Descriptive Statistics Project Part 0: Project Introduction Roughly every two years, the CDC (Center for Disease Control and Prevention) performs the National Health and Nutrition Examination Survey (NHANES). The following is information about the NHANES as stated on the CDC website 1 The National Health and Nutrition Examination Survey (NHANES) is a program of studies designed to assess the health and nutritional status of adults and children in the United States. The survey is unique in that it combines interviews and physical examinations...The survey examines a nationally representative sample of about 5,000 persons each year. These persons are located in counties across the country, 15 of which are visited each year. In this project, you will analyze the results of a fasting two hour oral glucose test taken from the 2015-2016 Assessment Period for sampled individuals that are aged 20 and older who were not "currently on diabetic medications or pregnant." It is also helpful to know the following information. Per the CDC2, a fasting glucose (blood sugar) test (in milligrams per deciliter) of 140 mg/dL or below is Normal 140 - 199 mg/dL indicates Prediabetes 200 mg/dL or above indicates that the patient may have diabetes The file you will use is named NHANES 2015-2016 Oral Glucose Tolerance Data and can be accessed using the link below. Note that you can not edit the file, so you will need to make a copy of it to your Drive in order to use it. NHANES 2015-2016 Oral Glucose Tolerance Data 1 https://www.cdc.gov/nchs/nhanes/about_nhanes.htm 2https://www.cdc.gov/diabetes/basics/getting-tested.html STAT 243Z Page 1 of 8 Descriptive Statistics Project

Project Reminders You can use Google Sheets, Excel, and/or Desmos to create graphs and compute values in this assignment. Also, be sure to include your graphs! For this project, the following functions in Google Sheets may be helpful. To read more about each, you can visit the corresponding links. COUNT https://support.google.com/docs/answer/3093620?hl=en COUNTIF https://support.google.com/docs/answer/3093480?hl=en COUNTIFS https://support.google.com/docs/answer/3256550?hl=en Additionally, here is a link to the Google Sheet help site https://support.google.com/docs/topic/9054603?hl=en&ref_topic=1382883 Keep in mind that answers to questions need to be justified. However, keep explana- tions/justifications BRIEF!!!! You must submit YOUR OWN WORK. What you submit must be a complete representation of your knowledge when working on your own. You may NOT use any resources outside of what is linked to or mentioned in Canvas or this document. You may NOT use the Math Center even though it is linked to in Canvas. You may NOT work with any other person or entity on this project Any inkling that this has occurred will result in a 0 and a referral to the Office of Student Conduct. STAT 243Z Page 2 of 8 Descriptive Statistics Project

Part 1: Design Introduction For this part, you will use the datafile NHANES 2015-2016 Oral Glucose Tolerance Data which consists of only those sampled individuals that are aged 20 and older who were not currently on diabetic medications or pregnant. It is also helpful to know the following information. Per the Sample Design procedures documentation the CDC states3 The NHANES sample represents the noninstitutionalized civilian population residing in the 50 states and the District of Columbia. Since 1999, the sample design has consisted of multi-year, stratified, clustered four-stage samples, with data release in 2-year cycles. The NHANES sample is drawn in four stages: (a) PSUs (counties, groups of tracts within counties, or combinations of adjacent counties), (b) segments within PSUs (census blocks or combinations of blocks), (c) dwelling units (DUs) (households) within segments, and (d) individuals within households. PSUs are sampled from all U.S. counties. Screening is conducted at the DU level to identify sampled persons (SPs), based on oversampling criteria. 1.1 Getting Started Use the information provided in the dataset, the Project Introduction, and the Design Introduction to answer the following in 1.1.1 and 1.1.2. 1.1.1 1) 2) 3) 4) 1.1.2 1) Understanding the Data Describe the intended population Describe the sample. Describe all variables. Be sure to state the type of variable for each. Of the variables described, which are "variables of interest." (i.e. variables corresponding to the focus of the oral glucose study) Understanding the Design Is the National Health and Nutrition Examination Survey an observational study or an experiment? 3https://www.cdc.gov/nchs/data/series/sr_02/sr02-184-508.pdf STAT 243Z Page 3 of 8 Descriptive Statistics Project

