Hypothesis Testing Flashcards
What does H0 mean
Null hypothesis
What does H1 represent
Research hypothesis
What is a type 1 error
When the null hypothesis is actually true
What is a type 2 error
When the research hypothesis is actually true
What does p <0.05 mean
When the probability of a score occurring is less than 5%. Then we interpret it that it is unlikely to be due to chance. Reject the null (tentatively)
When is a P value significant?
When it is less than 0.05
When a result is considered significant, what are you also accepting?
That there is a 5% chance of making an incorrect decision
Hypothesis testing definition
Systematic procedure for determining whether the results of a study, which examines a sample, provides support for a particular theory or hypothesis, which applies to a population
How many steps are there in the process of hypothesis testing?
Five
What does step one include?
Formulating research and null hypotheses
What does step two include?
Identifying the comparison distribution
What does step three include?
Determining the cut-off score
What does step four include?
Where does your sample score Sit on the comparison distribution?
What does step five include?
Decision time - should the null hypothesis be rejected?
At first is the null hypothesis always assumed to be true
Yes (we test against it)
When does the cut-off score need to be determined?
Before data is collected
What is the most common cut of score of a distribution?
5%
What happens when the obtain score exceeds the critical value (cutoff score)?
The null hypothesis is rejected and there is a statistically significant result
Explain steps three and four
- need to determine a cut-off
- Need to find a Z score that will give you the cut-off
- Need to calculate the score for the sample you collected
- Then compare your z score to Z score that leaves 5%
- this will determine whether you reject or keep the null
What is a directional hypothesis?
A hypothesis that predicts the direction of the difference between two means
It also specifies where on the curve we will be looking for our samples statistic
Is a one or two tailed test easier to reject the null when at the same significant level
One tail (critical value is less extreme)
When is a two tailed test used
When there is a non-directional null hypothesis (equal sign with a line through it)
When is a one tailed test used
When there is a directional null hypothesis (it is predicted that the dependent variable will most probably increase than decrease and vice versa)
When do u reject the null
When the sample score is greater than the cut off
When do u keep the null hypothesis
When the sample score is less than the cut off (closer to the mean)
What is the probability of making a type 1 error referred to as
Alpha (a)
What is the probability of making a type 2 error referred to as
Beta (B)
What is power
The probability of making a correct decision (reject the null) when the null hypothesis is false