DATA 606 Final Project: World Happiness Report
Kelly Shaffer
Part 1 - Introduction
The first World Happiness Report (WHR) was published in 2012. Each year, ~1,000 individuals are sampled from each of more than 150 countries. The following question is asked of each individual:
“Please imagine a ladder, with steps numbered from 0 at the bottom to 10 at the top. The top of the ladder represents the best possible life for you. On which step of the ladder would you say you personally feel you stand at this time?”
Six key variables are then calculated to assist with the explanation of the country’s happiness score. The six variables are as follows:
Freedom, generosity, health, social support, income, trustworthy governance.
I’m going to dig into the 2015 and 2016 data to see what I can find. I suspect the happiness scores will be different by region, but I am not hypothesizing that they changed in one direction or the other. I am simply interested in the movement.
Part 2 - Data
We’ll begin by loading the data into our environment and assigning appropriate column names.
#install.packages("rio")
#install.packages("RCurl")
#install.packages("bitops")
library(rio)
library(RCurl)
library(bitops)
#load 2015 data
x <- getURL("https://raw.githubusercontent.com/excelsiordata/DATA606/master/2015.csv")
WHR2015 <- read.csv(text = x, head=TRUE, sep=",", stringsAsFactors=FALSE, col.names = c("Country","Region","Happiness Rank","Happiness Score","Standard Error","Economy (GDP per Capita)","Family","Health (Life Expectancy)","Freedom","Trust (Government Corruption)","Generosity","Dystopia Residual"))
#load 2016 data
x2 <- getURL("https://raw.githubusercontent.com/excelsiordata/DATA606/master/2016.csv")
WHR2016 <- read.csv(text = x2, head=TRUE, sep=",", stringsAsFactors=FALSE, col.names = c("Country","Region","Happiness Rank","Happiness Score","Standard Error","Economy (GDP per Capita)","Family","Health (Life Expectancy)","Freedom","Trust (Government Corruption)","Generosity","Dystopia Residual"))Research question
You should phrase your research question in a way that matches up with the scope of inference your dataset allows for.
How have the average happiness scores changed by region between 2015 and 2016?
Cases
What are the cases, and how many are there?
Each case represents a country, and there are 158 countries in the 2015 report. There are 156 countries in the 2016 Report.
Data collection
Describe the method of data collection.
The World Happiness Report was created from the Gallup World Poll data. The Gallup data is collected through surveys done globally either face to face or over the phone.
Type of study
What type of study is this (observational/experiment)?
Being that the data was collected through a survey, this is an observational study. There was no manipulation of any variables, etc.
Data Source
If you collected the data, state self-collected. If not, provide a citation/link.
The World Happiness Report data can be found here: https://www.kaggle.com/unsdsn/world-happiness
Response
What is the response variable, and what type is it (numerical/categorical)?
Happiness score is the response variable, and it is numerical.
Explanatory
What is the explanatory variable, and what type is it (numerical/categorical)?
Country is the explanatory variable, and it is categorical.
