This type of validity refers to how generalizable your findings are to your intended population.
What is External Validity?
100
These are measurements of variables made on the subjects of a study, and you subject them to analysis in order to answer specific research questions.
What are data?
100
This is the average score within a distribution.
What is the Mean?
100
When we reject this we are saying that is statistically improbable that we would have found a score as extreme or more extreme due to chance alone.
What is the Null Hypothesis?
100
This test measures the magnitude (strength) and direction (positive or negative) of the linear association between 2 variables.
What is a correlation?
200
If you have this type of validity you can say the IV, versus a confounding variable (e.g., history, maturation), caused changes in the DV.
What is Internal Validity?
200
*Time* is an example of this first variable type, which in theory can have an infinite number of subunits between adjacent whole units. *Number of children* is an example of this second variable type, which can only take on a whole unit value, with no possible values in between.
What are continuous and discrete variables?
200
This is the average (unsquared) deviation of scores about the mean.
What is the Standard Deviation?
200
You commit this type of error when you FAIL TO REJECT the null when in reality it is FALSE; this is, you erroneously conclude that the means of your groups are NOT significantly different when in fact they are or that there is NOT a significant linear association when in fact there is.
What is a Type II Error?
200
Using the principle of linear association, this test uses information about one variable (e.g., height) *to predict* individuals' scores on another variable (e.g., weight). The *standardized coefficient* from this test is equal to Pearson's r when there is only 1 predictor.
What is Multiple Regression?
300
Having small sample sizes, weak effect sizes, or attrition all threaten this type of validity.
What is Statistical Conclusion Validity?
300
These are the *4* types of variable scales, *listed in ascending order*, which vary in terms of magnitude, equal interval, and absolute zero.
What are Nominal, Ordinal, Interval and Ratio Scales?
300
It is necessary to do this in order to compare two distributions that are created from measures with *different scales*.
What is Standardize Variables (or Convert to z-Scores)?
300
You commit this type of error when you REJECT the null when in reality it is TRUE; this is, you erroneously conclude that the means of your groups are significantly different when in fact they aren’t or that there is a significant linear association when in fact there isn't.
What is a Type I Error?
300
This first test evaluates the omnibus difference in values of the DV across 3 or more groups, and contrasts between pairs of groups can be planned a priori or made post hoc. The second test adds a second IV to the first test, such that the effect of each IV is tested across each level of the other IV, revealing main effects of each IV and the interaction between the two IVs.
What are a one-way and factorial ANOVA?
400
The absence of this type of validity calls into question whether or not your IV is what you thought it was (i.e., did the treatment cause changes in the DV or was it the therapist?)
What is Construct Validity?
400
The first of these refers to the *consistency* with which a construct is captured by a measure; the second of these refers to the *accuracy* with which a construct is captured by a measure. Both are necessary in order to have confidence in your data.
What are Reliability and Validity?
400
The first of these distributions results when MOST scores on the distribution are LOW, with outliers on the high-end; The second of these distributions results when MOST scores on the distribution are HIGH, with outliers on the low-end.
What are Positively and Negatively Skewed Distributions?
400
This is the probability that the results of an experiment WILL allow rejection of the null hypothesis if in reality it is FALSE. It increases as sample size increases, alpha increases (becomes less stringent), and effect size increases.
What is Power?
400
This family of tests is used to determine if differences exist between two groups. The *3 tests* within this family (as named in the question) assess the difference in DV values from (1) ONE group against a known population parameter; (2) TWO *related* groups against one another; and (3) TWO *unrelated* groups against one another.
What are a Single Sample t-Test, Dependent Samples t-Test, and Independent Samples t-Test?
500
These are the TWO BEST methodological ways to ensure Internal Validity, ruling out almost all other plausible explanations for changes in the DV.
What are "Inclusion of a Control Group" and "Random Assignment to Condition"?
500
This type of reliability varies from sample to sample, so you must calculate it for all scaled variables in each new study. It is said to be adequate if the coefficient value is greater than 0.70.
What is Internal Consistency Reliability?
500
These values correspond to the APPROXIMATE percentages found between the mean and 1 standard deviation from the mean; between 1 and 2 standard deviations from the mean; and between 2 and 3 standard deviations from the mean on *ONE SIDE* of a normal distribution.
What are 34%, 14% and 2%?
500
This is what happens when you run multiple inferential tests on the same data, and can be avoided through the use of a Bonferoni or Tukey's HSD correction.
What is Alpha (or Type I error) Inflation?
500
This test is used to evaluate if the *proportion* of individuals who are *observed* to fall in different categories is significantly different from what might be *expected*.