Part 2: Describing the Sample Overall Introduction For this part, you will continue to look at the data file NHANES 2015-2016 Oral Glucose Tolerance Data. You will focus on appropriately analyzing the Two Hour Glucose results given in mg/dL as well as the Generalized Results for the sample. 2.1 Numerical Summaries 1) Compute the mean. Round to two decimal places 2) Interpret the mean. 3) Compute the standard deviation. Round to two decimal places 4) Interpret the standard deviation. 5) Compute the five number summary. 6) Compute the IQR. 7) Compute the proportion that are "prediabetes." Round to three decimal places. 8) Compute the proportion that are "possible diabetic." Round to three decimal places. 2.2 Graphical Summaries 1) Create a boxplot of the data. 2) Create a histogram of the data. Make sure the minimum value is 0, and the bucket size is 25. 3) What is the shape of the data? 4) Create a bar chart of the data. STAT 243Z Page 4 of 8 Descriptive Statistics Project

2.3 Overall Summary 1) Use your results from 2.1 Numerical Summaries and 2.2 Graphical Summaries to write 3-5 sentences summarizing your findings. 2) Describe at least one potential lurking/confounding variable for the work done in this part. 3) Which are better measures of center and variability for the cholesterol totals: mean and standard deviation or median and IQR? STAT 243Z Page 5 of 8 Descriptive Statistics Project

Part 3: Describing the Sample by Age Introduction For this part, you will continue to look at the datafile NHANES 2015-2016 Oral Glucose Tolerance Data. You will first need to separate the data into three age groups: 20-39, 40-59, 60+. Then you will focus on on appropriately analyzing the Two Hour Glucose results given in mg/dL as well as the Generalized Results for each age group. 3.1 Summaries by Age 3.1.1 Summaries for the 20-39 Age Group 1) State the size of the sample consisting of those aged 20-39. 2) Compute the mean for those aged 20-39. Round to two decimal places 3) Compute the standard deviation for those aged 20-39. Round to two decimal places 4) Compute the five number summary for those aged 20-39. 5) Compute the IQR for those aged 20-39. 6) Compute the proportion that are "prediabetes" for those aged 20-39. Round to three decimal places. 7) Compute the proportion that are "possible diabetic" for those aged 20-39. Round to three decimal places. 8) Create a histogram of the data for those aged 20-39. STAT 243Z Page 6 of 8 Descriptive Statistics Project

3.1.2 Summaries for the 40-59 Age Group 1) State the size of the sample consisting of those aged 40-59. 2) Compute the mean for those aged 40-59. Round to two decimal places 3) Compute the standard deviation for those aged 40-59. Round to two decimal places 4) Compute the five number summary for those aged 40-59. 5) Compute the IQR for those aged 40-59. 6) Compute the proportion that are "prediabetes" for those aged 40-59. Round to three decimal places. 7) Compute the proportion that are "possible diabetic" for those aged 40-59. Round to three decimal places. 8) Create a histogram of the data for those aged 40-59. 3.1.3 Summaries for the 60+ Age Group 1) State the size of the sample consisting of those aged 60+. 2) Compute the mean for those aged 60+. Round to two decimal places 3) Compute the standard deviation for those aged 60+. Round to two decimal places 4) Compute the five number summary for those aged 60+. 5) Compute the IQR for those aged 60+. 6) Compute the proportion that are "prediabetes" for those aged 60+. Round to three decimal places. 7) Compute the proportion that are "possible diabetic" for those aged 60+. Round to three decimal places. 8) Create a histogram of the data for those aged 60+. STAT 243Z Page 7 of 8 Descriptive Statistics Project

3.1.4 Comparative Graphical Summary 1) Create a side-by-side boxplot that includes a boxplot for each age group. 2) Explain why a side-by-side boxplot is a better choice for a comparative graphical summary as opposed to looking at the histograms. 3) Create a bar chart of the age based proportions computed in 3.1.1-3.1.3 for those with "prediabetes." 4) Create a bar chart of the age based proportions computed in 3.1.1-3.1.3 for those with "possible diabetes." 3.2 Age Based Summary 1) Use your results from 3.1 Summaries by Age to discuss (in 3-5 sentences) what the data tells you about how diabetes prevalance changes with age. Be sure to support your claims 2) Can a causal relationship be established using the data? Why or why not? STAT 243Z Page 8 of 8 Descriptive Statistics Project