Threats to internal validity!
Study Designs
Sampling
Statistical Validity
MISC
100
When too many people start but do not complete the study.
What is Attrition threat?
100
Having 2 or more conditions where participants are randomly assigned to one condition.
What is an independent (or between-groups) experimental study design?
100
This term means that everyone in the population has an equal chance of being selected.
What is probability sampling?
100
This standard value of probability distinguishes findings that are statistically significant from those that are not-statistically significant.
What is p less than .05?
100
This is the easiest way to show that a questionnaire is reliable and internally consistent. It can be calculated with just one timepoint of data.
What is Cronbach's alpha?
200
A problem when beginning with very high or very low scores in a study.
What is Regression (or regression to the mean) threat?
200
Having a lot of detailed and in-depth knowledge about one person who experienced an unusual event, but without knowing precisely what led to the results.
What is a case study?
200
Named for its ease, this is also the most popular way to recruit participants.
What is Convenience sampling?
200
This is the range around which the true value must lie, when reporting results of an inferential analysis.
What is a Confidence Interval?
200
A question that interrogates this is "does the questionnaire actually measure what it intended to measure?"
What is Construct validity?
300
When you have the WORST timing, and something occurs around the time of your intervention that could impact your study.
What is History threat?
300
In a study where every participant receives all of the levels of the independent variable, this version is when participants experience all levels at the same time.
What is concurrent-measures design (or concurrent within group study design)?
300
Not a weather phenomenon, but a way to gather participants that isn't exactly random.
What is Snowball sampling?
300
This method of decision-making bases its conclusion on whether or not one can reject the opposite of what the researcher expects to happen.
What is Null Hypothesis Testing?
300
To show a change in behavior based on the belief that one is receiving an active treatment or intervention.
What is the Placebo Effect?
400
When a different measure is used at the beginning and at the end of the study.
What is Instrumentation threat?
400
Not related to airplanes, but this study design shows proof-of-concept before designing a more internally valid study.
What is a pilot study?
400
Not a geographical phenomenon, but a way to apply probability sampling with different meaningful groups.
What is stratified random sampling?
400
The general indicator that tells a researcher whether the findings are small or much larger.
What is an Effect Size?
400
Also known as a "study of studies," this is different from a literature review because it analyzes results across studies, thereby providing "new data."
What is a Meta-Analysis?
500
When being exposed to a measure more than one time influences ones ratings.
What is Testing threat (or, practice effects)?
500
Another name for the ABAB design- where you take a baseline, present an intervention, remove the intervention, and present it one more time.
What is Reversal design?
500
This term means that a careful selection was made as to where to recruit participants who are most suited for a particular study.
What is purposive sampling?
500
This is the specific type of indicator SPSS reports with one-way ANOVAs that reflects how large the findings are.
What is Eta Squared?
500
A key issue when determining causality, this criterion isn't about a clock or watch, but refers to what they measure.
What is the time-order or temporal precedence?