Hypothesis Testing
This represents no difference or effect
Null hypothesis
Shape of a normal distribution
Bell-shaped
Probability result occurred by chance
p-value
Test used for two different groups
Independent samples t-test
Test used for three or more groups
One-way ANOVA
This predicts a difference exists
Alternative (research) hypothesis
Mean, median, and mode relationship in normal distribution
Equal
Standard alpha level
.05
Test used for same group measured twice
Dependent (paired) samples t-test
Test statistic for ANOVA
F-value
This hypothesis specifies direction (e.g., higher/lower)
Directional hypothesis
Percent of data within ±2 SD
About 95%
Decision when p < .05
Reject the null hypothesis
Key assumption for independent t-test
Two independent groups
Two types of ANOVA degrees of freedom
Between-groups and within-groups
This hypothesis does not specify direction
Nondirectional hypothesis
Meaning of a positive z-score
Above the mean
This type of error rejects a true null hypothesis
Type I error
“t” represents this in output
Test statistic
Section of a paper that reports statistical findings
Results
This is written first in hypothesis testing
Null hypothesis
What a z-score indicates
Number of standard deviations from the mean
This type of error fails to reject a false null hypothesis
Type II error
df formula for independent t-test
n₁ + n₂ − 2
Required elements in APA statistical reporting
Test statistic, degrees of freedom, p-value, descriptive statistics