Methods Analysis and Techniques Coursework - Essay Example

METHODS ANALYSIS AND TECHNIQUES WORK – DATA ANALYSIS Motivation and background of the study In recent times, the relationship that exists between health status and total income of a family has been called into question. Previous studies have suggested and revealed a negative relationship between low income and life expectancy and appositive relationship between low family income and mortality across most developed countries. Other revised studies have indicated a positive relationship between high income and high health standards within the family settings. According to The National Poverty Center at the University report, there were 14.1 million children that lived in poverty in the USA by the year 2008. 19% of these children were below 18 years old. The official level of poverty changes and in relation to household size; however, in 2008 the definition of poor family changed if a single parent with two children had an income below $17,346. The Connecticut Commission on children indicates that most children living in poverty are more vulnerable to diseases than children from more a well off homes. According to this commission, most babies that are born to low-income parents experience lower birth weights hence are more vulnerable to infant mortality. Child Trends organization reports that children from poor families experience mental health such as impulsiveness, disobedience and difficulties in association with other people. Their self-esteem is low and this may affect them throughout their lives. It is clear that income is a determinant on the health conditions of a person. Therefore this study wants to determine the effect of total family income on the health condition of an individual. The research question guiding this study is: does total income a family positively impact the health conditions of an individual. Objective of the study The objective of the study is to determine the effects of total family income in the health condition/status of an individual. Importance of the study This study will be appropriate at making conclusions about the relationship that exist between income and health conditions. This study will be important to several stakeholders including the ministries of planning and health to make appropriate policies that will improve health conditions of the citizens. Statistical assumptions Researchers, who study the effect of income on health conditions, generally seek to generalize their results from the available data to some larger context by generalizing a sample to a population. Assumptions are important aspect of empirical studies. This study, just like any other study, applies some statistical assumptions in order to achieve at the desired results. The statistical assumptions are: The mean difference is zero The data is normally distributed The variance of the two variables are equal (homoscedasticity) Methodologies The independent variable of the study is income while the dependent variable is health. The study will mainly duel on correlation and regression for data analysis. Correlation coefficient is important in showing whether and how strongly income and health conditions are related. Because the study is linear in nature, Pearson product-moment correlation coefficient will be used to measure the direction and strength of the linear relationship between income and health conditions. The value of Pearson’s correlation coefficient is influenced by the distribution of the independent (income) variable in the sample. Regression analysis defines the relationship between income and health status. The following regression model will be used during the study: Y=Bo + BX + e (where; Y= heath conditions, X=income, Bo=constant, e=error term). T-test will be used to determine the significance of the regression model before using it to predict the value of health status given values for income. Apart from; t-tests, regression and correlation coefficient, tables and graphs will also be used to determine the effect of income on health conditions of an individual. Hypothesis The hypothesis of the study is derived from the research question above. Null hypothesis Ha: total income of a family positively impacts the health conditions of a person. Alternative hypothesis Ho: total income of a family does not positively impacts the health conditions of a person. RESULTS The mean for total family income is 26588 while the mean of health condition is 20073.06. The standard deviations for total family income and health condition are 15666.067 and 82067 respectively. Correlation results The correlation matrix is shown below RESULTS Correlation 15-24 years   Health status income Health status 1 income 1 1 Regression Statistics Multiple R 1 R Square 1 Adjusted R Square 1 Standard Error 835 Observations 25 ANOVA   df SS MS F Significance F Regression 1 3,814,326,656 3,814,326,656 5,467 0 Residual 23 16,047,650 697,724 Total 24 3,830,374,307         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 147,047 681 216 0 145,638 148,456 145,638 148,456 X Variable 1 2 0 