Week 2: Hypothesis Testing and Chi-Squared Tests Flashcards
What is a hypothesis test?
A test to determine whether there is enough statistical evidence in favour of a certain belief about a population parameter.
Describe the two types of hypotheses.
1) A null hypothesis: procedure begins by assuming null hypothesis is true. That is, there is no change in the population parameters.
2) An alternative hypothesis: goal is to determine whether there is enough evidence to infer that the alternative hypothesis is true. This is the research question.
Describe the two types of hypothesis errors.
Type 1: not enough evidence to reject null hypothesis by we do reject it.
Type 2: enough evidence to reject null hypothesis but we don’t.
What do the following symbols denote?
1) H0
2) Ha
3) α
4) dR
5) ts
6) tsv
7) con
1) The null hypothesis.
2) The alternative hypothesis.
3) The probability of a type 1 error occuring.
4) The decision rule i.e. when we can reject the null hypothesis.
5) The test statistic.
6) The test statistic value.
7) Comparing tsv to decision rule.
Describe the 6 step process for hypothesis testing.
1) Set up alternative hypothesis and null hypothesis.
2) Determine test statistic & its sampling distribution.
3) Specify significance level.
4) Define decision rule.
5) Calculate value of statistic.
6) Make a decision & answer the question.
Give a one tail and two tail test for the H0: the average mark of an exam is 71%.
One-tail: exam is less than 71%.
Two-tail: exam is different to 71% (could be less than or greater than.
When would you use a test of independence?
To find out if two categorical variables are independent of each other.
When would you use a goodness of fit test?
To find if data fits the assumed distribution or an alternative.
How do you calculate a test statistic chi squared value?
The sum of all (O-e)^2/e where O is the observed value and e is the expected value.
