PSYC 5801: Assignment 1

I. Research Purpose and Questions

The purpose of this study is to explore cross-sectional associations between geographic region (i.e., Europe, Oceania, the Americas, Africa, and Asia), Gross Domestic Product (GDP) per capita, and life expectancy in 2007.

Research Questions:

1. Does life expectancy differ across world regions (i.e., at the continent level)?

2. Is GDP per capita associated with life expectancy?

3. Do world regions (i.e., at the continent level) and GDP per capita interact to predict life expectancy?

II. Define the Variables and Report Descriptive Statistics

Data Source

Data were taken from the Gapminder dataset. Given that the research questions are cross-sectional in nature, only the most recent data were taken (i.e., 2007 wave).

Variables and Descriptive Statistics

• Continent: categorical (nominal) variable with 5 levels. The continents included in this study include Africa (n = 52 countries), the Americas (n = 25 countries), Asia (n = 33 countries), Europe (n = 30 countries), and Oceania (n = 2 countries). Given that Oceania has a small cell size, results should be interpreted with caution.

• GDP per Capita: numeric (continuous) variable. GDP per capita refers to the monetary value of goods and services made within a country in a given year. GDP is thus as estimate of a country’s economy and growth rate. Mean GDP per Capital in 2007 was $11,680.07 (SD = 12,859.95).

• Life Expectancy: numeric (continuous) variable. Life expectancy refers to the average number of years a person is expected to live based on the current population’s mortality rates. Mean life expectancy in 2007 was 67.01 (SD = 12.07).

Descriptive statistics are presented in Table 1.

Table 1

Descriptive Statistics

Variable	Mean	SD	Min	Max	Sum	Percentage
continentAfrica					52	36.62%
continentAmericas					25	17.61%
continentAsia					33	23.24%
continentEurope					30	21.13%
continentOceania					2	1.41%
gdpPercap	11680.07	12859.94	277.55	49357.19
lifeExp	67.01	12.07	39.61	82.6

Bivariate Correlations

Bivariate correlations were computed to assess the association between life expectancy, GDP per capita, and each continent. GDP per capita was positively associated with life expectancy such that higher GDP per capita was associated with a longer life expectancy (r = 0.68). Europe was more strongly correlated with life expectancy (r = 0.46) and GDP per capita (r = 0.54) as compared with other continents. Bivariate correlations for all variables are presented in Table 2.

Note. To statistically examine the research questions, Europe was selected as a reference group to determine how life expectancy might vary across continent (i.e., Africa, Asia, Oceania, and the Americas) alongside GDP, given its significant positive correlation with life expectancy when compared to other continents.

Table 2

Correlation Matrix

III. Research Question 1:

Does life expectancy differ across world regions (i.e., at the continent level)?

To address this research question, we first displayed average life expectancy in 2007 across world regions using a map. Although the map presented in Figure 1 is descriptive in nature, it suggests that life expectancy varies as a function of world region.

Figure 1

Average Life Expectancy by Continent in 2007

Note. This map displays the average life expectancy for each continent in 2007. Visual interpretation of the plot indicates that regions such as Oceania and Europe have the highest reported average life expectancy, compared to Africa, which has the lowest average life expectancy. Finally, the Americas and Asia have moderate average life expectancy.

Next, a linear regression was performed to test the association between continent and life expectancy with Europe acting as the reference group. Africa, the Americas, and Asia were negatively associated with life expectancy when compared with Europe. Thus, those living in Africa, the Americas, or Asia have a significantly lower life expectancy as compared to those living in Europe. Results are presented in Table 3.

Table 3

Model 1: Main Effect of Continent Predicting Life Expectancy

variable	estimate	std.error	p.value
Intercept	77.65	1.35	< .001
Africa	-22.84	1.7	< .001
Americas	-4.04	2	0.046
Asia	-6.92	1.87	< .001
Oceania	3.07	5.4	0.571

IV. Research Question 2:

Is GDP per capita associated with life expectancy?

To address this research question, we added GDP per capita as a predictor to the previous linear regression model. When controlling for the effect of continent on life expectancy, GDP per capita was positively associated with life expectancy such that higher GDP per capita was associated with longer life expectancy. Results are presented in Table 4.

Table 4

Model 2: Main Effect of Continent and GDP per Capita Predicting Life Expectancy

variable	estimate	std.error	p.value
Intercept	68.96	1.86	< .001
Africa	-15.23	1.96	< .001
Americas	0.83	1.95	0.671
Asia	-2.56	1.81	0.159
Oceania	1.42	4.81	0.768
GDP per Capita	0	0	< .001

To visualize the main effect of GDP per capita on life expectancy, a scatterplot was produced (see Figure 2, top). In addition, scatterplots were produced to assess the effect of GDP per capita of life expectancy divided by continent (see Figure 2, bottom).

Figure 2

The Association between GDP per Capita and Life Expectancy in 2007

Note. Scatterplots are presented for the association between GDP per capita (log10) and life expectancy. The top scatterplot displays the association between GDP per capita (log10) and life expectancy overall, whereas the scatterplot matrix on the bottom display the same association divided by continent.

V. Research Question 3:

Do world regions (i.e., at the continent level) and GDP per capita interact to predict life expectancy?

To address this research question, we added interaction terms for GDP per capita and each continent as a predictor to the previous linear regression model. The interaction for Africa and GDP per capita was statistically significant; however, all other interactions were not statistically significant. Results are presented in Table 5.

Table 5

Model 3: Interaction Between Continent and GDP per Capita Predicting Life Expectancy

variable	estimate	std.error	p.value
Intercept	72.27	2.8	< .001
Africa	-20.63	3.04	< .001
Americas	-1.64	3.42	0.633
Asia	-6.38	3.18	0.047
Oceania	5.13	29.83	0.864
GDP per Capita	0	0	0.036
Africa:GDP per Capita	0	0	0.003
Americas:GDP per Capita	0	0	0.743
Asia:GDP per Capita	0	0	0.183
Oceania:GDP per Capita	0	0	0.917

To better understand the interaction between continent and GDP per capita, simple slopes analysis was conducted and is presented in Figure 3. Visual inspection of the plot confirms that there are no significant differences between Europe, Asia, Oceania, and the Americas. However, there is a significant difference for those living in Africa, which has an overall lower GDP per capita and lower life expectancy as compared to Europe.

Figure 3

Simple Slopes Analysis: Interaction Between Continent and GDP per Capita Predicting Life Expectancy in 2007

Note. The scatterplot displays the interaction between continent and GDP per capita (log10) predicting life expectancy in 2007.