significance testing Flashcards
sampling Error
We try to estimate the population means using sample but there are likely to be errors as the sample may be slightly different
sample mean - population mean = sampling error
Multiple samples
-taking an average across multiple samples lowers the sampling error
small sample sizes
Likely that:
-most ppts score above or below the population mean
-more likely to get a skewed representation/sample mean
sampling error is higher
sample means are varied
large sample sizes
Likely that :
-some ppts score above + below mean
sampling error lower
sample means relatively Stable
Better estimate of the population mean
Type 1 Error
false pos
claim I V influences DV when it doesn’t in reality
reject the null hypothesis when its true
Type 2 Error
false neg
claim I V doesn’t influence the DV when it does in reality
Null hypothesis is not rejected when it should be
Hypothesis Testing
inferential statistics are about the probability that you would find your results if the null hypothesis is true in reality
-If in reality your I v has no effect on DVV
-probability that your results is due to chance
probability (p)values
probability of getting your results by chance = represented by a p value
smaller the p value smaller chance of type error/ results due to chance
Alpha level
= 0.05
Low probability of a chance result means your I V probably did affect your DV
significant = p < 0. 05
Reject null hypothesis
Non-significant = p>0.05
fail to reject the null hypothesis
2 tailed hypothesis
predict theres a difference ut don’t state the direction