Inferential statistics Flashcards
Why are inferential statistics used
They are used to help us make inferences about populations based on the samples tested.
Allow us to infer IF a pattern in the data was due to chance or not? (likelihood of chance)
What is a research hypotheses
A hypothesis is a predictive statement about what the researcher expects to find.
What is a alternate hypothesis
This hypothesis states the IV will affect the DV in some way in the research
Can be divided into:
One tailed directional – This hypothesis states that the IV will affect the DV and states exactly what the effect will be. It therefore makes a prediction and states the direction of the prediction. It is used when a researcher is confident about what they expect to find.
E.G: Participants will run faster after drinking a coffee than those who didn’t drink a coffee
OR
Two tailed non-directional – This hypothesis states the IV will affect the DV BUT does not state how, just that there is a difference that will be found. It is used when previous research is inconclusive OR a newer area of investigation.
E.G: There will be a difference in the speed of participants that drank a coffee compared to those who didn’t
What is the null hypotheses
Null – This hypothesis predicts that there will be no significant difference between the two conditions; the IV will have no effect on the DV.
What is done if the null hypothesis is correct
If the null hypothesis is correct as no significant difference is found. The null hypothesis is accepted and the alternative hypothesis is rejected.
What is done if the null hypothesis is incorrect
If the null-hypothesis is incorrect and a significant difference is found. The null hypothesis is rejected and the null hypothesis is accepted.
What is the general probability used in psychological research
P < 0.05
Probability less than 5% that results are due to luck
When is P < 0.01 used?
1% luck probability is often used when replicating research or for medical research such as testing a new drug
What is a ‘type 1 error’ in inferential statistics
A type 1 error is when when they thought the results were significant, when they actually weren’t, meaning that the Null hypothesis is rejected when it should have been accepted.
(alternative hypothesis wrongly accepted)
Likely to happen when using 10% luck chance as it is too lenient.
What is a ‘type 2 error’ in inferential statistics
A type 2 error is when when they thought the results were not significant, when they actually were significant, meaning that the Alternative hypothesis is rejected when it should have been accepted.
(null hypothesis wrongly accepted)
How do we find out if the results are significant?
The inferential statistic test calculates an OBSERVED VALUE.
This value is then compared to a Critical Values table to find out if the results are significant or not.
What four things do we need to know when testing if an OV is significant?
Degrees of freedom - aka number of participants (N)
Whether a one or two tailed hypothesis is used
Significance level (usually 0.05)
Whether the OV needs to be greater or less than the CV.
Describe when Spearman’s Row would be used
Test of association
The observed value has to be equal or greater than the critical value to be significant.
Uses numerical data only.
Describe when Chi-Squared would be used
Test of Difference
Used for non-numerical data
The observed value has to be equal or GREATER than the critical value to be significant
Independent groups design
Describe when Man-Whitney U would be used
Test of Difference
uses data involving numbers (more than nominal)
The observed value has to be equal or LESS than the critical value to be significant.
Used in independent groups design
