Data Analysis Tools: WEEK 4

Exploring Statistical Interactions

Assignment: Testing a Potential Moderator

Source: New data from OECD (“The Organisation for Economic Co-operation and Development”)

Variables Used: 

*Employment Rate -  It is the number of employed persons aged 15 to 64 over the population of the same age. (source: OECD)

*Life Satisfaction - This indicator considers people's evaluation of their life as a whole (source: OECD)

*Education - percent of people with at least High School Education. (source: OECD)

Introduction:

In the first week of Data Analysis Tools, I examined the correlation between Employment rate (explanatory variable) and Life Satisfaction (response variable). The analysis proved a strong positive relationship between the two variables. The higher Employment Rate, the higher Life Satisfaction.

This week I decided to test if the Education variable (percent of people with at least high school education) is a moderator of the said relationship. I categorized the Education into two levels: countries with low percent (1) and high percent (2) of educated people.
Similarly, I categorized Employment rate (explanatory variable) into two levels: low employment rate (1) and high employment rate (2). After categorization, I ran a new Anova procedure for the Employment Rate and Life Satisfaction relationship in two  "Education" moderator sub-groups:

CODE:

Output:

[Table 1]

[Table 2] 

Interpretation:

In this analysis, I was interested to see if the moderator “Education” affects the relationship between Employment Rate (explanatory variable) and Life Satisfaction (response variable).

It appeared that the moderator does not change the relationship between the two variables. The level of Life Satisfaction was still higher with the increase of Employment Rate in both moderator sub-groups.

Moreover, in each sub-group the relationship between Employment Rate and Life Satisfaction remained statistically significant.

Below, there are more details concerning the results of Anova procedure:

[Table 1]

ANOVA on Employment Rate compared with Life Satisfaction– in subgroup moderator “Countries with LOW percent of people with at least high school education”

F-statistic: 11.65
Prob (F-statistic): 0.0039

Since p is less than 0.05, we can reject the null hypotheses and say that there is a significant relationship between Employment Rate and Life Satisfaction in countries with LOW percent of people with at least high school education.

[Table 2]

ANOVA on Employment Rate compared with Life Satisfaction– in subgroup moderator “Countries with HIGH percent of people with at least high school education”

F-statistic: 7.11
Prob (F-statistic): 0.0176

Again, as p is less than 0.05, there is a significant relationship between Employment Rate and Life Satisfaction in countries with HIGH percent of people with at least high school education.

***

Taking everything into consideration, we can assume that the Education variable (percent of people with at least High School education) does not moderate the relationship between Employment Rate and Life Satisfaction. The said relationship remains positive and statistically significant in both sub-groups of the moderator.