Objectives
The study general objective was:
The specific objectives were:
Hypothesis
The study hypothesis was: Demographic and socio-economic status of an individual affect their standard of happiness in a relationship. This resulted in the following null and alternative hypothesis to be tested using the multinomial logistic regression model.
H0: Demographic and socio-economic status of someone have no influence on their level of happiness in rapport.
H1: Demographic and socio-economic status of someone have an impact on their level of happiness in rapport.
Descriptive Statistics
Descriptive statistics are inclusive of the measures of central tendency, frequencies, percentage, measures of dispersion, skewness, and kurtosis. The statistics are used to describe the sampled population characteristics from which a generalization is made to the entire population. The above-stated variables were collected from a sample of 400 respondents to be used in the analysis for this study. The site for data collection was available in different regions in the Unites States. The study targeted individuals within 21 to 63 age bracket. The average age of the respondents was 33.12 years while that of their parents was 34.30 years, their parents age ranged from 18 – 79 years, having a wider range of 61 years compared to that of the sampled population, which had a range of 42 years. There was an equal representation, that is, fifty-percent representation of each of the genders for both the study respondents and their parents. The dates, on which the test was completed ranged between 01/02/2004 to 21 /01 /2007, the completion date running for the duration of four calendar years. 1/02/2006 recorded the highest number or respondents completing their test. The sampled population had varied and unique occupations both for the study units and their parents ranging from professional to casual occupation while most dominating occupations being students from the sampled population.
The average income for the respondents was $ 47112. 92 with a minimum of $ 0 and a maximum of $ 285000. The majority of the population was earning an income of $ 30,000; this corresponds to 7.5 % of the income. The respondent’s parents had an average of $ 38602. 10 with a minimum of $ 0 and a maximum of $ 200,000. There were only two categories of the study unit financial comfort, 54 % of the sampled population were comfortable financially while 46% of them were struggling. Parental financial stability had a direct impact on their children’s well-being –socially, economically and this ultimately influenced the type of relationship they were in and the level of happiness derived from the particular relationships.
Eight regions in the United States were used as the study sites from which the 400 sampled respondents were located. Table 1: shows that 18.5% of the sampled respondents were from Pacific Coast with the highest regional percentage of sampled respondents, followed by the South East at 15.8 %, then the Mid-Central region, giving 10.5% of the respondents, while the other remaining five regions had a percentage that was less than 10 in representation, with the least sampled population being from the North Central region of the United States at 6.8%.
The sampled respondents’ level of education ranged from having no high school education to having a doctorate. The largest representation regarding educational attainment was from the study units who had some college education at 32.3%, followed closely by those who were college graduates at 29.0% and 15.3% representation of those who had a post-graduate degree in Art or Science. The other education levels given have a level than 10% representation of the sampled population as illustrated in Table 2.
Twenty-nine percent of the respondent’s parents had some college education, 31.8 % of them were college graduates, and 16.3% were high school graduates. Those who had some graduate school education were 4.5 % of the sampled population, with the same percentage representing those with no high school education, 9.8% of them had a master’s degree and 4% had a doctorate.
Fifty-nine percent of the sampled population or the study was married giving a real scenario of the type of relationship on an individual’s level of happiness. From the sampled population of 400 respondents, 40.7% were in a single type of relationship: 25.5% were in a long-term commitment, 13.0% were faithful to one partner while the remaining 2.3 % were dating several people.
Inferential Statistics
Both the independent and the dependent variable have different levels or categories. For instance, for the marital status, there are two levels being single or married, while education has five levels: no education, have some high school education, have a college education, have some graduate school education, have a post-graduate education as either masters or a doctorate. The dependent variable has five levels: being wretched, unhappy, mixed, happy and jubilant in a relationship. The variable categorical nature fits a multinomial logistic regression for inferential statistics analysis because it is best for modeling unordered categorical dependent variables. The logit equations of the Multinomial Logistic Regression model (MLR) from a comparison the log odds of each of the non-reference K response variables to the categorical variable of choice (Shakhawat et al., 2012), logit :
logπijπi0=logPY=jXiPY=0Xi=Xβ;
Xβ=j=0kxijβj
X=x1,x2,x3,⋯⋯,xn
β=β1,β2,β3,⋯⋯,βkT
The significance level for the study variables from the study variables coefficients tables were all positive and greater than the significance alpha level of 0.05, an indicator that all the variables tested had statistically significant impact on the level of an individual’s happiness in a relationship.
The -2 Log Likelihood of the null model is 1061.743 while that of the model is 568.391 (Table 7). The difference (Chi-square statistics) is1061. 743–568.391= 493.353 that is significant at α=0. 05. Hence, there exists an association between the level of happiness in a rapport and the independent variables used in the study. The regression output for parameter estimates had several important elements. The Walds statistics and the associated probabilities using the logit (β) provide the index of the significance of each predictor in the multinomial logistic equation. There was 4.166 more likelihood for an individual earned $ 0 per month to be very unhappy in their relationship compared to their counterparts who earned some cash. Hence, an increase in one’s economics and financial comfortability results in an increased level of happiness in a relationship. According to a study by Rand et. al., (2010), marital or a relationship gratification is one of the main predictors of a stable rapport. This is greatly influenced by the socio-economic position of an individual, the more economically fit an individual is, the more comfortable financially they are and ultimately they are happy in their relationships. Based on the regression output in the Appendix, for the study hypothesis, there exist a rapport between the dependent and independent variables. Hence, the null hypothesis is rejected and concludes that demographic and socio-economic status of an individual has an influenced on their rapport they establish. The more economically fit and socially vast an individual is, the more they are stable and serious in their relationship and thus implies that there will be a higher level of happiness compared to their counterparts who are not financially stable- struggling and have some issues in their demographic and social well-being.
References
Conger, R. D., Conger, K. J., & Martin, M. J. (2010). Socioeconomic status, family processes, and individual development. Journal of Marriage and Family, 72(3), 685-704.
Conger, R. D., Wallace, L. E., Sun, Y., Simons, R. L., McLoyd, V. C., & Brody, G. H. (2002). Economic pressure in African American families: a replication and extension of the family stress model. Developmental psychology, 38(2), 179.
Hossain, S., Ahmed, S. E., & Howlader, H. A. (2014). Model selection and parameter estimation of a multinomial logistic regression model. Journal of Statistical Computation and Simulation, 84(7), 1412-1426.
Appendix