Impact Of Education On Income: Analysis And Results

Purpose of the study

Education is the right of everyone. It is now added in the basic needs of human. Most of the things like standard of living, habits, social life, etc are affected by education. But the most affected thing is income. Although there are many instances in the world where this statement is not valid. Most of the richest are not so educated, some of them are school drop outs. But part from these most part of societies successes affected by education.

Prestige is one of thing which is highly affected by education. Marsh (1971) gives the explanation of occupational prestige hierarchies. Barringer, Takeuchi & Xenos (1990) studied tht education, occupational prestige, and income from Asian Americans. Jamison, Jamison & Hanushek (2007) studied how the income growth and mortality decline affected by education quality. Income inequality among the women are studied by many athors. (Kahn et al. (2000), Lynch et al. (2000), Kawachi et al. (1997) Dollar & Gatti  (1999), Cancian, Danziger & Gottschalk (1992).)

The main aim of this study is to determine the relation between level of education and income. This study is useful for social workers, educational institutes for their action for future generation.

The purpose of the study is two folded. The main purpose of the study is to examine the relationship between the level of education and income and other variables. The second aim is to study the relationship between income and percentage of women and prestige in occupation.

Hypothesis to be tested:

Following are the hypothesis tested in this study.

H₀₁: There is no significant relation between income level and education level.

H₁₁: There is significant relation between income level and education level.

H₀₂: There is no significant relation between income and independent variable prestige and women.

H₁₂: There is significant relation between income and independent variable prestige and women.

The secondary data is extracted from the prestige dataset from R console having 102 rows and 7 columns where each row is an observation that relates to an occupation. The data following variables:

Education: The average number of years of education for occupational incumbents.

Income: The average income of occupational incumbents, in dollars.

Women: The percentage of women in the occupation.

Prestige: The average prestige rating for the occupation.

Census: The code of the occupation used in the survey.

Type: Professional and managerial(prof), white collar(wc), blue collar(bc), or missing(NA)

Data analysis is incomplete without statistical analysis. In literature, there is huge number of statistical tools and techniques available but selection of proper tools and technique is the main part of analysis. For the purpose of study following methods are carried:

Descriptive summary: Descriptive summary of the variables under study is obtained.

Scatter Plot: Figure speaks 100 times than numbers. Scatter plot are plotted for studying the relation between variables.

Correlation Analysis: Correlation analysis is carried out to study the correlation between variables.

Regression Analysis: Simple and Multiple regression is used for testing the hypothesis given above.

Result and discussion are done in this section. R code for this section is given in Appendix. Data is read using read.csv( ) in R.  Dalgaard (2008), Crawley (2012), Verzani (2014), Field, Miles, Field (2012) and Hothorn & Everitt (2009) are referred for R coding.

Research Methodology

Bickel & Doksum (2015), Pillers Dobler (2002), Moyé, Chan & Kapadia (2017) and Baayen (2008) is used for this section.  Table 1 shows the mean, standard deviation, minimum and maximum for the education level, income level, prestige and percentage of women in occupation.

Table 1: Summary statistics for the education level, income level, prestige and percentage of women in occupation.

	Education Level	Income Level	Women Percentage	Prestige
Mean	10.74	6797.90	28.98	46.83
SD	2.73	4245.92	31.72	17.20
Minimum	6.38	611.00	0	14.80
Maximum	15.97	25879.00	97.51	87.20

The average education level is 10.74 years with standard deviation 2.73.

• The average income level is $6797.90 with standard deviation 4245.92.
• On an average 28.98% women in the each occupation.
• Mean prestige rating is 46.83 with max rating 87.2. 

• There is increasing trend in income as education level increases.
• Prestige has linearly related with income and education.
• Women percentage is not related with other variables.

Table 2 shows the correlation matrix for education level, income level, prestige and percentage of women in occupation. From Table 2, it is observed that

• Income is positively related with education and prestige whereas it is negatively correlated with women percentage.
• Prestige is highly positively correlated with education and income.

Table 2: Correlation matrix for education level, income level, prestige and percentage of women in occupation.

	Education	Income	Women	Prestige
Education	1.000	0.578	0.062	0.850
Income	0.578	1.000	-0.441	0.715
Women	0.062	-0.441	1.000	-0.118
Prestige	0.850	0.715	-0.118	1.000

Pedhazur & Kerlinger  (1973), Glantz,  Slinker  & Neilands (2016),  Kleinbaum et al. (2013) and Seber & Lee  (2012) are referred for this section.

Simple Linear Regression:

Here income level is regressed on education level to test the hypothesis that whether there is any significant relation between income level and education level. Following null and alternative hypothesis are tested:

Null hypothesis: There is no any significant relation between income level and education level.

Alternative hypothesis: There is significant relation between income level and education level.

Simple linear regression is carried out in R (code and output given in appendix). p value = 2.079e-10 < 0.0001 is observed suggest that there is significant relation between income level and education level. R2 is 0.3386 suggest that only 33.86% variation in income level is explained by education level. There may be other variables which affect income as years of experience, skills, etc.

By observing the p value of significance of coefficient, both the coefficient (intercept and coefficient of education level) is found to be significant at 5%.

The fitted regression line is

Income = -2853.6 + 898.8 × Education

There is significant positive relation between income and education. Each unit incline in education results in increment of $894.8in income. If education level is 0 then there is negative income.

