Abstract
The study was performed to determine the effect of marital status on the body mass index. The data used contained a sample size of 6878 adults from England aged above eighteen years. The data was collected through a Health Survey that was performed in 2011. The data was analysed using t-test, one-way ANOVA with Tukey HSD post hoc test, and the linear regression. The study found that there was no correlation between marital status and the body mass index. However, the effect of marital status on the body mass was statistically significant even though marital status accounted for 0.2% of changes in the body mass index. The resultant model was Body Mass Index = 27.156 + 0.114(marital status).
Introduction
The body mass index serves as an indicator of body fatness by dividing their weight in kilograms with the square of their height (Sattar, Baig, Rehman & Bashir, 2013, p.956). It is an indirect measure of determining body fatness. The body mass index is a product of the lifestyle one leads, particularly with regards to choices in dietary intake and physical activity (French, Story & Jeffery, 2001, p.309; Burke et al., 2003, p.421; Centers for Disease Control and Prevention, 2011, p.28). Nonetheless, the body mass index is also affected by numerous other factors as argued by (Sattar et al., 2013, p.956). Exploring the factors that affect the body mass index is of social importance because as reported in the empirical study performed by Yetubie, Haidar, Kassa & Fallon, (2010, p.321), a high or low body mass index is associated with poor health outcomes.
Literature
The effect that the body fat has on health has necessitated scientific inquiries that different factors have on the body mass index. One such study was performed by Asil et al., (2014, p.255) in which the sought to determine the factors that influenced the body mass index in adults. In a sample of 498 adults, the researchers found that the predictors of body mass index included parity, age, marital status, educational level, sleep duration, and smoking status.
Asil et al., (2014, p.255) also used a regression model to rank the predictors of body mass index based on their importance. Asil et al., (2014, p.255) reported that the most important predictor was age, educational level, and marital status for which the p-value was less than 0.001. Smoking status and sleep duration followed in importance and their p-value was less than 0.05. the study by Asil et al., (2014, p.255) identified some age, educational level, marital status, smoking status, and sleep duration as some of the factors that affect an individual’s body mass index.
Other studies into the predictors of the body mass index have also highlighted the factors identified by Asil et al., (2014, p.255) as significant to a person’s body mass index. For instance, Sattar, Baig, Rehman & Bashir (2013, p.956) also identified some of these factors as predictors of one’s body mass index. In their study, the researchers used a descriptive cross sectional design and a sample of 493 people. One of the factors influencing the body mass index identified in the study was sex with more men (25.4%) of men having a body mass index of 18.5 and below compared to 11.5% of the women respondents (Sattar, Baig, Rehman & Bashir 2013, p.956)
The study also found that one’s marital status was a significant influencing factor as evidenced by 22.9% of married people scoring above 30 on their body mass index compared to 6.6% of the single people who were no married. Smoking was also identified a s significant predictor of body mass index. The empirical findings showed that 21.1% of the people who smoked recorded a body mass index of above 30 compared to 18.3% of people who did not smoke.
The findings by the researchers above showed that marital status was a significant influencer of the body mass index. More married people than unmarried people were obese, meaning that they had a high body mass index. Owing to these findings, it is important to explore marital status as a specific factor that influences body mass index. Numerous studies have focused their research efforts on this specific factor (The, Gordon-Larsen, 2009, p.1441).
A study by Teachman (2013, p.74) found that marital status was a significant factor influencing one’s body mass index. More precisely, the findings of the study found the people who did not live with partners because they were either not married or were divorced were associated with a low body weight. A low body weight is associated with a low body mass index (Teachman, 2013, p.74). The findings also showed that people who were married and cohabiting typically weighed more compared to their divorced and unmarried counterparts. More weight, especially if it located on the abdomen region is associated with a higher body mass index (Teachman, 2013, p.74).
A study by Umberson, Liu & Powers (2009, p.327) also advanced the argument on the effect of marital status on the body mass index by measuring the effect of both the status and transitions in marital status on one’s body weight. The findings showed that people who were continuously widowed was associated with continued weigh loss when compared to people who were continuously married (Umberson, Liu & Powers, 2009, p.327). The researchers also found that those who transitioned into marriage weighed more and had a higher body mass index compared to people who remained unmarried in the same period. The researchers also found that people who transitioned from marriage to divorce or widowhood weighed less when compared to the married counterparts (Umberson, Liu & Powers, 2009, p.327).
