Validity
The extent to which a study demonstrates a cause-and-effect relationship.
What is internal validity?
This sampling method gives every member of the population an equal chance of being selected.
What is random sampling?
This hypothesis states that there is no effect or no difference.
What is the null hypothesis?
This statistical test compares the means of two groups.
What is an independent sample t-test?
A researcher studies gender differences in math ability by comparing men and women.
What is a quasi-experiment?
The extent to which results generalize to other people or settings.
What is external validity?
This sampling method divides the population into subgroups and randomly samples from each subgroup.
What is stratified sampling?
The probability threshold used to decide whether to reject the null hypothesis.
What is alpha level?
This statistic describes the magnitude or practical importance of a finding.
What is effect size?
Within-subjects designs improve statistical power primarily because they reduce this source of variability.
What is individual differences (error variance)?
The extent to which a measure actually captures the theoretical construct.
What is construct validity?
A researcher recruits participants from students walking through the student union.
What is convenience sampling?
A researcher predicts that therapy will improve mood compared to no therapy. This uses this type of test.
What is a one-tailed test?
A statistical test is considered statistically significant when this probability is below alpha.
What is the p-value?
Between-subjects designs avoid this threat that occurs when participants experience multiple conditions.
What are order effects? (or practice/fatigue effects)
Experimental manipulation primarily improves this type of validity.
What is internal validity?
This sampling method intentionally selects participants with unusually high or low scores on a variable.
What is extreme group sampling?
Using a one-tailed test instead of a two-tailed test increases this statistical property if the prediction is correct.
What is power?
The critical value of a t-test varies as a function of this number.
What is degrees of freedom? (Sample size?)
An uncontrolled variable that systematically influences the dependent variable.
What is a confound?
Controlled laboratory experiments often sacrifice this type of validity.
What is external validity?
This type of sampling improves external validity but may be difficult or expensive to implement.
What is random sampling?
A study has a real treatment effect, but the researcher fails to find significant results.
What is a Type II error?
Standard error reflects variability due to this source.
What is sampling error (or random sampling variability)?
Random assignment helps control this category of variables.
What are confounding variables? (or individual differences)