Chapter 5 - Sampling & Probability

Question 1

What is Probability?

Answer 1

the likelihood that a particular outcome (out of ALL possible outcomes) will occur

Required individual trials to be independent, meaning the outcome of each trial can not depend on the outcome of the trial before or after it.

Question 2

Difference between Analytical Probability & Relative Probability?

Answer 2

1. Outcomes are known. Do calculations based on information I know, like with a known dice that has 6 sides.
2. Outcomes are based on what was actually performed, like the actual 10 times I rolled the dice. Proportion should be used here instead of Probability. Proportions do not reflect the underlying probability that will show up when a 1000 trials or 2000 trials will eventually show – it will eventually show the Analytical Probability of 1/6.

Proportion will only equal probabilities OVER THE LONG-TERM (‘Law of Large Numbers’)

Question 3

3rd view: What is Expected Relative Frequency Probability?

Answer 3

Uses past performances. NOT like Analytical Probability where it’s using info about the trials to analyse probabilities – it’s using info about trials that have already been done and we have outcomes.

Ex: there is no known info on the internet that tells us how many red skittles are per bag. We have to do our own trials. So grab a bag and count out the colours.

SO, THIS IS WHERE STATISTICS COMES IN - it uses expected relative frequency probabilities all the time, with MANY trials to create predicted probabilities. To make conclusions about a population based on a sample.

Question 4

Illusory Correlation?

Confirmation Bias?

Answer 4

See below

Question 5

Relative Frequency Distribution?

Answer 5

The likelihood of an event occurring based on the actual outcome of many, many trials.

• we usually don’t know the population distribution
• we draw a sample and make out best guess (inference) based on that
• as sample sizes INCREASE, gets closer to true distribution (law of large numbers – gets closer to actual, underlying probability)

Relative indicates that this number is relative to the overall number of trials, and expected indicates that it’s what we would anticipate, which might be different from what actually happens.

Question 6

Setting up an experiment:

Answer 6

Question 7

EXPERIMENTAL & NULL Hypothesis

Answer 7

Question 8

DECISIONS using the HYPOTHESIS:

Answer 8

Question 9

Type 1 and Type II Errors:

Answer 9

Q: A study reports that eating dark chocolate has larger excitatory effects on the body than kissing a romantic partner. If this finding can’t be replicated, it implies the researchers made a ____ error.

TYPE 1.