Part 3 - Exploratory Data Analysis (code portion):
summary(WHR2015$Happiness.Score)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.839 4.526 5.232 5.376 6.244 7.587mean(WHR2015$Happiness.Score)## [1] 5.375734var(WHR2015$Happiness.Score)## [1] 1.311048median(WHR2015$Happiness.Score)## [1] 5.2325sd(WHR2015$Happiness.Score)## [1] 1.14501plot(WHR2015$Happiness.Score, main = "2015 Happiness Score by Frequency", xlab = "Frequency", ylab = "Happiness Score")
summary(WHR2016$Happiness.Score)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.905 4.404 5.314 5.382 6.269 7.526mean(WHR2016$Happiness.Score)## [1] 5.382185var(WHR2016$Happiness.Score)## [1] 1.303418median(WHR2016$Happiness.Score)## [1] 5.314sd(WHR2016$Happiness.Score)## [1] 1.141674plot(WHR2016$Happiness.Score, main = "2016 Happiness Score by Frequency", xlab = "Frequency", ylab = "Happiness Score")
head(WHR2015)## Country Region Happiness.Rank Happiness.Score Standard.Error
## 1 Switzerland Western Europe 1 7.587 1.39651
## 2 Iceland Western Europe 2 7.561 1.30232
## 3 Denmark Western Europe 3 7.527 1.32548
## 4 Norway Western Europe 4 7.522 1.45900
## 5 Canada North America 5 7.427 1.32629
## 6 Finland Western Europe 6 7.406 1.29025
## Economy..GDP.per.Capita. Family Health..Life.Expectancy. Freedom
## 1 1.34951 0.94143 0.66557 0.41978
## 2 1.40223 0.94784 0.62877 0.14145
## 3 1.36058 0.87464 0.64938 0.48357
## 4 1.33095 0.88521 0.66973 0.36503
## 5 1.32261 0.90563 0.63297 0.32957
## 6 1.31826 0.88911 0.64169 0.41372
## Trust..Government.Corruption. Generosity Dystopia.Residual
## 1 0.29678 2.51738 NA
## 2 0.43630 2.70201 NA
## 3 0.34139 2.49204 NA
## 4 0.34699 2.46531 NA
## 5 0.45811 2.45176 NA
## 6 0.23351 2.61955 NAhead(WHR2016)## Country Region Happiness.Rank Happiness.Score Standard.Error
## 1 Denmark Western Europe 1 7.526 1.44178
## 2 Switzerland Western Europe 2 7.509 1.52733
## 3 Iceland Western Europe 3 7.501 1.42666
## 4 Norway Western Europe 4 7.498 1.57744
## 5 Finland Western Europe 5 7.413 1.40598
## 6 Canada North America 6 7.404 1.44015
## Economy..GDP.per.Capita. Family Health..Life.Expectancy. Freedom
## 1 1.16374 0.79504 0.57941 0.44453
## 2 1.14524 0.86303 0.58557 0.41203
## 3 1.18326 0.86733 0.56624 0.14975
## 4 1.12690 0.79579 0.59609 0.35776
## 5 1.13464 0.81091 0.57104 0.41004
## 6 1.09610 0.82760 0.57370 0.31329
## Trust..Government.Corruption. Generosity Dystopia.Residual
## 1 0.36171 2.73939 NA
## 2 0.28083 2.69463 NA
## 3 0.47678 2.83137 NA
## 4 0.37895 2.66465 NA
## 5 0.25492 2.82596 NA
## 6 0.44834 2.70485 NAtable(WHR2016$Region)##
## Australia and New Zealand Central and Eastern Europe
## 2 29
## Eastern Asia Latin America and Caribbean
## 6 24
## Middle East and Northern Africa North America
## 19 2
## Southeastern Asia Southern Asia
## 9 7
## Sub-Saharan Africa Western Europe
## 38 21hist(WHR2015$Happiness.Score, main = "Histogram of Happiness Score Worldwide in 2015", xlab = "Happiness Score")
hist(WHR2016$Happiness.Score, main = "Histogram of Happiness Score Worldwide in 2016", xlab = "Happiness Score")
boxplot(WHR2015$Happiness.Score ~ WHR2015$Region, xlab = "Region", ylab = "Happiness Score", main = "Boxplot of Happiness Score by Region for 2015")
boxplot(WHR2016$Happiness.Score ~ WHR2016$Region, xlab = "Region", ylab = "Happiness Score", main = "Boxplot of Happiness Score by Region for 2016")
#I noticed here that there was a huge shift in Sub-Saharan Africa between 2015 and 2016
#This is what made me choose to dig into it below
#Dig into the Sub-Saharan Africa data
SSA2015 <- subset(WHR2015, Region == "Sub-Saharan Africa")
SSA2016 <- subset(WHR2016, Region == "Sub-Saharan Africa")
