One sample t tests Flashcards
One-sample hypotheses
-Tests whether the mean of a sample is different to a pre specified comparison value
e.g. attendance in class is more than 80%.
T test equation
t value = mean of observed data - comparison value/ SEM
When does t test value grow
t-value grows as difference between observed and comparison gets bigger - top of fraction gets larger whilst bottom stays the same
When does t test value shrink
t-value shrinks as variance of observed data gets bigger - bottom of fraction gets larger, top remains the same
What did we do before t tests
Estimate of SD is more unreliable than estimate of mean so original answer was to restrict analyses to very large samples where we can be confident in SD
Student t test
camping distribution of t-values is t-distributed - not normally distributed
accounts for additional uncertainty in smaller samples with wider tails