Investigating the Impact of Educational Attainment on Life Expectancy in Canada: Empirical Report
In this study, educational attainment is measured through access to education and the number of graduates. To measure these two elements, the study relied on the enrolment totals for youth programs as measures and the graduate levels of students in Canada. These two variables informed the choice of the analytical model to be used. The right choice is the OLS regression model. This model is best suited for data that is distributed in a linear manner. It is an appropriate style, especially in situations where the assumptions hold. This model is more concerned with the squares of the errors finding the line that goes through the sample data. It is appropriate for studies that have more than one independent variable. In this study, the independent variables are two. Therefore, the OLS regression model is the right choice. It will allow the comparison between the two independent variables and the dependent variables. The coefficient of determinants will be revealed for each variable so that the net effect can be identified. The net results will be the overall significance of the model and the answer to the research problem. The model will rely on the data to identify the impact that educational attainment has on life expectancy in Canada.
The OLS regression model, in this case, worked with two predictor variables and one criterion variable. The relationship between the variables was measured using the following equation.
∑ = β + β1M + β2C + u
Where ∑ represented Life Expectancy
β represented the intercept or constant in the equation
β1,2 represented the coefficients of determinant
M represented the accessibility to education
C represented the number of graduates
u represented the error term in the equation
The central premise in using this equation is that the predictor variables all have statistical significance enough to allow the study to analyze the overall relationship. The main assumptions in this paper are that the moderating and mediating factors do not have an overbearing impact on the research. This would lead to inaccurate findings or skewness in the data. Thus, so long as these assumptions hold true, the OLS regression equation should be able to provide the net relationship effect between the predictor variables and the criterion variable.
In this investigative report, the focus was on the impact of educational attainment on life expectancy in Canada. To arrive at the inferences, data secondary data was collected. The study relied on time-series data to answer the research questions. The main data sources used were government databases for secondary data. The benefits of using secondary data include the time and cost efficiency inherent to the process. The researcher does not have to travel to acquire the needed data nor engage in the tenuous activities of cleaning the data. All those important steps in data collection are already taken care of by experts. The data was ready for analysis.
The study hypothesizes that the two predictor variables have adequate statistical significance to be included in this study. Further, it is hypothesized that the variables are strong enough to stand on their own without interfering with each other. The two main hypotheses in this study are two. The null hypothesis holds that educational attainment has no impact on life expectancy in Canada. The second hypothesis, alternative hypothesis, holds that educational attainment has an impact on life expectancy. It is also expected that access to education leads to high life expectancy. Again, the number of graduates is expected to have a positive and significant relationship with life expectancy in Canada. These hypothesized results were put to the test, and the data was analyzed. The following section gives the actual results following the analysis sing the OLS regression model.
In this section, the actual results are presented. The data used was largely organized into blocks and categories that allowed for direct analysis. The data cleaning part of the analysis part was skipped as the data was already prepared for analysis. The regression model gave out three main tables. The first table detailed the regression model summary, the second had the ANOVA test results, and the last one showed the coefficient of determinants for each variable.
Table 1: Regression Model Summary
|Model||R||R Squared||Adjusted R Squared||Std. Error of the Estimate|
From the above table, the R, R Squared, Adjusted R Squared, and the Standard Error of the Estimate are presented. In this particular analysis, the R-value was positive. This indicates that the overall effect of the variables used was positive. Also, it means that in the period that the data collected covered, the life expectancy increased by 0.951 units.
The R squared value was also positive. It gives the proportion of the variance that the independent variables can explain within the dependent variable (Schroeder et al. 2016). In this case, that proportion is 0.904 or 90.4%. This means that the mediating and moderating factors in this study’s central relationship under investigation were responsible for the remaining proportion. The adjusted R squared shows the impact that the added variable could have on the current results (Brook and Arnold 2018). In this case, the adjusted R squared value was 0.902, which means that adding other variables such as the mediating or moderating variables could have had a 90.2% impact on the final results. Further, the standard error of the estimate was found to be low at 2.672, which means that the accuracy of the predictions is high.
Table 2: ANOVA Results
|Model||Sum of Squares||df||Mean Square||F||Sig.|
|a. Dependent Variable: Life Expectancy|
From Table 2, the ANOVA results show the values of the F value and the overall significance of the model in this case. The F value was 489.640, which is much higher than the p-value at 0.05. This indicates that the variation in the sample data is within the acceptable range to allow the author to pull out accurate answers to the research problem (Daoud 2017). Further, the overall significance of this data was found to be 0.000, which means that the model’s level of significance allowed the study to investigated the coefficient of determinants.
Table 3: Coefficients of Determination
|Model||Unstandardized Coefficients||Standardized Coefficients||t||Sig.|
|Accessibility to education||0.696||0.261||0.577||2.667||0.010|
|Number of graduates||0.522||0.203||0.508||2.571||0.013|
- Dependent Variable: Life Expectancy in Canada
In table 3, the coefficient of determinants is revealed for the two predictor variables. First off, the constant’s coefficient stood at 1.253 and a significance of 0.001. This means that the two variables were statistically significant and positive. Based on the values for the coefficient, it is clear that a unit increase in accessibility to education led to an increase of 0.696 units in life expectancy at a significance of 0.010. Also, it was revealed that a unit increase in the number of graduates led to an increase of 0.522 units in life expectancy at a significance of 0.013. Accessibility to education had the least significant of the two variables, which means it had the greatest influence on life expectancy.
Interpretation of the Results
Based on the results outlined above, the study found that the model was highly significant to this study. This validates the selection of the OLS regression model to analyze the data collected. Moreover, the study found that the accuracy of predictions was high, which means that the inferences made were highly correct. Further, it was found that the second hypothesis, alternative hypothesis, holds true that educational attainment has an impact on life expectancy. As was expected, accessibility to education leads to high life expectancy. Again, the number of graduates was found to be as was expected and had a positive and significant relationship with life expectancy in Canada.
These results imply that Canada’s life expectancy rise in the period analyzed in this study increased due to these two predictor variables. Further, it means that in that period, the number of graduates increased across Canada while the accessibility to education increased, allowing more people to graduate. Thus, with more people having high education attainment levels, their access to high income-earning job opportunities also increased. This implies that more people in the Canadian economy were earning high incomes allowing them to access medical and basic amenities to increase their life expectancy.
In a nutshell, the study established that the second hypothesis, alternative hypothesis, holds true that educational attainment has an impact on life expectancy. As was expected, accessibility to education leads to high life expectancy. Besides, the number of graduates was found to be as was expected and had a positive and significant relationship with life expectancy in Canada. These results imply that Canada’s life expectancy rise in the period analyzed in this study increased due to these two predictor variables. Further, it means that in that period, the number of graduates increased across Canada while the accessibility to education increased, allowing more people to graduate.
The study was based on secondary data collected over a period in the past. To improve the study’s relevance, raw data should have been used, focusing on the last two years. In addition, the study could have been improved by adding other variables to explain the variances that were not accounted for, as revealed by the ANOVA test. This includes mediating and moderating variables that would have helped bolster the understanding of the research problem focused on in this paper. In this way, it would have been possible to understand why life expectancy in Canada has increased. Using other factors that are not education-related would have also diversified the study’s analysis making it possible to cover a large scope on the factors that determine life expectancy in Canada.