Week 5 Flashcards
(25 cards)
How can we define statistics?
The common method used to draw reliable conclusions from quantitative data
What is the confidence interval (CI)?
A measure of how confident we are in our results - typically we adopt a confidence level of 95% and calculate a range
What is the difference between descriptive statistics and inferential statistics?
Descriptive is means, modes etc but inferential is correlations and forming a hypothesis
What is nominal data?
Categorical data
What is ordinal data?
Ordered or ranked data with undefined distance between steps. For example, is ice cream tasty/neutral/not tasty?
What is interval data?
Ordered data with equal distance between steps
What is standard deviation?
The mean deviation from the mean - a measure of how dispersed the data is
What is the difference between interval data and ratio data?
Ratio data has a defined 0
How are interval and ratio data often grouped?
Parametric or contiguous
What is hypothesis testing?
Defining a hypothesis and a null hypothesis with the goal of determining if we can reject the null hypothesis
What does a significant result mean, given a significance level of 5%?
The chance of getting the results we get, even is the null hypothesis is true, is below 5%
What is a Type 1 error?
Reject a true null hypothesis (false positive)
What is a Type 2 error?
Accept a false null hypothesis (false negative)
Which type of errors are we more likely to accept?
We accept a higher risk of type 2 errors, in order to have a lower risk of type 1 errors
What is a T-test?
Used to test if the mean is different from a reference value - provides a 95% CI
What is an independent sample T-test?
Provides a 95% CI of THE DIFFERENCE
What is the difference between an independent sample t-test and a paired sample t-test?
A paired sample test is usually used to test differences from pre-post tests, for example after a measure is implemented
What is one way to assess distribution form in normal distribution tests?
Looking at curves in graphs
What is a CHI2 test?
Used to test if the frequency of observations is different from some reference frequency (commonly even distribution)
What does a return value of 1 mean in a correlation test?
Perfect correlation
What does a return value of -1 mean in a correlation test?
No correlation
What does positive results bias mean?
Research is prone to favour positive outcomes
What is p-hacking?
Shotgunning statistical tests until something comes back positive
What is thematic analysis?
A systematic procedure for generating codes and themes from qualitative data
