Stat 30100 Media Project

Big Picture

Overview: You need to post 1 unique article/explanation about an experiment OR 1 unique article/explanation about an observational study. You must also make 4 statistically intelligent and valid comments on other people’s posts. These comments should be split between observational studies and experiments (read two of each).

Read over this entire document to get the details of exactly what is required. Each article/explanation is worth up to 25 points, and each statistically intelligent comment that moves the discussion forward is worth 5 points. The whole assignment is worth 45 points.

Purpose: Statistics are everywhere in the media. Some are used well, and some are not. As part of the goal of this course, you will become a savvy media statistics consumer. This project is not meant to be completed in one night. You are expected to keep your eyes open for statistics in the media over the whole semester.

Please note the articles must be news stories—they cannot be course materials designed to teach statistical methods. The stories do not have to be recent. Your posting has to be UNIQUE. It won’t count if another student has posted a link to the same story or if your instructor has mentioned it in class. This means a Washington Post story about a research project will count as exactly the same as an MSNBC story about the same research. You will have to read what other students have posted and written in order to make sure that your posting is unique. We will check postings across sections, too, to make sure you aren’t sharing posts with students in other sections.

Your Discussion Post

You should begin a new thread with your topic, and include:

1.) A link to the website or upload the relevant file.

2.) The requested explanation or calculation (See below.)

3.) Format your post like this: “Observational Study: In this article, Gallup polled 200 people about ….” so that it is clear to everyone which topic you have chosen.

4.) Use “Observational Study” or “Experiment” in the title (no quotation marks necessary). This may help you ensure that you haven’t duplicated anyone else’s post.

Commenting on Other Posts

You must make at least 4 comments on posts from other people. (Note: Only 1 comment per person per post will receive credit, so don’t try 4 comments on just one post. The goal is to get you reading and commenting on lots of different posts.) These comments must be thoughtful, indicate that you have thoroughly read the article and any previous comments on that article, and must add to the level of discussion on the statistical concepts. Balance your comments between experiment articles and the observational study articles. The instructor may post articles also, and a comment on those are acceptable for earning your discussion points.

Just as with everything we do in this course, please keep in mind that it is important to be respectful of each other and to engage in professional dialogue.

Examples of bad comments on posts which will not receive credit:

· “This was an interesting article. It reminded me of junior high days!”

· “I love everything that has to do with monkeys and I find the way they act to be fascinating. It is so crazy to me that they are intelligent species, and act SO similarly to humans.”

· “This was a great experiment.” (This is especially bad if the article was actually an observational study and you didn’t catch it! Not all the students who post articles will label them correctly, and you should be on the lookout for that too).

Examples of good comments on posts which will receive credit:

· “You didn’t label this post, but I’m pretty sure you meant to label it “Experiment” based on your discussion. Anyways, this is definitely not an experiment, it is an observational study. No treatments were applied. It’s true that they gave everyone a hearing test, but that was simply a method of collecting data since you can’t just ask everyone if they have hearing problems and get reliable feedback. They didn’t divide the people tested into groups and give them any sort of different treatment to see if there was an effect on the outcome. In other words, they were just looking for the data for a single variable as it was, as opposed to changing another variable, to see what it’s effect on the measured variable would be.”

· “This graph is interesting, but pretty hard to read. I had to look very closely to see if red was used in more than one place. I like that it is interactive, but it is difficult to put your cursor over the extremely thin red lines. I wonder why it raised back up in the 1940’s? I do find it interesting how drastically the name popularity has declined over the years. It hasn’t become popular again. The graph shows that for a majority, the name has been on a decline.”

You should check back on your old posts from before to see if the instructor or your peers had comments/questions for you to answer.

See the course schedule for due dates. It is strongly recommended that you do not wait until the last minute to try and find links to articles. Since your article needs to be unique (Stories not posted by any other student) and your comments need to be thoughtful, it is to your advantage to work steadily throughout the semester. With the uniqueness requirement, this assignment will be much easier for the people who get an earlier start!

Details

Here are the details on the articles and discussion you should post to a new thread (choose only ONE):

1.) Observational Study:

a. How was the sample collected? (Use appropriate statistics terminology: Simple Random Sample, Stratified Random Sample, Cluster Sample, Voluntary Response)

b. What is the sample size?

c. What was the response rate?

d. What variables were measured?

e. Were the results displayed in a graph? If so, what type of graph was used? If not, what type of graph would be appropriate for displaying this data?

f. What are potential sources of bias?

g. What could be done to improve this study?

2.) Experiment:

a. Who or what is in the sample? What is the sample size?

b. What is/are the treatments?

c. What is the design of the experiment? Is it one we learned in class?

d. Were the three principles of good experimental design (Randomization, Replication, Comparison/Control) used?

e. What are the response variables? Are these variables categorical or quantitative?

f. Were the results shown with a graph? If so, what kind of graph was used? If not, what kind of graph would be appropriate for displaying this data?

g. What are some potential lurking variables? Explain your reasoning.

Grading

Choose ONE TOPIC OR THE OTHER, NOT BOTH.

Portion of the Assignment Points Possible
Unique Article 11
Each question answered (a-g) is worth 2 points 14
Statistically intelligent comment 5 points each, 4 comments 20
Total 45

3