Hypothesis Testing and Categorical Inference I Flashcards
Null Hypothesis (Ho) & Alternative Hypothesis (Ha)
Ho: there’s no special relationship (default hypothesis)
Ha: there’s a relationship, an effect!
How can we reject or fail to reject the null? Ex. Can we say with confidence that people do worse than random chance (Ha), or is this within a range one would expect from sampling variation(Ho)?
We can check this by constructing a confidence interval!
In hypothesis testing, we either reject the null or fail to reject the null, we never
accept the null
Type 1 Error
rejecting the No when the No = T
Type 2 Error
Failing to reject No when No = F
P-value
the probability of observing the sample data (or something more extreme) under the assumption that the null hypothesis is true.
If p-value < a you
reject the null hypothesis
if pvalue > or = to a you
fail to reject the null hypothesis