Similar findings were also reported by Averett, Sikora & Arqys (2008, p.330) in studty where the authors sought to establish the relationship between body mass index and one’s relationship status. The researchers found that the body mass index increase during the period of marriage for both men and women. The same findings were reported in relationships where the partners were cohabiting (Averett, Sikora & Arqys 2008, p.330).
The studies have shown that one’s marital status is an important indicator of their body mass index. The review of literature also showed that the transition from marriage also affects one’s body mass index. More precisely, transitioning into a marriage or cohabiting relationship results in a higher body mass index while transitioning out of a marriage into either divorce or widowhood was associated with a reduction in the body mass index.
Theory
The aim of the paper is to determine the marital status differences as a factor influencing one’s body mass index. To achieve the purpose of the paper, the paper will be guided by the following objectives.
The working null and alternative hypothesis are as follows:
Objective 1
Null hypothesis: There are no differences in body mass index by marital status
Alternative hypothesis: There are differences the body mass index by marital status.
Objective 2
Null hypothesis: There is no statistically significant difference in the body mass index by marital status.
Alternative hypothesis: There is a statistically significant difference in the body mass index by marital status.
Objective 3
Null hypothesis: The marital status has no statistically significant effect on one’s body mass index.
Alternative hypothesis: Marital status has a statistically significant effect on one’s body mass index.
The hypotheses that are tested in this paper are derived from the information gained from the review of literature. As highlighted in the last paragraph of the literature review, the studies showed that there were significant differences in the body mass index of the unmarried people when compared to the married and cohabiting people.
Methodology & Data
The analysis of the data will seek to test all the hypotheses outlined above so as to meet the objectives that guide the paper. The analysis of data will use both descriptive and inferential statistics. For instance, testing the null and alternative hypothesis of objective 1 will require descriptive statistics where mean as a measure of central tendency will be used in determine the differences in the body mass index by the marital status. The test of the null and alternative hypothesis in objective two will be done using inferential statistics. This is because much more than the description of the data is required (Park, 2005).
There is a need to determine if the difference between the body mass index by marital status is statistically significance. For this purposes, a one-way ANOVA will be performed. ANOVA is an appropriate statistical test for determining whether the difference between calculated means of independent groups are statistically significant (Park, 2005). ANOVA tests will offer information on whether the mean differences within and between groups are statistically significant. However, they do not tell which of the variables within and between the groups had significant variances in their means. Performing a Tukey HSD post hoc test offers information on the variables where there are statistically significant mean differences (Park, 2005).
Testing the hypothesis in objective 3 will be done using a linear regression. Linear regression is appropriate for determining the effect of one variable on another by calculating the relationship between the explanatory variables and the scalar dependent variable (Park, 2005). The linear regression uses the following equation, Y = a + bX. In this equation, X represents the explanatory variable while Y represents the dependent variable. While b represents the gradient of the line, a represents the y intercept. The argument is that the body mass index as the dependent variable is affected by marital status, the independent or explanatory variable multiplied by the gradient plus the y intercept. These calculations help determine the effect that marital status has on the body mass index.
Results
One of the objectives was to determine if there were mean differences between the body mass index of the various participants as grouped by their marital status. The analysis was performed using t-tests to show that there were differences in the mean body mass index for the various groups as illustrated in Table 1 below. There was also an objective to determine whether the mean differences between the single participants and the other groups based on marital status were statistically significant. The application of the ANOVA test and a TUKEY HSD post hoc test showed that there were statistically significant differences in the body mass index of some of the groups. For instance, Table 2 shows that the mean differences between the body mass index of the single participants and the body mass index of the separated, divorced, widowed, and cohabitees were statistically significant.
A linear regression was performed to achieve the third objective. The R statistic from the model summary was 0.043. This is an indication that there was no correlation between marital status and the body mass index. The R Square statistic was 0.002 as shown on Table 3. This is an indication that the independent variable (Marital Status) explained the dependent variable (body mass index) by 0.2%.
An ANOVA test was also performed to determine whether the regression model that was used in the analysis of the data predicted the dependent variable in a statistically significant manner. As shown in Table 4 the p-value in the sig. column showed the model fitted the data well and the manner in which it predicted the dependent variable was statistically significant. As highlighted previously, the format for the regression equation was Y = a + bX. Using the coefficients, it is discernable that the model for predicting the effect of marital status on the body mass index is Body Mass Index = 27.156 + 0.114(marital status) as shown in Table 5.