summary(SSA2015)## Country Region Happiness.Rank Happiness.Score
## Length:40 Length:40 Min. : 71.0 Min. :2.839
## Class :character Class :character 1st Qu.:115.8 1st Qu.:3.756
## Mode :character Mode :character Median :132.0 Median :4.272
## Mean :127.9 Mean :4.203
## 3rd Qu.:146.2 3rd Qu.:4.581
## Max. :158.0 Max. :5.477
## Standard.Error Economy..GDP.per.Capita. Family
## Min. :0.0000 Min. :0.0000 Min. :0.0000
## 1st Qu.:0.2039 1st Qu.:0.6767 1st Qu.:0.1651
## Median :0.3084 Median :0.8784 Median :0.2982
## Mean :0.3805 Mean :0.8091 Mean :0.2823
## 3rd Qu.:0.4831 3rd Qu.:1.0016 3rd Qu.:0.3721
## Max. :1.0602 Max. :1.1847 Max. :0.7095
## Health..Life.Expectancy. Freedom Trust..Government.Corruption.
## Min. :0.1008 Min. :0.03060 Min. :0.06822
## 1st Qu.:0.3013 1st Qu.:0.07231 1st Qu.:0.18260
## Median :0.3829 Median :0.10387 Median :0.20730
## Mean :0.3659 Mean :0.12388 Mean :0.22114
## 3rd Qu.:0.4620 3rd Qu.:0.13289 3rd Qu.:0.24334
## Max. :0.5920 Max. :0.55191 Max. :0.50318
## Generosity Dystopia.Residual
## Min. :0.6704 Mode:logical
## 1st Qu.:1.6693 NA's:40
## Median :1.9501
## Mean :2.0200
## 3rd Qu.:2.4583
## Max. :3.0514summary(SSA2016)## Country Region Happiness.Rank Happiness.Score
## Length:38 Length:38 Min. : 66.0 Min. :2.905
## Class :character Class :character 1st Qu.:117.5 1st Qu.:3.745
## Mode :character Mode :character Median :133.5 Median :4.130
## Mean :129.7 Mean :4.136
## 3rd Qu.:144.8 3rd Qu.:4.433
## Max. :157.0 Max. :5.648
## Standard.Error Economy..GDP.per.Capita. Family
## Min. :0.0000 Min. :0.0000 Min. :0.0000
## 1st Qu.:0.2800 1st Qu.:0.4819 1st Qu.:0.1605
## Median :0.3945 Median :0.6312 Median :0.2419
## Mean :0.4743 Mean :0.5937 Mean :0.2399
## 3rd Qu.:0.6265 3rd Qu.:0.7615 3rd Qu.:0.3144
## Max. :1.1585 Max. :0.9605 Max. :0.6619
## Health..Life.Expectancy. Freedom Trust..Government.Corruption.
## Min. :0.0000 Min. :0.03050 Min. :0.06244
## 1st Qu.:0.2551 1st Qu.:0.06682 1st Qu.:0.18328
## Median :0.3402 Median :0.09586 Median :0.21621
## Mean :0.3154 Mean :0.12038 Mean :0.22635
## 3rd Qu.:0.4132 3rd Qu.:0.12895 3rd Qu.:0.25789
## Max. :0.5678 Max. :0.50521 Max. :0.51479
## Generosity Dystopia.Residual
## Min. :0.9674 Mode:logical
## 1st Qu.:1.9384 NA's:38
## Median :2.1231
## Mean :2.1664
## 3rd Qu.:2.4415
## Max. :3.8377#2015
#all of these pieces are within the summary, but I wanted to pull them out
mean(SSA2015$Happiness.Score)## [1] 4.2028range(SSA2015$Happiness.Score)## [1] 2.839 5.477range(SSA2015$Happiness.Rank)## [1] 71 158hist(SSA2015$Happiness.Score, main = "Histogram of Happiness Score in Sub-Saharan Africa in 2015", xlab = "Happiness Score", breaks = 40)
boxplot(SSA2015$Happiness.Score ~ SSA2015$Country, main = "Boxplot of Happiness Score in Sub-Saharan Africa in 2015", xlab = "Country", ylab = "Happiness Score")
plot(SSA2015$Happiness.Score ~ SSA2015$Trust..Government.Corruption., main = "Scatterplot of Happiness Score and Trust in 2015 Sub-Saharan Africa", xlab = "Trust", ylab = "Happiness Score")
plot(SSA2015$Happiness.Score ~ SSA2015$Family, main = "Scatterplot of Happiness Score and Family in 2015 Sub-Saharan Africa", xlab = "Family", ylab = "Happiness Score")
