Research: Module 6 Flashcards
(58 cards)
parametric measures
nonparametric measures
T-tests
a statistical method used to determine if there is a significant difference between means of 2 groups
3 basic types of t-tests
comparing a sample mean to a population mean (we are not doing this)
comparing the means of 2 independent samples
comparing the means of paired dependent samples (same subject repeated testing)
independent t-test
used when comparing 2 unrelated groups
look for differences between 2 separate, unrelated groups
dependent t-test/paired t-test
used when comparing related groups, often measurements from the same subjects at different times
look for changes within the same subjects or matched pairs
what type of t-test would you run for the following scenario: comparing test scores of students in 2 different classes
independent t-test
what type of t-test would you run for the following scenario: measuring the same students’ performance on test before and after a training program
dependent t-test
what type of t-test would you run for the following scenario: comparing the blood pressure of 2 different groups
independent t-test
what type of t-test would you run for the following scenario: comparing a person’s blood pressure before and after taking medication
dependent t-test
one tail tests
looks for a difference in a specific direction (either greater than or less than)
use if you have a clear explanation about the direction of the effect
two tail test
looks for any difference (either greater than or less than)
use when under about the direction or effect or if you are interested in ANY difference
“there is going to be a difference PERIOD”
critical region
In one-tailed tests, the critical region lies entirely on one side of the probability distribution — either the left tail or the right tail — depending on your alternative hypothesis.
The critical region is where the test statistic must fall for you to reject the null hypothesis (H₀) in favor of the alternative hypothesis (H₁)
pros and cons to a one-tail test
pros: can be more powerful (require smaller sample size) to detect an effect in the predicted direction
cons: may miss significant effects if the actual effect is in the opposite direction
pros and cons to a two-tail test
pros: more versatile and can detect effects in either direction, in any difference
cons: requires a larger sample size to achieve the same level of significance compared to a one-tail test
would you use a one tail or two tail test with this hypothesis: Recovery time is less than the current average recovery time
one tail test
-specifies a direction of the effect (greater than/less than)
would you use a one tail or two tail test with this hypothesis: the new method’s scores are different from the old method’s score
two tail test
-simply states there is a difference, not if it will be greater than/less than
type 1 error
rejecting the null hypothesis when it is actually true (false positive)
concluding there is a difference when there is NOT a difference
type 2 error
failing to reject the null hypothesis when it is actually false (false negative)
concluding there is not a difference between groups when there IS a difference
power
the probability that a hypothesis will correctly reject a false null hypothesis
(probability of finding a real effect/difference if one truly exists)
probability of avoiding a type 2 error
larger sample sizes → ______ power
higher power
larger effect size → ________ power
higher power
higher alpha → _______ power
increases power AND risk of type 1 error
lower variability → _________ power
higher
lower variability = less spread out data
