Inferential stats intro part 1
- 1. Learning Objectives: •Understand the nature of probability • Understand the role of probability in statistical testing • Describe and choose the level of significance • Explain the difference between type 1 + type 2 errors. Inferential Statistics Outcomes: ALL – Complete own MEMORABLE notes using mnemonics / mindmaps on each objective. Key words you need definitions for: MOST – Complete activities set • Probability • Level of significance SOME – Complete Tea Test problem by • Type 1 error end of the lesson •Type 2 error
- 2. Inferential Statistics Tests Make inferences about the populations from which the samples are drawn
- 3. Descriptive Statistics vs. Inferential Statistics Allow us to say whether Allows us to draw difference is significant conclusions Through use of graphs This difference Is significant
- 4. Inferential Stats Watch the clip – the tea test. Task: Why are inferential statistical tests needed? (Also see Pg 286)
- 5. Probability How likely is it that something will happen? A number between 0 and 1 0 = something DEFINITELY will NOT happen 1 = something DEFINITELY will happen NUMBER OF PARTICULAR OUTCOMES Probability is PROBABILITY = expressed as NUMBER OF POSSIBLE OUTCOMES “p” Task: What is the probability of a coin landing heads up? How would you express this as a decimal?
- 6. Probability Inferential tests use probability to ascertain the likelihood that a pattern of results could have arisen by chance. If the probability of the results occurring by chance is below a certain level we assume these results to be significant
- 7. Chance We can state how certain we are the results are not Real due to chance difference
- 8. Key questions for Psychologists… •How far does what we have found in our sample reflect the general population? •Could differences shown in our test have occurred by chance? E.g. In a study of 10 yr old boys a positive correlation is found between time spent playing aggressive computer games and observed levels of aggression? Is this the case for all 10 yr old boys? Inferential tests will tell us how probable it is that the correlation could have occurred by chance.
- 9. Watch the clip – P Values Task: Also referring to “Chance” pg 286 What does a p value of p ≤ 0.05 mean? Explain this both as a % but also what it tells us about the results of the study / correlation.
- 10. P-levels/Significance Levels P ≤0.10 C H P ≤0.05 A N P ≤0.01 C E P ≤0.001 We can also write these as 10%, 5%, 1%, 0.1%
- 11. Significant? If our test is significant we can Reject our null hypothesis and accept our alternative/experimental hypothesis If our test is not significant we can Accept our null hypothesis and reject our alternative/experimental hyp “If P is low…null must go.”
- 12. Type 1 and Type 2 Errors Type 1 error Rejecting a null hypothesis when we should not P level too tight Type 2 error Accepting a null hypothesis when we should not P level too loose
- 13. Errors Throwing a coin 10 times there is a 17% probability of getting a head If we set our p level too low it looks like there is phenomena there is not Throwing a coin 100 times there is a 0.005% chance of getting a head If we set it too high we may miss phenomena
- 14. Why do we make errors? Type 1 – if we allow ourselves a p=.05 sig level then we allow yourself a 1 in 20 chance of making an error Type 2 – too stringent a p level means we may miss something
- 15. Watch the Type 1+2 errors video Task In your own terms explain the difference between a type 1 and type 2 error
- 16. Refer to Pg 287 Type 1+2 errors Task: Why might researchers choose to use p≤0.01 in preference to p≤0.05?
- 17. Finished? •Check in with Mr Beech. •Re-vsit any clips you are less certain on. •Re-visit and test yourself on your schizophrenia cue cards.