Looks like no one added any tags here yet for you.
Quasi-experiments
a type of research design where a comparison is made, as in an experiment, but no random assignment of participants to groups occurs.
Why are quasi-experiments used?
When a researcher wants to compare the behavior of groups of individuals cut are unable to manipulated the characteristics on which the groups differ as the manipulation would be unethical.
What makes a quasi-experiment different from an experiment?
The random assignment of participants to groups is missing in a quasi-experiment.
Random assignment
Assignment of participants at random to levels of the independent variable in an experiment to control for individual differences as an extraneous variable.
Pre-test posttest design
a type of research design (often a quasi-experiment) where behavior is measure both before and after a treatment or condition is implemented.
Solomon four group design
A pretest-posttest design with two sets of nonequivalent groups, one set that takes the pretest and post test and one that only takes the post test.
Time series design
A research design where patterns of scores over time are compared from before a treatment is implemented and after a treatment is implemented.
What makes a pretest-posttest design a quasi-experiment?
The inclusion of a randomly assigned control group is optional as they are examining attitudes before and after a time frame where other confounding variables could cause error with final attitudes.
How do we control for bias in a pretest-posttest design?
Adding a control group.
What is a nonequivalent pretest-posttest design?
When we have a control group but participants are not randomly assigned to the control group. (ex. When looking at effects of the new teaching standards model we chose schools that already fit our criteria to be our control group rather than randomly assigning them to a control group.)
Which design allows the researcher to measure the effect of taking the pretest on the scores of the posttest (testing effects)?
The Solomon four group design.
Interrupted time series design
A time series design with a treatment is an independent event such as a historic event, a study testing the effect of a treatment not implemented by the researcher.
Example of interrupted time series design.
The reduction of the work shifts in a factor from 10 hours to 8 hours and the study measured the effects of this change on employees and the business itself.
Non-interrupted time series design
a time series design where the treatment is implemented by the researcher
Longitudinal design
A developmental design where a single sample of participants is followed over time and tested at different ages
Cross sectional design
A developmental design where multiple samples of participants of different ages are tested once
Cohort sequential design
A developmental design where multiple samples of participants of different ages are followed over time and tested at different ages.
Name the three main types of developmental designs.
Longitudinal, cross sectional, and Cohort sequential designs.
What is a drawback of a cross-sectional design?
Generation effects (cohort effects)
Generation effects (Cohort effects)
A confound that can occur in cross-sectional designs due to different experiences that different generations have had.
What is the primary disadvantage of a longitudinal design?
They take a very long time to carry out and they often have error with attrition/mortality and history effects.
What are the sources of bias that can affect a longitudinal design?
Testing effects and attrition/mortality.
When is a small n design used?
Who is famous for conducting a small n design on forgetting?
Herman Ebbinghaus
What study did Ebbinghaus do on forgetting and what did he learn?
He taught himself nonsense syllable to perfection and then let time pass and kept testing himself multiple times to see how forgetting goes down. He found that most of our forgetting happens in the first 48 hours.
Baseline measurement
Measurement of behavior without a treatment used as a comparison for during and after the treatment has begun.
Baseline design
A small n design that involves baseline measurements of behavior as compared with measures of behavior during the implementation of a treatment.
What is the main type of baseline design?
A-B-A design/ reversal design
A-B-A design/ reversal design
A small n, baseline design where the baseline behavior is measured, followed by implementation of a treatment, followed by another baseline measure after the treatment has ended.
Example of an A-B-A design/ reversal design.
School psychologist was called in to work with a child that was disrupting class as they are trying to help the teacher come up with an effective treatment, in order to test for its effectiveness (possibly positive reinforcement) we need to do an ABA design.
Discrete trial designs
A small n design that involves a large number of trials completed by one or a few individuals and conducted to describe basic behaviors.
Carryover effects
Effects that occur when participants’ experience in one condition affects their behavior in another condition of a study.
How are small n designs presented?
They are presented with no identifying information to protect their confidentiality.
Null hypothesis
The hypothesis that an effect or relationship does not exist, or exists in the opposite direction of the alternative hypothesis, in the population
Alternative hypothesis
That there is a relationship (or a specific type of relationship of a one-tailed test is used)
Power
Likelihood that a statistical test will detect an effect or relationship in a sample when one exists.
Sampling error
The difference between the observations in a population and in the sample represents that population in a study.
Alpha level
The probability level used by researchers to indicate the cut-off probability level (highest value) that allows them to reject the null hypothesis.
What is a variable with more than 2 levels known as?
Multivalent independent variable
What can cause the sample mean difference?
Can be caused by the effect, by sampling error, or both.
Sources of bias for a pretest-posttest design.
History effect, regression towards the mean, testing effects.
History effect
Events that occur during the course of a study to all participants or to individual participants that can result in bias.
Testing effects
Effects that occur when participants are tested more than once in a study, with early testing affecting later testing.
Regression toward the mean
Can occur when participants score higher or lower than their personal average, with scores more likely near their personal average upon additional testing.
When should you use an ANOVA?
When more than two samples or MORE THAN TWO sets of scores from the same individual are compared; an effect.
When should a one-sample t test be used?
When the population mean without the treatment is known and is compared with a single sample; an effect
When should an independent samples t-test be used?
When two samples of different individuals are compared; an effect.
When should a linear regression be used?
When you want to predict an individual’s score on one variable from the score on a second, related variable; a relationship.
When should a Pearson r be used?
When the relationship between two sets of scores is being tested; a relationship.
When should a chi square be used?
When the relationship between two sets of categorical responses is being tested; a relationship.
When should a repeated measure/ paired samples t test be used?
When two related samples or two sets of scores from the same individual are compared; an effect.
Multivalent independent variable
An independent variable that includes three or more levels.
When is a small n design used?
The goal of this design is to understand an individual behavior rather either to better explain it for other people or change a negative behavior for a small group of people/ an individual.