Project 5: Analyzing Efficacy of Handwashing Methods

Course: Public Health Computing
Instructor: Dr. Robert Parker, PhD
Semester: Spring 2022

Background:

Washing hands is an essential part of our everyday life in all fields and circumstances to keep ourselves, healthy, clean, and prevent infectious diseases. We have learned this more so than ever during the COVID-19 pandemic. Removing bacteria, in general, is significant to our overall health. The data is gathered from an experiment. The four different methods that have been tested are: washing with water only, washing with regular soap, washing with antibacterial soap, and spraying hands with antibacterial spray (containing 65% ethanol as an active ingredient). Hence, the purpose of this study is to determine the effectiveness of these four methods in removing bacteria from our hands through statistical analyses.

Source and Sample:

The dataset for this statistical analysis comes from the Data and Story Library (DASL). This specific dataset stems from a student investigator who wanted to determine the most effective hand washing method experimentally. The dataset contains 32 cases, with each treatment being repeated four times.

Methodology:

To carry out the appropriate statistical tests for studying the relation between bacterial counts and the methodology of washing, R was used. In this programming, we specifically conducted a one-way ANOVA test, the Tukey’s Honestly Significant Difference (HSD) test, and the Kruskal-Wallis test. The ANOVA test helped us determine any differences in the means of the methodology of handwashing at each level. The Kruskal-Wallis test was conducted to confirm the assumptions of ANOVA. Since the ANOVA statistical analysis showed for significant differences, it did not show where those differences lie between the methods. Hence, Tukey’s HSD test was executed to determine those differences.

Key Results:

The summary statistics of the data showed that the average bacterial count from all methods of hand washing was 88.25, but the count itself ranged from 5 to 207. The Method variable in the printed code shows a length of 32, which is the number of cases. Water had the highest mean of 117, which translates to the highest average bacterial counts, making it likely an insufficient method to remove bacteria when washing. The one-way ANOVA test showed that there is a difference in the mean bacterial counts among the four washing methods. The p-value of 0.001667 from the Kruskal-Wallis test indicates that there is enough evidence to reject the null hypothesis that the group medians are equal. Tukey's HSD showed that Alcohol Spray seems to be the most effective since all pairwise comparisons with it show significant differences with the other means.