05. Hypothesis Testing Flashcards
Null hypothesis
HYPOTHESIS BEING TESTED; Hypothesis to be nullified or rejected; Hypothesis the researcher HOPES TO REJECT! Rejecting it favors the alternative hypothesis; If null hypothesis is accepted “Fail to reject the Null Hypothesis”
Alternative hypothesis
Research hypothesis; Any hypothesis that does not conform to the one being tested
One directional null
Null hypothesis predicts direction of results
Two directional null
Non-directional; Does not predict direction of results
Alpha levels
Level of acceptable risk set by researcher before any values are calculated; possibility that results happen by chance; Most common ? = 0.05 (5% chance that results happened by chance); smaller = more significant finding
P values
Probability value; risk associated with not being 100% confident; Probability of making a Type 1 error
p
Reject the Null Hypothesis
p > Alpha
Accept the Null Hypothesis
Calculated “t” values
Computer value generated with results
Critical “t” values
Table value
Crit t
Reject the Null Hypothesis
Crit t > Calc t
Fail to reject the Null Hypothesis
Statistical power
Power = 1 - ?; Probability that if there is an effect, it will be detected; Lower sampling error = higher power; Increasing sample size reduces sampling error; Tells about Type II error
Correlation co-efficiencies
Pearson’s Correlation Coefficient = r; Range is -1 to +1. Shows correlation.
Positive correlation
+ Pearson’s Correlation Coefficient showing that one variable goes UP as the other variable goes UP
Negative correlation
- Pearson’s Correlation Coefficient showing that one variable goes UP as the other variable goes DOWN
Independent variable
Variable that is manipulated. Cause of change in another variable; Also called factors.
Dependent variable
Effect of the treatment
Confounding variable
variables that the researcher failed to control, or eliminate, damaging the internal validity of an experiment.
Extraneous variable
Any condition not part of the study. Could have an effect of the DV. Control by randomizatio of subjects
Type 1 Error
False alarm; saying finding is significant when it isn’t; Decreased by lower critical value
Type 2 Error
More serious error; Saying a finding is insignificant when it’s significant; Decreased by increasing sample size