Introduction
The happiness index is a classic national parameter of well-being that summarizes information about the quality of life, civil rights, and economic opportunities of citizens within a country. In short, citizens of regions with a higher happiness index are themselves happier, and their quality of life is higher. For the present assignment, a data set was selected that assesses the consolidated happiness index in different countries according to several criteria over the past fifteen years. Thus, the chronological data allow us to evaluate not only the countries among themselves according to this criterion but also to provide the dynamics of the change in the happiness index within the country (WHR, 2021). Nine different types of variables are represented in the array, a description of which is indicated by Helliwell et al. (2021). Each variable assesses an individual measure of perceived and objective quality of life within a country, and their consolidation yields a single happiness index.
Data
In this paper, the key research question was to determine the relationship between two variables from this data set. Specifically, the first variable is the Life Ladder, which is the perceived standard of living as assessed by respondents. To obtain it, respondents in different countries were asked how they rated their quality of life on a ten-point scale, where 10 was the maximum level. The second variable chosen was “Freedom to make life choices,” which measures the extent to which respondents feel that their civic self-determination is essential and valuable. This criterion indicates the ability to make critical decisions in their lives independently without interference. This variable ranges from 0 to 1, where 1 is the strongest self-determination. Thus, the working hypothesis is that there is a strong positive correlation between the two variables, which means that an increase in one variable is associated with an increase in the other.
Since the current data set included historical information, it was decided to pre-process it. To do this, all repeated entries for each country were excluded so that only one row remained for each country, with the most recent year of measurement chosen. The total number of rows (countries) in the data set was 163. In addition, additional variables that were not measured in this paper were removed. After these operations, it was found that for several countries, no data were available for recent years, so older data were used. Thus, the final data set was a 4×164 table and contained three types of records for each country, namely Year of Measurement, Life Ladder, and Freedom.
Methods
Statistical analysis was performed in this paper using the built-in functions of MS Excel. Elements of descriptive statistics included determining measures of central tendency and characteristics of the scatter of the data. Regression analysis was also performed automatically and included not only the determination of linear regression equation coefficients between the variables but also the Pearson correlation coefficient.
Analysis
Measures of central tendency were calculated for the distribution of the two variables. The Life Ladder was found to have a mean of 5.494 (SD = 0.086), and for Freedom, 0.787 (SD = 0.010). In addition, kurtosis values were also determined to be: -0.293 for Life Ladder and 1.753 for Freedom, respectively. The skewness values were -0.210 for Life Ladder and -1.210 for Freedom. In terms of data scatter, the IQR measurement showed that the data for Life Ladder had an IQR of 1.485 and Freedom 0.169. An interesting finding concerned the year of the last measurement – for developing countries, it was found that the last measurement was taken the earliest, 2006-2014, whereas, in developed countries, studies were taken almost every year. The results of the analysis of the descriptive statistics are shown in Table 1 below.
Table 1: Results of Descriptive Statistics for Two Variables
The working hypothesis, recall, was the possibility of the existence of a strong positive correlation between the two selected variables. In addition, it was interesting to perform a regression analysis to try to assess the relationship between them. Using both types of tests at once was expected to not only determine the strength of the association between them but also to reveal the direction of this potential relationship. The results of the consolidated analysis are demonstrated below in Table 2.
Table 2: Results of Data Regression Analysis
Results
Turning to the results shows that the Pearson correlation coefficient for the two variables is 0.628. This value indicates the existence of a high positive correlation between Life Ladder and Freedom. Consequently, it is proper to emphasize that an increase in Life Ladder value led to a natural increase in freedom to make life choices by citizens. In addition, the results of regression analysis show, first, that the model constructed is statistically significant (p<.001). However, the proposed linear regression equation had an R2 value of 0.395 – meaning that only about 40 percent of the variance of all data in the distributions was covered by the proposed equation. According to the value in Table 2, the final regression equation can be formulated as follows:
The equation clearly shows that an increase in Life Ladder values for every one unit led to a 0.075 increase in Freedom values. Consequently, there was a positive slope between the variables, confirming their consolidated growth. Referring back to Figure 1 perfectly demonstrates the relationship described: the variables grow together, with the linear model not fully capturing the variation in the data.
Conclusion
In emphasizing the conclusion, it should first be noted that a high positive correlation was found between the two variables, which supports the stated working hypothesis. An increase in the ability to make critical decisions in life is strongly related to an increase in the perception of the quality of one’s own life, according to the analysis. Regression analysis showed parameters for this growth, but it is fair to note that the linear regression only covered about forty percent of all variances. The findings mean that if national leaders want to increase the happiness of their citizens, they should grant them more political freedoms. For future research, it would be helpful to expand the input metrics and conduct multiple regression analyses.
References
Helliwell, J. F., Huang, H., Wang, S., and Norton, M. (2021) Statistical appendix 1 for chapter 2 of world happiness report 2021 [PDF document]. Web.
WHR. (2021). Data Panel [PDF document]. Web.