Data
Stats/Methods
Stats/Methods 2
Design
Validity
100
The scale of measurement that is categorical, with no properties of the real number system.
What is nominal data?
100
Used to summarize data obtained from a sample, e.g. pie chart, frequency chart, etc.
What are descriptive statistics?
100
The degree to which an instrument consistently measures the same variable.
What is reliability?
100
When the participant and the data collector do not know which treatment condition the participant belongs to.
What is a double-blind study?
100
Changes in the dependent variable due to normal growth and development.
What is maturation?
200
The type of data that can only take on values that are separate (e.g. nominal and ordinal).
What is discrete data?
200
Enables us to make inferences about sample data that we generalize to other samples within the population.
What are inferential statistics?
200
The acceptable alpha level most often used in the social sciences.
What is .05?
200
Involves the sampling strategy used, e.g. sample of convenience, cluster, random stratification, as well as the criteria for inclusion in the study.
What is participant selection?
200
The cross contamination of an intervention to other research groups/interventions.
What is diffusion of treatment?
300
Data that only has two categories.
What is dichotomous data?
300
Stating the null was false when it was indeed true.
What is a Type I error?
300
Selected in order to gather data and use the data to make inferences about the population.
What is a sample?
300
The design in which participants serve as their own controls.
What is a "within" design?
300
To clearly define a variable by making it as measurable as possible.
What is an operational definition?
400
Occurs when the measure does not allow for distinction in top scores.
What is the ceiling effect?
400
This determines the design of the study; its measures, data types or scaling, and the type of statistical analysis.
What is the research question or hypothesis?
400
What one does when the obtained probability or p-value is greater than alpha.
What is "retain the null hypothesis"?
400
The design in which different groups of subjects are compared via inferential statistics.
What is a "between design"?
400
Two techniques used to ensure inter-rater reliability?
What is training and percent agreement?
500
When extreme scores tend to be more similar to the normal distribution upon retesting.
What is regression to the mean?
500
When all of the error (Alpha = .05) is in one tail of the distribution.
What is a one-tailed test or a directional hypothesis?
500
The type of problem that occurs when one runs too many statistical analyses on the same data.
What is experiment-wise error?
500
The design in which analyses are undertaken both within and between participants.
What is a "mixed" design?
500
When the experimenter limits confounding variables within the environment via research design and selection of measures.
What is control through procedure?
M
e
n
u