74 0 2 2 2 2 figure 1 Correlation 35-44   health status income health status 1 income 1 1 SUMMARY OUTPUT Regression Statistics Multiple R 1 R Square 1 Adjusted R Square 1 Standard Error 4,501 Observations 38 ANOVA   df SS MS F Significance F Regression 1 1,467,599,201 1,467,599,201 72 0 Residual 36 729,265,340 20,257,371 Total 37 2,196,864,541         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 19,522 1,888 10 0 15,694 23,350 15,694 23,350 X Variable 1 1 0 9 0 0 1 0 1 Correlation 45-54   health status income health status 1 income 1 1 SUMMARY OUTPUT Regression Statistics Multiple R 1 R Square 1 Adjusted R Square 1 Standard Error 795 Observations 25 ANOVA   df SS MS F Significance F Regression 1 1,020,487,342 1,020,487,342 1,616 0 Residual 23 14,525,164 631,529 Total 24 1,035,012,506         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 5,194 728 7 0 3,688 6,700 3,688 6,700 X Variable 1 1 0 40 0 1 1 1 1 figure 2 Correlation 54-64   health status income health status 1 income 1 1 Multiple R 1 R Square 1 Adjusted R Square 1 Standard Error 2,405 Observations 38 ANOVA   df SS MS F Significance F Regression 1 770,057,270 770,057,270 133 0 Residual 36 208,284,294 5,785,675 Total 37 978,341,564         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 13,236 922 14 0 11,367 15,106 11,367 15,106 X Variable 1 0 0 12 0 0 0 0 0 figure 3 Correlation over 75 years   health status income health status 1 income 1 1 Regression SUMMARY OUTPUT Regression Statistics Multiple R 1 R Square 1 Adjusted R Square 1 Standard Error 460 Observations 25 ANOVA   df SS MS F Significance F Regression 1 108,012,224 108,012,224 511 0 Residual 23 4,860,980 211,347 Total 24 112,873,204         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 5,936 389 15 0 5,132 6,740 5,132 6,740 X Variable 1 0 0 23 0 0 1 0 1 figure 4 The Pearson correlation coefficient is -0.191. This means that there is a weak negative relationship between total family income and health condition. The negative sign shows that if one variable increase, the other variable decreases. For example, increases in total family income will results to a decline in health conditions. Generally, total family income and health conditions exhibit negative correlation. The Sig. (2-Tailed) value from correlation output is 0.00. This value is less than 0.05. Hence there is a statistically significant correlation between total family income and health conditions. Regression The regression results exhibit similar results to those of the correlation results. Checking the correlation results from every groups-using the model above-then we find the following regression line: 15-24 years: Y= 147,047 + 2X 35-44 years: Y=19,522 + X 45-54 years: Y=5,194 +X 54-64 years: Y=13,236 + 0X 75+ years Y= 5,936 + 0X The independent factor’s coefficient (health status) drops from 2-0 as the ages increases. This shows that income positively influence the health status of an individual. For example between the years 15-24, 35-44 and 45-54. However, this trend is opposite for the person’s between 54 and over 75 years. The general regression equation when all these ages are incorporated together is: Y= 8,356 + 1X +e This means that holding other factors constant, an increase in income by one unit will increase heath status by one. 8,356 is the constant of the regression line which shows that holding other factors constant an individual’s health status should be 8,356 units despite the age. The coefficient for income is statistically significant because the p-value is 0.00 which is less than 0.05. The intercept is significantly different from 0 at the 0.05 alpha levels. Hypothesis testing Ha: total income of a family positively impacts the health conditions of a person. Ho: total income of a family does not positively impacts the health conditions of a person. The one way Anova gives a p-value of 0.00. The study uses 95% confidence interval that means the significance level is 0.05. Therefore, the P-value (0.00) < significance level (0.05). Therefore we reject the null hypothesis and accept the alternative hypothesis. We conclude therefore that income does not positively impact the health of a person. Conclusion The objective of the study was to determine the impact of total family income in health conditions. The regression analysis shows that income does not positively impact health conditions of an individual. The hypothesis testing also indicates similar results while Pearson correlation coefficient also a negative correlation between income and health conditions. Therefore, from the study it is necessary to conclude that income does not positively impact health conditions. This is in line with other studies but at the same time contradicts other earlier studies. But the conclusion of this study is solely based on the results attained from analysis of the available data. References Lewis-Beck, Michael S. 1997. Data analysis: an introduction. Thousand Oaks [u.a.]: Sage. McKinney, Wes. 2012. Python for data analysis. Sebastopol, CA: OReilly Media. http://proquest.safaribooksonline.com/?fpi=9781449323592. Read More