plot(SSA2015$Happiness.Score ~ SSA2015$Health..Life.Expectancy., main = "Scatterplot of Happiness Score and Health in 2015 Sub-Saharan Africa", xlab = "Health", ylab = "Happiness Score")
#These relationships look linear - proceeding with quantifying the
#strength of the correlation coefficient
plot(SSA2015$Happiness.Score ~ SSA2015$Economy..GDP.per.Capita., main = "Scatterplot of Happiness Score and Economy in 2015 Sub-Saharan Africa", xlab = "Economy", ylab = "Happiness Score")
plot(SSA2015$Happiness.Score ~ SSA2015$Generosity, main = "Scatterplot of Happiness Score and Generosity in 2015 Sub-Saharan Africa", xlab = "Generosity", ylab = "Happiness Score")
cor(SSA2015$Happiness.Score, SSA2015$Economy..GDP.per.Capita.)## [1] 0.5769741cor(SSA2015$Happiness.Score, SSA2015$Generosity)## [1] 0.5829218SSA2015e <- lm(Happiness.Score ~ Economy..GDP.per.Capita., data = SSA2015)
summary(SSA2015e)##
## Call:
## lm(formula = Happiness.Score ~ Economy..GDP.per.Capita., data = SSA2015)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.78143 -0.44288 -0.04344 0.40806 1.04291
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.1403 0.2567 12.234 9.49e-15 ***
## Economy..GDP.per.Capita. 1.3132 0.3016 4.355 9.73e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5044 on 38 degrees of freedom
## Multiple R-squared: 0.3329, Adjusted R-squared: 0.3153
## F-statistic: 18.96 on 1 and 38 DF, p-value: 9.728e-05plot(SSA2015$Happiness.Score ~ SSA2015$Economy..GDP.per.Capita., xlab = "Economy (GDP per Capita)", ylab = "Happiness Score", main = "Scatterplot with a LSR Line for 2015 Sub-Saharan Africa")
abline(SSA2015e)
SSA2015g <- lm(Happiness.Score ~ Generosity, data = SSA2015)
summary(SSA2015g)##
## Call:
## lm(formula = Happiness.Score ~ Generosity, data = SSA2015)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.17777 -0.23711 0.05247 0.23073 1.44017
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.9060 0.3038 9.566 1.15e-11 ***
## Generosity 0.6420 0.1452 4.422 7.91e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5018 on 38 degrees of freedom
## Multiple R-squared: 0.3398, Adjusted R-squared: 0.3224
## F-statistic: 19.56 on 1 and 38 DF, p-value: 7.913e-05plot(SSA2015$Happiness.Score ~ SSA2015$Generosity, xlab = "Generosity", ylab = "Happiness Score", main = "Scatterplot with a LSR Line for 2015 Sub-Saharan Africa")
abline(SSA2015g)
SSA2015le <- lm(Happiness.Score ~ Health..Life.Expectancy., data = SSA2015)
summary(SSA2015le)##
## Call:
## lm(formula = Happiness.Score ~ Health..Life.Expectancy., data = SSA2015)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.36308 -0.41911 0.02015 0.42122 1.17458
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.0156 0.3137 12.801 2.35e-15 ***
## Health..Life.Expectancy. 0.5116 0.8151 0.628 0.534
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6143 on 38 degrees of freedom
## Multiple R-squared: 0.01026, Adjusted R-squared: -0.01578
## F-statistic: 0.394 on 1 and 38 DF, p-value: 0.534plot(SSA2015$Happiness.Score ~ SSA2015$Health..Life.Expectancy., xlab = "Health (Life Expectancy)", ylab = "Happiness Score", main = "Scatterplot with a LSR Line for 2015 Sub-Saharan Africa")
abline(SSA2015le)
#Just double checking to make sure these are not any more significant than they look
#cor(SSA2015$Happiness.Score ~ SSA2015$Trust..Government.Corruption.)
