The chi-square test is used with the categorical data while the t-test is used for continuous data. In both tests, a p-value is determined to ensure the significance of the tests. In this report, the tables comparing the population of a category that has diabetes and that do not have (McDonald, 2014). The data presented in the tables are obtained from gender, race, salary, education, height, weight, BMI, allergies, family history diabetes, family history allergies.
Chi-Square Test
The purpose of a Chi-square test is to test the dependency of the variables from two groups of data. It shows the difference that exists among the observed and expected counts of data and if the populations are related. The following table shows the values of the categories of data.
Delegate your assignment to our experts and they will do the rest.
| Gender | Chi-Square Value | P Value | |||
| Female | Male | ||||
| Diabetes | Total | ||||
| Yes | 56 | 53 | 109 | 0.09 | 0.7642 |
| No | 103 | 88 | 191 | ||
| % with Diabetes (Yes) | 51% | 49% | 100% | ||
There are significantly more women (51%) who have diabetes than men (49%). The p-value is 0.7642, which is greater than 0.05, implying that the null hypothesis is not rejected (McDonald, 2014) . Thus, there is an insignificant difference among the groups.
| Race | ||||||||
| White | African Amer. | Hispanic | Asian Amer. | Native Amer./Other | Total | Chi-Square Value | P Value | |
| Diabetes | ||||||||
| Yes | 57 | 13 | 24 | 7 | 8 | 109 | 1.74 | 0.7834 |
| No | 107 | 22 | 31 | 14 | 17 | 191 | ||
| % with Diabetes(Yes) | 52% | 12% | 7% | 22% | 6% | 100% | ||
There are significantly more whites (52%) who have diabetes than African American (12%), Hispanic (24%), and Asian and Native Americans (15%). The p-value is greater than 0.05, implying that there is an insignificant difference between the groups.
| Education | |||||||
| High School | College | Masters | Professional | Chi-Square Value | P Value | ||
| Diabetes | Total | ||||||
| Yes | 16 | 42 | 27 | 24 | 109 | 2.67 | 0.4453 |
| No | 30 | 59 | 62 | 40 | 191 | ||
| % with Diabetes(Yes) | 15% | 39% | 25% | 22% | 100% | ||
There are significantly (highest percentage of college tier (39%) who have diabetes. The p-value is greater than 0.05, implying that there is an insignificant difference between the groups.
| Allergies | Chi-Square Value | P Value | |||
| Yes | No | ||||
| Diabetes | Total | ||||
| Yes | 67 | 42 | 109 | 0.01 | 0.9203 |
| No | 115 | 76 | 191 | ||
| % with Diabetes(Yes) | 61% | 39% | 100% | ||
There are significantly more people with allergies (61%) who have diabetes than those who do not have allergies (39%) (McDonald, 2014). The p-value is greater than 0.05, implying that there is an insignificant difference between the groups.
| Family History of Diabetes | Chi-Square Value | P Value | |||
| Yes | No | ||||
| Diabetes | Total | ||||
| Yes | 89 | 20 | 109 | 140.77 | <0.0001 |
| No | 23 | 168 | 191 | ||
| % with Diabetes(Yes) | 82% | 18% | 100% | ||
There are significantly more people with a family history of diabetes (82%) who have diabetes than people who do not have family history of diabetes (18%). The p-value is less than 0.05, implying that there is significant difference among the groups.
| Family History of Allergies | Chi-Square Value | P Value | |||
| Yes | No | ||||
| Diabetes | Total | ||||
| Yes | 0 | 109 | 109 | 0.09 | 0.7642 |
| No | 148 | 43 | 191 | ||
| % with Diabetes(Yes) | 0 | 100% | 100% | ||
There are no people with a family history of allergies (0%) who have diabetes and all people who do not have a family history of allergies (100%) have diabetes. The p-value is greater than 0.05, implying that there is an insignificant difference between the groups.
T-Test
The t-test is used to compare means for two groups (McDonald, 2014). The following are the tables used to compare the means for the different pairs of data.
| Salary | |||
| Mean | T Value | 2 Tail P Value | |
| Diabetes | 7.79 | <0.0001 | |
| Sample A Yes | $70,226.45 | ||
| Sample B No | $45,522.11 | ||
The average salary of those with diabetes is $70,226.45 while for those without diabetes is $45,522.11 .Those with diabetes were significantly good earners (p<0.05) (Van Eck, 2014) .
| Height | |||
| Mean | T Value | 2 Tail P Value | |
| Diabetes | 16.63 | <0.0001 | |
| Sample A Yes | 70.422 in | ||
| Sample B No | 65.0209 in | ||
The average height of those with diabetes is 70.422 inches while for those without diabetes is 65.021 inches . Those with diabetes were significantly taller (p<0.05).
| Weight | |||
| Mean | T Value | 2 Tail P Value | |
| Diabetes | 16.33 | <0.0001 | |
| Sample A Yes | 187.87 lbs | ||
| Sample B No | 142.70 lbs | ||
The average weight of those with diabetes is 187.87 lbs while for those without diabetes is 142.70 lbs . Those with diabetes significantly weighed more than the other group (p<0.05).
| BMI | |||
| Mean | T Value | 2 Tail P | |
| Diabetes | 14.35 | <0.0001 | |
| Sample A Yes | 26.5239 kg/m2 | ||
| Sample B No | 23.5628 kg/m2 | ||
The average BMI of those with diabetes is 26.5239 kg/m2 while for those without diabetes is 23.5628 kg/m2 . Those with diabetes had significantly higher BMI than the other group (p<0.05).
References
McDonald, J. H. (2014). Paired t-test. In Handbook of biological statistics (3rd ed.). Sparky House Publishing, Baltimore, Maryland. Accessed at http://www.biostathandbook.com/pairedttest.html
McDonald, J. H. (2014). Chi-square test of independence. In Handbook of biological statistics (3rd ed.). Sparky House Publishing: Baltimore, Maryland. Accessed at http://www.biostathandbook.com/chiind.html
Van Eck, N. (2014). The decision tree for statistics. Based on Andrews, F. M., Klem L., Davidson, T. N., O'Malley, P., and Rodgers, W. L.: A guide for selecting statistical techniques for analyzing social science data: The decision tree for statistics. The University of Michigan. Retrieved from http://www.microsiris.com/Statistical%20Decision%20Tree/how_many_ variables.htm