Here income is regressed on prestige and women variables to test the hypothesis that whether there is any significant relation between income and independent variables prestige and women. Following null and alternative hypothesis are tested:

Null hypothesis: There is no any significant relation between income and independent variables prestige and women.

Alternative hypothesis: There is significant relation between income and independent variables prestige and women.

Multiple linear regression is carried out in R (code and output given in appendix). p-value: < 2.2e-16 is observed suggest that there is significant relation between income and prestige and women variable. R² is 0.64 suggest that 64% variation in income is explained by prestige and women.

By observing the p value of significance of coefficient, both the coefficient (prestige and women) are found to be significant at 5% whereas intercept is found to be non-significant at 5%.

The fitted regression line is

Income = 431.574 + 165.876 × Prestige – 48.385 × Women

There is significant positive relation between income and prestige whereas significant negative correlation between income and women. Each unit incline in prestige results in increment of $165.876 in income when the percentage of women is constant. Each unit incline in women percentage in occupation results in decrement of $48.385 in income when the prestige is constant.

Result and Discussion

Conclusion 

From the descriptive statistics of the variables under study found that the average education level is 10.74 years with standard deviation 2.73 whereas the average income level is $6797.90 with standard deviation 4245.92. On an average 28.98% women in the each occupation are reported. Average prestige rating is 46.83 with standard deviation 17.20 and max rating 87.2.

From the scatter plot and correlation analysis, it is observed that there is increasing trend in income as education level increases and prestige has linearly related with income and education.  Income is positively related with education and prestige whereas it is negatively correlated with women percentage. Prestige is highly positively correlated with education and income.

From the simple linear regression where income is regressed on education level suggests that there is significant relation between income and education. Each unit incline in education results in increment of $894.8in income.

From the multiple linear regression it is observed that there is significant relation between income and prestige and women variable. R² is 0.64 suggest that 64% variation in income is explained by prestige and women. Both the coefficient (prestige and women) are found to be significant at 5% whereas intercept is found to be non-significant at 5%. There is significant positive relation between income and prestige whereas significant negative correlation between income and women. Each unit incline in prestige results in increment of $165.876 in income when the percentage of women is constant. Each unit incline in women percentage in occupation results in decrement of $48.385 in income when the prestige is constant.

References:

Baayen, R. H. (2008). Analyzing linguistic data: A practical introduction to statistics using R. Cambridge University Press.

Barringer, H. R., Takeuchi, D. T., & Xenos, P. (1990). Education, occupational prestige, and income of Asian Americans. Sociology of Education, 27-43.

Bickel, P. J., & Doksum, K. A. (2015). Mathematical statistics: basic ideas and selected topics, volume I (Vol. 117). CRC Press.

Cancian, M., Danziger, S., & Gottschalk, P. (1992). Working wives and family income inequality among married couples. [Unpublished] 1992. Presented at the Annual Meeting of the Population Association of America Denver Colorado April 30-May 2 1992..

Crawley, M. J. (2012). The R book. John Wiley & Sons.

Dalgaard, P. (2008). Introductory statistics with R. Springer Science & Business Media.

Dollar, D., & Gatti, R. (1999). Gender inequality, income, and growth: are good times good for women? (Vol. 1). Washington, DC: Development Research Group, The World Bank.

Field, A., Miles, J., & Field, Z. (2012). Discovering statistics using R. Sage publications.

Glantz, S. A., Slinker, B. K., & Neilands, T. B. (2016). Primer of applied regression & analysis of variance. McGraw-Hill Medical Publishing Division.

Hothorn, T., & Everitt, B. S. (2009). A handbook of statistical analyses using R. Chapman and Hall/CRC.

Jamison, E. A., Jamison, D. T., & Hanushek, E. A. (2007). The effects of education quality on income growth and mortality decline. Economics of Education Review, 26(6), 771-788.

Kahn, R. S., Wise, P. H., Kennedy, B. P., & Kawachi, I. (2000). State income inequality, household income, and maternal mental and physical health: cross sectional national survey. BMJ: British Medical Journal, 321(7272), 1311.

Kawachi, I., Kennedy, B. P., Lochner, K., & Prothrow-Stith, D. (1997). Social capital, income inequality, and mortality. American journal of public health, 87(9), 1491-1498.

Kleinbaum, D. G., Kupper, L. L., Nizam, A., & Rosenberg, E. S. (2013). Applied regression analysis and other multivariable methods. Cengage Learning.

Lynch, J. W., Smith, G. D., Kaplan, G. A., & House, J. S. (2000). Income inequality and mortality: importance to health of individual income, psychosocial environment, or material conditions. BMJ: British Medical Journal, 320(7243), 1200.

Marsh, R. M. (1971). The explanation of occupational prestige hierarchies. Social forces, 50(2), 214-222.

Moyé, L. A., Chan, W., & Kapadia, A. S. (2017). Mathematical statistics with applications. CRC Press.

Pedhazur, E. J., & Kerlinger, F. N. (1973). Multiple regression in behavioral research. New York: Holt, Rinehart and Winston.

Pillers Dobler, C. (2002). Mathematical statistics: Basic ideas and selected topics.

Seber, G. A., & Lee, A. J. (2012). Linear regression analysis (Vol. 329). John Wiley & Sons.

Verzani, J. (2014). Using R for introductory statistics. CRC Press.