**Calculate the appropriate descriptive statistics for thefollowing variables comparing diabetes with no diabetes status: gender, race, salary,**

education, height, weight, BMI, allergies, family history diabetes, family history allergies.

For chi-square tests, report the chi-square value and the p-value (if p-value < 0.05, then the

test is significant). For t-tests, report the t-test value and the p-value. Include a 2-3 page

description of the descriptive statistics including tables of the summarized data, similar to a

“Results” section in a published manuscript or journal article. Use the following online

calculators to obtain the results for this analysis.

Chi-Square for Categorical Data: (Choose “Frequency Data”

from the far left, then “Chi-Square, Cramer’s V, and Lambda” from the middle of the

page)

Enter in the number of people in each category (e.g. number of women who have diabetes,

number of men with diabetes, etc.). Example of a table below:

SLP Assignment

Descriptive Statistics

Using the provided dataset in Excel, descriptive statistics that are appropriate for variables

concerning to diabetes including gender, race, salary, height, weight, as well as BMI. The

descriptive statistics calculated using the provided dataset specifically include mean, standard

deviation, variance as well as media. These descriptive statistics are mainly concerned with

analysis of measurement of central tendency i.e. mean and median as well as measurement of

variation i.e. standard deviation and variance.

Table 1: Descriptive Statistics

Age Salary Height Weight BMI To feel

depressed

during the

winter

To

exercise

during

the

summer

To overeat

when

stressed out

Mean 50 $54,498 66.98333 159.1133 24.63867 2.993333 3.373333 2.77

Standard

Deviation

20 28923.783 3.750102 31.66517 2.231375 1.226773 1.227046 1.315116638

Variance 401 836585199 14.06327 1002.683 4.979035 1.504972 1.505641 1.729531773

Median 50 $50,012 67 161 25 3 4 3

In particular, this SLP assignment will be analyzed the provided dataset using chi-square tests

and t-test. For the chi-square tests apart from the descriptive statistics, the report will also include

chi-square value as well as the p-value. On the other hand, for the t-tests the report will include

the t-test value as well as the p-value.

In addition, the specific numbers of people in the provided the dataset within their specific

category i.e. diabetes and no diabetes are determined in order to enable the data analysis to be

carried out. A summary of those statistics is presented in the table shown below:

Table 2: Data Summary

Diabetes No Diabetes Total Percentages

Female 56 103 159 53%

Male 53 88 141 47%

Total 109 191 300

Percentages 36.3% 63.7% 100%

Based on the statistics presented in the above table concerning the chi-square obtained from the

VassarStats website which is used for statistical computation, particularly in the context of Chi-

Square for Categorical Data and specifically using Chi-Square, Cramer’s V, and Lambda in a 2 x

2 table; there are some inferences that can already be done. Some of the inferences based on

percentages include:

There are significantly more women (53%) who have diabetes than men (47%).

Additionally, the results of the chi-square test show that the chi-square value is 0.09 and the p-

value is <0.0001 an indication that the test is significant meaning that there a significant

difference between the number of women who are diabetic compared to men who are diabetic.

T-Tests for Continuous Data

The t-test was used to compare the two groups i.e. Sample A (no diabetes) and Sample B

(diabetes) and the t-test reported the t-test value as well as p-value. The t-test values for variables

such as age, height, weight as well as BMI are reported in the table shown below. In addition, the

two-tailed p-values are also shown and the are all below <0.05 and indication that the tests are

significant which means there are significant differences between the two groups (i.e. Sample A

and Sample B) with regards to the considered variables.

Table 3: Data Summary

A B Total t-test

value

Two-

tailed p-

value

N 191 109 300

Age Mean 39.0052 70.5229 50.4567 -20.69 <0.0001

Height Mean 65.0209 70.422 66.9833 -16.63 <0.0001

Weight Mean 142.7016 187.8716 159.1133 -16.33 <0.0001

BMI Mean 23.5628 26.5239 24.6387 -14.35 <0.0001

The average age of those without diabetes is 39 years and for those with diabetes is 70.5 years.

Those with diabetes were significantly older/younger (p<0.05).

The average height of those without diabetes is 65.02 centimeters and for those with diabetes is

70.4 centimeters. Those with diabetes were significantly shorter/taller (p<0.05).

The average weight of those without diabetes is 132.7 lbs and for those with diabetes is 187.8

lbs. Those with diabetes were significantly heavier/lighter (p<0.05).

The BMI of those without diabetes is 23.6 and for those with diabetes BMI is 26.5. The BMI for

those with diabetes is significantly higher/lower (p<0.05).

References

Corder, G. W. & Foreman, D. I. (2014). Nonparametric Statistics: A Step-by-Step Approach.

New York, NY: Wiley.

Greenwood, P. E. & Nikulin, M. S. (1996) A guide to chi-squared testing. New York, NY:

Wiley.

Markowski, C. A. & Markowski, E. P. (1990). Conditions for the Effectiveness of a Preliminary

Test of Variance. The American Statistician, 44(4), 322–326.

Sawilowsky, S. S. (2005). Misconceptions Leading to Choosing the t Test over the Wilcoxon

Mann–Whitney Test for Shift in Location Parameter. Journal of Modern Applied

Statistical Methods, 4(2), 598–600.

VassarStats (2015). Procedures Applicable to Categorical Frequency Data. (Accessed on November 26 2015).

VassarStats (2015). t-Tests & Procedures.

Zimmerman, D. W. (1997). A Note on Interpretation of the Paired-Samples t Test. Journal of

Educational and Behavioral Statistics, 22(3), 349–360.