Material

🎥 Watch Scientific Studies and Confounding

• Slides

Today's Goal

• Understand statistical process terminology
• Understand the types of conclusions that can be made given the underlying statistical process
• Learn to simulate from a binomial distribution

Quiz

Q - What is an observational study? Any examples?

• Collect data in a way that does not interfere with how the data arise
• Establish associations (!= causation)
• Data often cheaper and easier to collect
• e.g., Smoking behavior and health outcomes

Quiz

Q - What is an experimental study? Any examples?

• Randomly assign subjects to treatments vs. control
• Establish causal connections
• Often more expensive
• Sometimes impossible or unethical to design an experiment
• e.g., Covid-19 vaccine tests in developmental phase

Quiz

In the prep video, the following terms are introduced:
response variables and explanatory variables

Q - What is a response variable?

• "Dependent variable"
• Denoted by $y$
• The focus of a research question

Q - What is an explanatory variable?

• "Independent variable"
• Denoted by $x$
• Variables that might affect or explain changes in the response variable

Quiz

Q - Identify a response variable and an explanatory variable regarding the following examples.

Do larger homes in good locations lead to higher home selling prices?

• $y$: home selling price
• $x$: size of a house, neighborhood

Which procedure prolongs life more for breast cancer patients, chemo or anti-estrogen treatment?

• $y$: survival time
• $x$: type of treatment
• In real life, you would include several more explanatory variables such as age, general health conditions, weight, etc.

Quiz

Q - What is a confounding variable?

• Extraneous variable that affects both the explanatory and the response variable, and that makes it seem like there is a relationship between them.
• If you fail to account for confounding variables, you may not correctly observe the actual relationship between explanatory variables and a response variable.

Quiz

Q - Identify the confounding variable in each of the following statements.

As the amount of ice cream sales increases, the number of shark attacks also increases.

• Confounding variable: summer
• Both increase in the summer, when people go to the beach.

The higher the number of firefighters at a fire is, the greater the amount of damage caused by that fire.

• Confounding variable: size of a fire
• More firefighters are sent to larger and severer fires that cause greater damage.

Quiz

In the prep video, the following terms are introduced:
random sampling and non-random sampling

Q - What is random sampling?

• A sampling method that allows for randomization of sample selection.
• Conclusions from a random sample are generalizable.
• Simple random sampling: every element in a whole population has an equal chance of being chosen for the sample

Quiz

Q - What is non-random sampling?

• In a non-random sample, individuals are selected based on non-random criteria, and not every individual has a chance of being included.
• Conclusions from a non-random sample are not generalizable.
• e.g., I make my close friends complete a survey and analyze the survey data for a final project.
• e.g.. An online survey available to anyone who wishes to provide responses

Let's Practice Together!

Go to AE 12: Foundations of Inference

Bulletin

• Exam 01

  • to be released today at 1pm
  • due Monday, May 30 at 12:59pm

• May 30

  • Memorial day holiday. No class.
  • Have fun after submitting Exam 01!

• Nothing to prepare for May 31!

• Lab 04 due Tuesday, May 31 at 11:59pm.

• Submit your ae12 before class on Tuesday, May 31.