Conclusions & Recommendations
The review of studies showed that the body mass index is influenced by various factors. Marital status was one of the factors that was highlighted as a significant predictor of body mass index. Various studies showed that that there were statistically significant mean differences of body mass index between single people and married people. The themes discussed in the literature showed that married people were heavier, and hence a higher body mass index compared to their unmarried counterparts.
Some of the studies reviewed also highlighted the effect of transitions of one’s marital status on their weight, and by extension, their body mass index. One particular finding was that people transitioning into marriage tended to gain weight over a given period while people transitioning out of marriage, either through divorce, separation, or death lost weight during the same period. This would indicate that the people transitioning into marriage were associated with a higher body mass index compared to people who transitioned out of marriage, either through divorce, separation, or death who were associated with a smaller body mass index.
Based on these findings, the paper analyzed a data set containing information of over 6000 participants. The information was collected through a Health Survey performed by adults aged above eighteen years in England in 2011. The analysis of the data was performed using t-tests, one-way ANOVA, and linear regression. The analysis showed that the mean of the body mass index of the various groups as per their marital status were statistically significant.
The ANOVA tests showed that mean differences of the body mass index of the single and the married, separated, divorced, widowed, and cohabitees was statistically significant. The inclusion of the separated, divorced, and widowed can be explained by citing one of the limitations of the data set. The data set did not contain information on the length of time between the survey and the separation, divorce, and widowhood of the participants in these classifications. There was a time aspect in the arguments in the review of literature regarding to the statistical difference in the body mass index of the married and those transitioning out of marriage through separation, divorce, and widowhood.
The regression analysis showed that there was no correlation between marital status and body mass index, and that any explanation of the body mass index that was attributed to marital status was only by by 0.2%. However, the ANOVA test showed that the regression model [Body Mass Index = 27.156 + 0.114(marital status)] predicted the body mass index in a statistically significant manner. It is recommended to determine how marital status influences the body mass index.
References
Asil, E., Surucuoglu, M., Cakiroglu, F., Ucar, A., Ozcelik, A., Yilmaz, M., and Akan, L., 2014. Factors That Affect Body Mass Index of Adults. Pakistan Journal of Nutrition, 13: 255- 260.
Averett, S., Sikora, A. and Arqys, L. 2008. For better or worse: relationship status and body mass index. Economics and Human Biology. 6(3): 330-349.
Burke, V., Giangiulio, N., Gillam, H., Beilin, L. and Houghton, S. 2003. Physical activity and nutrition programs for couples: a randomized controlled trial. Journal of Clinical Epidemiology. 56(5): 421–432,
Centers for Disease Control and Prevention. 2011. School health guidelines to promote healthy eating and physical activity. Morbidity and Mortality Weekly Report. 60(5): 28–33.
French, S., Story, M. and Jeffery, R. 2001. Environmental influences on eating and physical activity. Annual Review of Public Health, 22: 309–335.
Park, H. 2005. Comparing Group Means: The T-test and One-way ANOVA Using STATA, SAS, ~ and SPSS. [PDF] Available at :< http://stat.smmu.edu.cn/DOWNLOAD/ebook/statistics_course.pdf> [Accessed 4 January 2017]
Sattar A, Baig S, Rehman, N. and Bashir B. 2013. Factors affecting BMI; Assessment of the effect of sociodemographic factors on BMI in the population of Ghulam Mohammad Abad Faisalabad. Professional Medical Journal. 20(6): 956-964.
Teachman, J. 2013. Body Weight, Marital Status, and Changes in Marital Status. Journal of Family Issues. 37(1): 74-96.
The, N. and Gordon-Larsen, P. 2009. Entry into romantic partnership is associated with obesity. Obesity. 17 (7) 1441–1447.
Umberson, D., Liu, H. and Powers, D. 2009. Marital Status, Marital Transitions, and Body Weight. Journal of Health and Social Behavior. 50(3): 327-343.
Yetubie, M., Haidar, J., Kassa, H., and Fallon, F. 2010. Socioeconomic and Demographic Factors Affecting Body Mass Index of Adolescents Students Aged 10–19 in Ambo (a Rural Town) in Ethiopia. International Journal of Biomedical Science. 6(4): 321-326.