#cor(SSA2015$Happiness.Score ~ SSA2015$Family)
#As I suspected, the correlation coefficient is really low
#Therefore, not moving forward in analyzing these variables
#2016
#all of these pieces are within the summary, but I wanted to pull them out
mean(SSA2016$Happiness.Score)## [1] 4.136421range(SSA2016$Happiness.Score)## [1] 2.905 5.648range(SSA2016$Happiness.Rank)## [1] 66 157hist(SSA2016$Happiness.Score, main = "Histogram of Happiness Score in Sub-Saharan Africa in 2016", xlab = "Happiness Score", breaks = 40)
boxplot(SSA2016$Happiness.Score ~ SSA2016$Country, main = "Boxplot of Happiness Score in Sub-Saharan Africa in 2016", xlab = "Country", ylab = "Happiness Score")
plot(SSA2016$Happiness.Score ~ SSA2016$Trust..Government.Corruption., main = "Scatterplot of Happiness Score and Trust in 2016 Sub-Saharan Africa", xlab = "Trust", ylab = "Happiness Score")
plot(SSA2016$Happiness.Score ~ SSA2016$Family, main = "Scatterplot of Happiness Score and Family in 2016 Sub-Saharan Africa", xlab = "Family", ylab = "Happiness Score")
plot(SSA2016$Happiness.Score ~ SSA2016$Health..Life.Expectancy., main = "Scatterplot of Happiness Score and Health in 2016 Sub-Saharan Africa", xlab = "Health", ylab = "Happiness Score")
#These relationships look linear - proceeding with quantifying the
#strength of the correlation coefficient
cor(SSA2016$Happiness.Score, SSA2016$Economy..GDP.per.Capita.)## [1] 0.3276954cor(SSA2016$Happiness.Score, SSA2016$Generosity)## [1] 0.493634SSA2016e <- lm(Happiness.Score ~ Economy..GDP.per.Capita., data = SSA2016)
summary(SSA2016e)##
## Call:
## lm(formula = Happiness.Score ~ Economy..GDP.per.Capita., data = SSA2016)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.9548 -0.3356 -0.1209 0.3183 1.5019
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6793 0.2362 15.578 <2e-16 ***
## Economy..GDP.per.Capita. 0.7700 0.3700 2.081 0.0446 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5349 on 36 degrees of freedom
## Multiple R-squared: 0.1074, Adjusted R-squared: 0.08259
## F-statistic: 4.331 on 1 and 36 DF, p-value: 0.0446plot(SSA2016$Happiness.Score ~ SSA2016$Economy..GDP.per.Capita., xlab = "Economy (GDP per Capita)", ylab = "Happiness Score", main = "Scatterplot with a LSR Line for 2016 Sub-Saharan Africa")
abline(SSA2016e)
SSA2016g <- lm(Happiness.Score ~ Generosity, data = SSA2016)
summary(SSA2016g)##
## Call:
## lm(formula = Happiness.Score ~ Generosity, data = SSA2016)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.1995 -0.2465 -0.0217 0.3110 1.4932
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.0280 0.3351 9.035 8.69e-11 ***
## Generosity 0.5117 0.1502 3.406 0.00164 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4924 on 36 degrees of freedom
## Multiple R-squared: 0.2437, Adjusted R-squared: 0.2227
## F-statistic: 11.6 on 1 and 36 DF, p-value: 0.001636plot(SSA2016$Happiness.Score ~ SSA2016$Generosity, xlab = "Generosity", ylab = "Happiness Score", main = "Scatterplot with a LSR Line for 2016 Sub-Saharan Africa")
abline(SSA2016g)
#Just double checking to make sure these are not any more significant than they look
#cor(SSA2016$Happiness.Score ~ SSA2016$Trust..Government.Corruption.)
#cor(SSA2016$Happiness.Score ~ SSA2016$Family)
#cor(SSA2016$Happiness.Score ~ SSA2016$Health..Life.Expectancy.)
#Life Expectancy is showing a much stronger linear relationship in 2016. Let's take a look.
