 -1.566372843 4.0108173 0.8485874
## 4-1 -0.05882353 -2.725453196 2.6078061 1.0000000
## 5-1 0.40000000 -2.193445981 2.9934460 0.9992790
## 6-1 0.77108434 -1.781504016 3.3236727 0.9722118
## 7-1 1.23684211 -1.351064291 3.8247485 0.7883174
## 3-2 2.53496763 1.078671355 3.9912639 0.0000110
## 4-2 1.25392188 0.047501478 2.4603423 0.0358307
## 5-2 1.71274541 0.678096453 2.7473944 0.0000355
## 6-2 2.08382975 1.156336508 3.0113230 0.0000000
## 7-2 2.54958752 1.528903468 3.5702716 0.0000000
## 4-3 -1.28104575 -2.751550134 0.1894586 0.1330367
## 5-3 -0.82222222 -2.155425663 0.5109812 0.5244864
## 6-3 -0.45113788 -1.703007150 0.8007314 0.9351015
## 7-3 0.01461988 -1.307775245 1.3370150 1.0000000
## 5-4 0.45882353 -0.595729815 1.5133769 0.8530070
## 6-4 0.82990787 -0.119738383 1.7795541 0.1305289
## 7-4 1.29566563 0.254810114 2.3365212 0.0049823
## 6-5 0.37108434 -0.347856743 1.0900254 0.7216399
## 7-5 0.83684211 0.001121762 1.6725624 0.0494478
## 7-6 0.46575777 -0.232936540 1.1644521 0.4268469#Anova test with two indepdendent factors using Insurance type, value perception and future purchase 2.
#Honda
futurerep_value_relation2<- aov(Futu_Pur_2~as.factor(Valu_Percp_2)
+as.factor(Insur_Type), var.equal=FALSE, data = Honda_Data)
summary(futurerep_value_relation2)## Df Sum Sq Mean Sq F value Pr(>F)
## as.factor(Valu_Percp_2) 6 80.00 13.333 9.372 1.02e-08 ***
## as.factor(Insur_Type) 2 0.75 0.376 0.264 0.768
## Residuals 149 211.98 1.423
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 1 observation deleted due to missingness#Toyota
futurerep_value_relation2T<- aov(Futu_Pur_2~as.factor(Valu_Percp_2)
+as.factor(Insur_Type), var.equal=FALSE, data = Toyota_Data)
summary(futurerep_value_relation2T)## Df Sum Sq Mean Sq F value Pr(>F)
## as.factor(Valu_Percp_2) 7 40.6 5.800 3.552 0.00112 **
## as.factor(Insur_Type) 2 3.1 1.569 0.961 0.38374
## Residuals 281 458.8 1.633
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 1 observation deleted due to missingness#Ford
futurerep_value_relation2F<- aov(Futu_Pur_2~as.factor(Valu_Percp_2)
+as.factor(Insur_Type), var.equal=FALSE, data = Ford_Data)
summary(futurerep_value_relation2F)## Df Sum Sq Mean Sq F value Pr(>F)
## as.factor(Valu_Percp_2) 6 86.82 14.469 10.055 1.19e-09 ***
## as.factor(Insur_Type) 2 1.07 0.536 0.372 0.69
## Residuals 191 274.85 1.439
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 2 observations deleted due to missingness