research methods: probability & significance Flashcards
Null vs alternative hypothesis
Null - no sig diff
Alt - will be a sig diff but no direction
Probability
Likelyhood of something occurring by chance
The higher the probability the greater… so therefore …
likelihood the results were down to chance
Therefore, you want as low probability as possible.
Significance level
The level of probability at which it is agreed that the null hypothesis will be rejected and the experimental hypothesis accepted.
Usual level of sig
P<_ 0.05 (5%)
Probability that our results are due to chance is less than 0.05
What is a type 1 error
Null hypothesis is rejected and alt hypothesis is accepted when it should have been other way round, null hypothesis is true (false positive - found. A correlation but it down exist)
More likely if significance level is too high (10% or 0.1)
Type II error
Null hypothesis is accepted and alt is rejected, but the alt was true -false neg
Too low struct significance evel - eg 0.01 or 1%
Distinguish between a type I and type ll error in psychological research (3 marks)
A Type I error occurs when a researcher wrongly rejects the null hypothesis, thinking there is an effect when there isn’t one (a false positive). A Type Il error happens when the researcher wrongly accepts the nuil hypothesis, missing an effect that actually exists (a false negative). These errors are linked to the level of significance and statistical pawer.
What is the accepted level of significance in psychological research? (1 mark)
P<- 0.05
Explain why a psychologist might prefer a lower level of significance. (2 marks)
A psychologist might prefer a lower level of significance (e g., 0.01) to reduce the likelihood of making a Type I error. This is important in research where false positives could have serious consequences, such as medical or forensic psychology