SSA2016le <- lm(Happiness.Score ~ Health..Life.Expectancy., data = SSA2016)
summary(SSA2016le)##
## Call:
## lm(formula = Happiness.Score ~ Health..Life.Expectancy., data = SSA2016)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.97992 -0.37802 0.01915 0.27196 1.28174
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6400 0.2079 17.508 <2e-16 ***
## Health..Life.Expectancy. 1.5739 0.6026 2.612 0.0131 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5191 on 36 degrees of freedom
## Multiple R-squared: 0.1593, Adjusted R-squared: 0.1359
## F-statistic: 6.821 on 1 and 36 DF, p-value: 0.01305plot(SSA2016$Happiness.Score ~ SSA2016$Health..Life.Expectancy., xlab = "Health (Life Expectancy)", ylab = "Happiness Score", main = "Scatterplot with a LSR Line for 2016 Sub-Saharan Africa")
abline(SSA2016le)
summary(SSA2015$Health..Life.Expectancy.)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.1008 0.3013 0.3829 0.3659 0.4620 0.5920summary(SSA2016$Health..Life.Expectancy.)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.2551 0.3402 0.3154 0.4132 0.5678hist(SSA2015$Health..Life.Expectancy., breaks = 20, main = "Histogram of Health in 2015 Sub-Saharan Africa", xlab = "Health (Life Expectancy)")
hist(SSA2016$Health..Life.Expectancy., breaks = 20, main = "Histogram of Health in 2016 Sub-Saharan Africa", xlab = "Health (Life Expectancy)")
#Quick peek at eastern and central europe
ECE2015 <- subset(WHR2015, Region == "Central and Eastern Europe")
ECE2016 <- subset(WHR2016, Region == "Central and Eastern Europe")
