NHST Basics
what is the hypothesis that states that there is NO effect/difference between groups?
the null hypothesis
a statistic that tells you how big the effect was in our sample by comparing the sample mean to the hypothesized population mean
raw mean difference
What distinguishes a z-test from a one sample t-test?
you know the population standard deviation in a z-test
you know the sample standard deviation in a one sample t-test
the type of test that specifies a difference but not the direction of an effect
two-tailed test
if p= .07 and your significance level is .05, how would you interpret this result?
what is the probability of obtaining results as extreme as the observed data if the null hypothesis is true
the p-value
a statistic that tells you how big an observed effect is, relative to the spread of observations in our sample
Cohen's d (effect size)
a probability distribution similar to the normal distribution but with fatter tails for small sample sizes
student's t-distribution
the type of test that specifies the direction of the effect
one-tailed test
the real-world importance, magnitude, and meaningfulness of research findings, beyond mere statistical probability
the statistical decision made when the p-value is smaller than your alpha value
reject the null hypothesis
a statistic that tells you how large the observed effect is relative to the standard error
t-statistic
the formula used to calculate degrees of freedom for a one sample t-test
df = n - 1
the greek symbols used for population mean AND sample mean
population mean: mu
sample mean: x bar
the p-value represents the probability of observing results this extreme if this hypothesis were true
the null hypothesis
what is the predetermined probability of making a Type 1 error?
your alpha value
p-value
the statistic that tells you how much sample averages tend to vary just by chance
standard error
an error that occurs when researchers fail to detect a real effect and incorrectly conclude there is no difference or relationship
type 2 error
is a small effect size always unimportant?
No! A result with a small effect size may still have practical significance, especially in medicine
the error that occurs when researchers reject a true null hypothesis
type 1 error
the statistic that gives us a range of plausible values for the population mean based on the sample
confidence interval
what are the other two types of t-tests, and when would you use them?
independent samples: used to compare the means of two separate groups
paired samples: used to compare two measurements taken from the same participants
if the p-value from a statistical test is .03 and the significance levels is .05, what is the correct interpretation of the p-value?
the result is statistically significant, and you would reject the null
a result with a very small p-value but a tiny effect size may still lack this
practical significance