Question

Part 1. The insulin pump is a device that delivers insulin to a diabetic patient at regular intervals. It presumably regulates insulin better than standard injections. However, data to establish this point are few, especially in children. The following study was set up to assess the effect of use of the insulin pump on HbgA1c, which is a long-term marker of compliance with insulin protocols. In general, a normal range for HgbA1c is <7%. Data were collected on 256 diabetic patients for 1 year before and after using the insulin pump. A subset of the data for 10 diabetic patients is given in the table below. ID HbA1c 1 year HbA1c 1 year before(W) after (1) 1 6.7 2 74 70 74 3 92 8.6 &1 5 74 6 8.1 7 10.8 8 7.1 9 79 10 10.8 70 8.5 77 7.7 Test the hypothesis that the measured HbA1c 1 year after the insulin pump will be different from HgbA1c 1 year before the insulin pump. Set your p-value to indicate statistical significance at <0.05. Type up a report that has the following information: 1. Write out the null and alternative hypothesis 2. Identify the types of variables that are being used - Nominal, ordinal, interval, or ratio? 3. Report the mean and standard deviation for each group 4. Display the data for each group using a histogram with the normal curve included (get the graph from SPSS and insert it into this document) a. List the conditions that should be met for conducting a hypothesis test of a mean b. Determine whether the data are appropriate for conducting a hypothesis test of a mean and indicate how you came to this conclusion 5. Choose and identify the statistical test you will use to answer the hypothesis question assuming the conditions for conducting a hypothesis test of a mean are met 6. Indicate the alpha level, the df and whether it is one- tailed or two-tailed 7. Report t-value, p-value, and 95% confidence interval 8. Make a decision regarding the hypothesis 9. State your conclusion regarding the results of the study

Answer 1

Solution:

NULL HYPOTHESIS H0:

ud = 0

ALTERNATIVE HYPOTHESIS Ha:

07 prl

t cal=1.432

Paired Samples Test Paired Differences 95% Confidence Interval of the Difference Mean 65000 df Std. Error Mean 45393 Std. Deviation 1.43546 Lower - 37687 Upper 1.67687 Sig. (2-tailed) .186 Pair 1 BEFORE - AFTER 1.432 9

P value= 0.186

Since P value > 0.05 therefore NOT significant

Decision: DO NOT REJECT H0.

Conclusion: We don't have sufficient evidence to conclude that the measured HgbA1c 1 year after the insulin pump will be different from HgbA1c 1 year before the insulin pump.

The type of variable that are being used is RATIO

4 Mean =8.50 Std. Dev. =1.506 N =10 3- Frequency 27 09 6.00 7.00 8.00 9.00 10.00 11.00 BEFORE

3- Mean -7.85 Std. Dev. =0.898 N =10 2- Frequency 1 0 6.00 7.00 8.00 9.00 10.00 AFTER

a) the conditions that should be met for conducting a hypothesis test of a mean:

DATA MUST BE NORMALLY DISTRIBUTED The population data are symmetric, uni modal, without outliers, and the sample size is 15 or less.

b) The data is appropriate to use . All the conditions have met.