Does Exercise Help Lower Sugar Levels?

Running head: APPLICATION OF DATA ANALYSIS 1

APPLICATION OF DATA ANALYSIS 2

Application of Data Analysis
Serena Mahoney
C214
01/14/2016

Most health care professionals who focus on patient safety are very much familiar with statistics that are alarming and frequently cited from the Medicine institute. Errors made by medical practitioners result to death of patients ranging from 44,000 to 98,000 in a year. One of the ways to provide good care and treatment of diabetics is by exercising and having appropriate diet. Hospitals advocate the use of correct diet and exercising (self-care) as one of the ways of lowering blood sugar levels (Agresti & Kateri, 2011). To test this hypothesis, a certain hospital designed an experiment to test if diabetic self-care would help lower patients’ blood sugar levels. The situation will benefit from data analysis because a real impact of the effects of diabetics will be known and how exercise helps to reduce the sugar levels.
Data was collected from interviewing patients and medical practitioners, a sample population responded to questionnaires that fielded various topics to enable the gathering of raw data from the population. Observation of various trends, reading documentation, research on mainstream and other media also provided evidence on the situation and data to be evaluated. To test if this was effective, 50 diabetic patients were selected randomly from different hospitals to enroll in a six-month course while other fifty served as a control group. After the course, blood sugar tests were done on both groups.
The data that will be collected will focus on socioeconomic, demographic, dietary and health-related questions and the six-month will begin from February to August. In this case, data analysis will provide detailed and deep information by recording patient’s feelings and any improvements and enhance risk analysis (Casella & Berger, 2002). Data analysis will also lead to accuracy by lowering assumptions and errors since it is a health situation.
The data collected is quantitative data where 50 patients are selected randomly for each group. Thus, our sample size is 50 for both groups. Interviews are carried out on the sample of patients selected while some questionnaires are developed to help develop detailed information about the situation. Observation is also a great method that was used in the gathering methodology. We also used experimental method to gather information for analysis or rather to collect data required since our study aims to understand cause-and-effect relationship and it is ‘controlled’.
T-test as an analysis technique for this type of data, as the sample size is small to estimate the sample parameters. This is because we have two independent samples, and the frequency in each group is normally distributed. The t-test is used because we can compare the means of the two samples we are using in the analysis, this technique is appropriate because it is used to compare the actual difference that exist between the two means of the sample as it is related to the variation in the data. The data was collected in intervals, and the standard deviation of each group is different from the other making a t-test the best technique to analyze the data collected. The t test gives the ability to formulate the hypothesis upon determination of the p value. Thereon we can either go by the alternative hypothesis or accept the null hypothesis. A one sample t test will be undertaken to assist in the hypothesis testing.

Some of the information collected is shown below;

Mean
Standard deviation
n

Diabetic course group
7.1
0.9
50

Control group
6.5
0.7
50

Diabetic course group = n1
Control group = n2
Given ? = 0.01
We have
H0: ?1 = ?2 (diabetic course has no significant difference)
H1:?1 < ?2 (diabetic course has a significant effect) The results are, t-value = -3.72, p-value =0.0002 p< ? that is, 0.0002<0.01, thus we reject the null hypothesis and accept the alternative hypothesis. The calculated t value exceeded the tabulated value and therefore the means are significantly different at the probability level used (Nelson, 2009). The level of significance is 0.01 that means that when the patients exercise the blood sugar levels will decrease with time. This implies that the diabetic course lowers patients’ blood sugar level at 0.01 level of significance. Hence the diabetic course should be highly recommended for helping the diabetic patients recover. References Agresti, A., & Kateri, M. (2011). Categorical data analysis (pp. 206-208). Springer Berlin Heidelberg. Casella, G., & Berger, R. L. (2002). Statistical inference (Vol. 2). Pacific Grove, CA: Duxbury. Nelson, W. B. (2009). Accelerated testing: statistical models, test plans, and data analysis (Vol. 344). John Wiley & Sons.