summary(ECE2015)## Country Region Happiness.Rank Happiness.Score
## Length:29 Length:29 Min. : 31 Min. :4.218
## Class :character Class :character 1st Qu.: 59 1st Qu.:4.959
## Mode :character Mode :character Median : 77 Median :5.286
## Mean : 79 Mean :5.333
## 3rd Qu.: 95 3rd Qu.:5.813
## Max. :134 Max. :6.505
## Standard.Error Economy..GDP.per.Capita. Family
## Min. :0.3905 Min. :0.3856 Min. :0.5389
## 1st Qu.:0.8015 1st Qu.:0.9056 1st Qu.:0.6509
## Median :1.0122 Median :1.1061 Median :0.7313
## Mean :0.9424 Mean :1.0530 Mean :0.7188
## 3rd Qu.:1.1225 3rd Qu.:1.2279 3rd Qu.:0.7736
## Max. :1.1850 Max. :1.3404 Max. :0.8734
## Health..Life.Expectancy. Freedom Trust..Government.Corruption.
## Min. :0.09245 Min. :0.00227 Min. :0.00199
## 1st Qu.:0.25883 1st Qu.:0.02652 1st Qu.:0.10686
## Median :0.35068 Median :0.04212 Median :0.15275
## Mean :0.35827 Mean :0.08667 Mean :0.15226
## 3rd Qu.:0.44888 3rd Qu.:0.14296 3rd Qu.:0.20951
## Max. :0.65821 Max. :0.38331 Max. :0.30030
## Generosity Dystopia.Residual
## Min. :0.8999 Mode:logical
## 1st Qu.:1.7393 NA's:29
## Median :2.0250
## Mean :2.0214
## 3rd Qu.:2.2464
## Max. :3.1071summary(ECE2016)## Country Region Happiness.Rank Happiness.Score
## Length:29 Length:29 Min. : 27.00 Min. :4.217
## Class :character Class :character 1st Qu.: 60.00 1st Qu.:5.145
## Mode :character Mode :character Median : 74.00 Median :5.488
## Mean : 78.45 Mean :5.371
## 3rd Qu.: 91.00 3rd Qu.:5.813
## Max. :129.00 Max. :6.596
## Standard.Error Economy..GDP.per.Capita. Family
## Min. :0.4884 Min. :0.1925 Min. :0.4401
## 1st Qu.:0.9014 1st Qu.:0.7417 1st Qu.:0.5739
## Median :1.1131 Median :0.9316 Median :0.6408
## Mean :1.0475 Mean :0.8619 Mean :0.6316
## 3rd Qu.:1.2323 3rd Qu.:1.0469 3rd Qu.:0.6810
## Max. :1.3092 Max. :1.1681 Max. :0.7915
## Health..Life.Expectancy. Freedom Trust..Government.Corruption.
## Min. :0.09511 Min. :0.00000 Min. :0.02025
## 1st Qu.:0.19770 1st Qu.:0.03586 1st Qu.:0.09929
## Median :0.29091 Median :0.04762 Median :0.16840
## Mean :0.30053 Mean :0.08807 Mean :0.17090
## 3rd Qu.:0.40212 3rd Qu.:0.12721 3rd Qu.:0.22567
## Max. :0.60848 Max. :0.31880 Max. :0.38432
## Generosity Dystopia.Residual
## Min. :1.154 Mode:logical
## 1st Qu.:1.979 NA's:29
## Median :2.275
## Mean :2.270
## 3rd Qu.:2.493
## Max. :3.380Part 3 - Exploratory Data Analysis (text portion):
I began looking at the data by region, since there are too many countries to focus on. When I began to compare 2015 and 2016 by region, I noticed a shift in happiness score in Sub-Saharan Africa. Once I started looking into the relationships between happiness score and the 6 primary variables which calculate it, I noticed a linear relationship between Happiness Score and Economy (GDP per capita), and Happiness Score and Generosity for 2015. These results are reported below.
Sub-Saharan Africa:
2015:
Economy: Correlation coefficient: 0.5769741 Least squares regression line for the linear model: \[ \hat{y}(Sub-Saharan Africa 2015) = 3.1403 + 1.3132 * Economy (GDP per Capita) \]
\[ R^2 = 0.3329 \]
For this model, 33.29% of the variability in happiness score is explained by economy.Generosity: Correlation coefficient: 0.5829218 Least squares regression line for the linear model: \[ \hat{y}(Sub-Saharan Africa 2015) = 2.9060 + 0.6420 * Generosity \]
\[ R^2 = 0.3398 \]
For this model, 33.98% of the variability in happiness score is explained by generosity.Interestingly, when I looked at the same data for 2016, I discovered that Economy (GDP per Capita) was not as significant as it was in 2015, and Health (Life Expectancy) was moreso. These results are reported below.
2016
Economy: Correlation coefficient: 0.3276954 Least squares regression line for the linear model: \[ \hat{y}(Sub-Saharan Africa 2016) = 3.6793 + 0.77 * Economy (GDP per Capita) \]
\[ R^2 = 0.1074 \]
For this model, 10.74% of the variability in happiness score is explained by economy.Generosity: Correlation coefficient: 0.493634 Least squares regression line for the linear model: \[ \hat{y}(Sub-Saharan Africa 2016) = 3.0280 + 0.5117 * Generosity \]
\[ R^2 = 0.2437 \]
For this model, 24.37% of the variability in happiness score is explained by generosity.Health (Life Expectancy): Correlation coefficient: 0.3991246 Least squares regression line for the linear model: \[ \hat{y}(Sub-Saharan Africa 2016) = 3.64 + 1.5739 * Health (Life Expectancy) \]
\[ R^2 = 0.1593 \]
For this model, 15.93% of the variability in happiness score is explained by health (life expectancy).Another shift I noticed on the opposite end of the spectrum was an increase in happiness score across Central and Eastern Europe. Digging into an additional region is outside of the scope of this project, but a quick analysis found that the Economy (GDP per Capita) and Generosity variables increased between 2015 and 2016 across the region.
Part 4 - Inference:
Major crises have the power to impact the scores of the six key variables in any given country. Sub-Saharan Africa is no stranger to “major crises”, with some countries living in what seems to be a perpetual state of crisis. Some regions can bounce back from crisis, bonding amongst the chaos and coming together to rise above it all, producing increases in happiness scores despite the circumstances. It seems the challenges faced by many countries in Sub-Saharan Africa are too severe to rise above.
Part 5 - Conclusion:
Happiness score declined in Sub-Saharan Africa between 2015 and 2016, due to drastic swings in the scores of two variables; Economy (GDP per Capita) and Generosity. Given what we’ve seen between 2015 and 2016, it’s possible that the happiness scores will continue to decline across Sub-Saharan Africa in 2017 if drastic improvements aren’t made which have a direct impact on quality of life. Even so, drastic improvements don’t happen overnight, so it’s very possible this downward trend will continue.
References:
Helliwell, J., Layard, R., & Sachs, J. (2016). World Happiness Report 2016, Update (Vol. I). New York: Sustainable Development Solutions Network.