Comparing Two Groups

Posted: August 26th, 2021

Comparing Two Groups

Name

Institutional Affiliation

Comparing Two Groups

Comparing two groups involves applying a statistical evaluation using an independent t-Tests. As such, an independent t-test refers to a test that conducts a comparison of two groups by utilizing their respective mean values(Montgomery, Peck & Vining, 2012). The variables are continuous, that is, interval or ratio, and normally distributed. The implementation of this model bases on the assumption that there is a difference in the mean scores if the dependent variable established due to the influence of the explanatory or independent variable(Montgomery, Peck & Vining, 2012). Thus, these differences are the once that distinguish the two groups.

Subsequently, the t-test is a family of t-distribution since differences in the mean distribution for a variable that is normally distributed estimates the t-distribution(Montgomery, Peck & Vining, 2012). Therefore, this assessment conducts a comparison of two groups using an independent t-test model on a sample of two groups containing continuous and dichotomous variables, as demonstrated in subsequent parts.

Sample Data

The data below shows the results of two treatments for drinking coffee or soda. It is assumed that drinking coffee makes the hands shakier than drinking soda. The following is the data collected from a study of those who participated in the examination;

Table 1: Data from the Study

Hypothesis test

The null hypothesis, H0: µ1 – µ2 = 0

The alternative hypothesis, HA: µ1-µ2 ≠ 0

Excel analysis is used to assess whether the differences in the means of the two groups are significant at 0.05.

Table 2: Independent t-test results

Table 2 shows the results of the t-test, assuming unequal variances. In this case, the rule of thumb for a two-tail t-test is that if t Stat < -t Critical two-tail or the t Stat > t Critical two-tail, we reject the null hypothesis. As in the case, t Stat is -2.961 while t Critical two-tail is 2.262. Hence, t Stat (-2.961) < -t Critical two-tail (-2.262), reject null hypothesis. It thus implies that the difference in the effect of coffee and soda are statistically significant at the 0.05 significance level.

Reference

Montgomery, D., Peck, E. & Vining, G. (2012). Introduction to linear regression analysis. Hoboken: Wiley.