Goals

This homework serves two goals

Assignment

Type: Individual

Tasks:

For this homework you will be using RStudio on a local machine, either your own personal computer or a school computer. Note that RStudio should be available on most computers across campus.

All this time, we have been working with R Markdown files and knitting them to pdf documents on the Cloud. Whenever you click File > New File > R Markdown it gives you three different file types as a choice, HTML, PDF, and Word. I normally use HTML and teach with HTML. However, the reason we have been using PDF all this time was because Gradescope only accepts pdf file uploads!

For this homework we will be using HTML and one the homework is completed we will make it a pdf. The reason is because pdf files requires extra installation on your computer. However, if you are interested in making these installations such as installing TinyTex you are free to work with pdf. You can find instructions here on how to download TinyTex.

Make sure to watch the video on how to create R projects.

Homework Questions

  1. OpenIntro 6.22 Sleep deprivation, CA vs. OR Part I
  2. OpenIntro 6.24 Sleep deprivation, CA vs. OR Part II
  3. Given the following information, calculate p-value and decide whether to the null hypothesis would be rejected at \(\alpha = 0.05\). Assume conditions have been met.

  1. n = 135, T = 4.32, \(H_A: \mu \neq 0.5\)

  2. n = 9, T = -1.67, \(H_A: \mu < 0.3\)

  3. n = 24, T = -5.64, \(H_A: \mu > 0.2\)

  1. For this question we will be working with Breast Cancer Coimbra Data Set. Click on the Data Folder to download the data in csv format.

Make sure to familiarize yourself with the data by reading about the variables on the website. Note that the data comes this study. It is reported in the study that the blood draws were taken when participants were fasting. According to Mayo Clinic a fasting blood sugar level less than 100 is mg/dL is normal.

  1. In the data provided to you, in other words in the sample, do breast cancer patients have normal glucose level on average? Calculate sample statistic. Make sure to share your code and comment on your finding.

  2. Estimate 95% confidence interval for average glucose level of breast cancer patients.

  3. Set up and test hypotheses to determine whether breast cancer patients have higher than normal glucose levels or not.

Scoring: 25 points

Submission

In your project folder, you should see at least 4 things

  1. an R project
  2. the dataset as a csv file
  3. an R Markdown document
  4. an HTML file that you get when you knit the R Markdown.

Gradescope expects each answer to be on a separate page. So you will have to add page breaks to your document. Between each of your responses, add the following code <p style="page-break-before: always"> to add page breaks. Add this between each response. That means this code should come after your response to Question 1 and before your response to Question 2a. See this Rmd example file for further assistance. Note that HTML will not show page breaks. However, it will show the page breaks when you can save your file as a pdf. To do that open the HTML file in Google Chrome (or any web browser) and try to print the document (not really) by clicking Ctrl + P (Windows) or Cmd + P (Mac). For Destination, instead of choosing a printer, choose Save as PDF save the file.

If you have been using pdf files all along, then you will have to use \pagebreak instead of the aforementioned code to add page breaks. See this Rmd example

Upload the pdf file on Gradescope.

Come to our office hours if you need help with questions from the book, R, or how to submit your homework.




Title
Copyright © Dr. Mine Dogucu. All Rights Reserved.