Sampling
What is the difference between a population and a sample?
A population is the broad demographic we aim to understand, the sample is who from that group we can specifically measure. (Population= College students in general; Sample= OU college students who sign up for my study)
In a true experiment, we aim to explain __________________________ relationships.
Cause-and-effect
Which research design are you using?: If you are just explaining the frequency of something, with no attempt to understand any relationship.
Descriptive
What is external validity?
The extent to which my findings generalize to others in that population
Doctor A wants to know how many students garden. He asks students if they garden or not.
What research design is being used here? Why?
Descriptive, no attempt to understand a relationship
What is the main difference between probability sampling and non-probability sampling?
In probability sampling, we know all the members of group; in non-probability sampling we do not know all the members of the group
Explain the difference between experimental condition (group) and control condition (group). Explain the purpose of those conditions
Experimental group experiences the manipulation, control group does not. The control group serves as a comparison to the manipulated group to show the effect.
Which research design are you using?: if you are attempting to prove a cause and effect relationship, but make no effort to control for extraneous variables?
Non-experimental
What is internal validity?
Doctor A wants to know if students who garden are more empathetic. He asks students how much they garden and gives them an empathy measure. He is interested in the relationship between empathy and gardening.
What design? Why?
Correlational, wants to understand a relationship but not making a causal claim.
Within probability sampling, there is random sampling. What is the difference between simple random sampling with replacement and simple random sampling without replacement?
In replacement one member is randomly picked out of a hat, and their name is put back in the hat.
In without replacement one member is randomly picked out of a hat, and their name is not put back in the hat.
What is the third variable problem? How can you address it?
The third variable problem is when an unmeasured variable is causing changes in your IV and DV, making it seem that there is a relationship between them.
Utilize control and manipulation
Which research design are you using?: if you are attempting to prove a cause and effect relationship, you rule out all extraneous variables, and you utilize manipulation, control, comparison, and measurement?
True experimental design
Are internal and external validity at odds with each other? Why?
Yes, the more you control internally the less generalizable the findings. The more generalizable your results the higher threat to internal validity.
Doctor A wants to know if students who garden are more empathetic than those who dont garden. He asks students how much they garden and gives them an empathy measure. He groups students based on gardener or not, and compares the mean empathy of the groups. He makes no attempt to control for any extraneous variables.
What design? Why?
Non-experimental, a comparison is being proposed as well as a casual claim. However, he is not controlling for any EV.
Within probability sampling explain the difference between cluster sampling and systematic sampling.
In systematic sampling we take every Nth person on the list (every 3rd person on the class roster)
In cluster sampling, we recruit groups (tables of people at OU: Table 1 would be a cluster, Table 2 would be a cluster, etc...)
What is the directionality problem?
We do not know if X is changing Y, or if Y is really changing X. We attempt to prove a direction in true experimental designs
Which research design are you using?: if you are attempting to prove a cause and effect relationship, but unable to control for all extraneous variables, but still rule out as many as possible?
Quasi-experiment
What are threats to external validity?
Generalizing across subjects and participants (Individual differences), Features of the Study (affluence of the area studied), Features of the measure (english to german)
Doctor A wants to know if students who garden are more empathetic than those who dont garden. He asks students how much they garden and gives them an empathy measure. He groups students based on gardener or not, and compares the mean empathy of the groups. He controls for all extraneous variables statistically or through inclusionary criteria.
What design? Why?
True experiment. a comparison is being proposed as well as a casual claim. He is controlling for all EV.
What is the difference between simple convenience sampling and quota convenience sampling?
In convenience sampling, we just take whoever signs up. (SONA)
In quota we attempt to balance the groups. (50 men and 50 women from SONA)
Name and explain the 4 essential elements for an experimental design.
Manipulation-Deliberate changes in the independent variable to "cause" a change in the dependent variable.
Measurement- Have clear and valid measurements for what you want to compare.
Control- ruling out possible extraneous variables that could explain the results
Comparison- drawing a distinction, not just describing the observation. There is a difference between those who take the memory pill and those who dont.
Which research design are you using?: if you are attempting to establish a relationship between two variables?
Correlational
What are threats to internal validity? What are the three types mentioned?
Extraneous/Confounding variables
Environment (cold in morning, warm in the evening)
Assignment Bias (putting all the sports players in one condition for a running task)
Time-related variables (Fatigue, finals rush for SONA)
Doctor A wants to know if female students who garden are more empathetic than those who dont garden. He asks students how much they garden and gives them an empathy measure. He groups students based on gardener or not, and compares the mean empathy of the groups. He controls for almost all extraneous variables statistically or through inclusionary criteria, but can not control for gender.
What design? Why?
Quasi-Experiment, a comparison is being proposed as well as a casual claim. He is controlling for all EV besides gender because you can not manipulate gender in an experiment. Cant rule out all EV, but makes an attempt to