Sampling
What is the difference between a sample and the population?
The population includes all people your study findings apply to. The sample includes only the people in your actual study.
What is one major difference between descriptive and correlational research?
Descriptive research does not seek to test hypotheses.
Explain the primary difference between the Independent variable and Dependent variable in an experiment.
The IV is what the research manipulates, the DV is the outcome of interest.
How do we control for order effects (e.g., carryover effects) in a within-subject experiment?
Counterbalancing
Explain the difference between a Type I and Type II error in reasoning.
Type I Error – False Positive
Type II Error – False Negative
Some researchers are polling students on a college campus about their preferences in the upcoming election. The researchers are careful to only recruit people in a way that would lead them to have perfectly balanced proportions of each demographic in relation to the actual population characteristics.
Which sampling approaches does this describe?
Representative sampling
Correlational research can describe relationships but it cannot ________ them.
Explain
What is the purpose of a "pilot study"?
For a researcher to test/validate a new experimental manipulation.
Why do within-subjects studies usually have higher internal validity?
Within-subjects experiments are always comparing the outcome of interest against a *true* baseline (i.e., the same participant) instead of another group.
What is a major logistical disadvantage to within-subjects designs
Participant attrition
Why are we confident that a large enough sample can accurately reflect the population?
The Central Limit theorem
Name the three distinct research strategies we discussed that all center around comparing groups.
- Experimental
- Quasi-experimental
- Non-experimental
What are the two types of study control we discussed and when do they occur?
Statistical control (post-hoc)
Experimental control (a priori)
When a participant is asked to complete the same task (or a very similar task) multiple times as in a within-subjects experiment, it could lead to _____________.
Practice effects
Why do within-subjects designs have more statistical power than between-subjects designs?
Because within-subjects experiments have more observations per participant. Between-subjects experiments only have one per person.
Which type of probability sampling involves the use of a rule for selecting cases from the sample frame?
Systematic Sampling (e.g., choose every 4th person)
Exploratory research has no formal hypothesis, but may have a general research question. Confirmatory research must be done to replicate the initial exploratory findings.
A researcher wants to know if tutoring helps students’ writing skills in a High School writing course. They have 100 participants, half the participants receive tutoring for 6 hours a week, and the other half receive no additional help. After 3 weeks their writing scores are compared for differences.
What type of experimental design is being described in this study?
Between-subjects
Which type of threat to internal validity is specific to within-subjects designs?
Time-related threats
Explain the Experimenter's Dilemma
The Experimenter’s Dilemma refers to the fact that external validity and internal validity are always in conflict with one another.
Increasing the external validity of a lab study always deceases the internal validity and vice-versa.Which sampling approach involves splitting people up into groups based on some demographic characteristic (e.g., age)?
Stratified random sampling
What do we mean by construct validity in terms of experimental manipulations?
Does the manipulation actually change what we think it does? For example, does an anger manipulation actually make people angry?
Why are experiments that only use an active control problematic?
You cannot be sure that any observed change is due to one condition over the other (e.g., happiness could decrease aggression instead of anger increasing it).
Name and describe two of the time-related threats to validity that can arise in within-subjects designs.
History effects
Maturation effects
Instrumentation effects
Order effects
What are the three general categories that threats to external validity can come from?
Participants
Study design
Measurement
What is Berkson's paradox?
Berkson's paradox occurs when data become censored due to a flawed sampling procedure (e.g., only including highly aggressive people in a study of aggression). This leads can lead to erroneous conclusions.
What type of distribution exists when all possible values have the same chance of being part of your dataset?
Uniform
Researchers interested in studying the effects of different emotional states on aggression know that their laboratory space tends to get hot during the summer. This is an issue because there are known effects of heat on anger and aggression.
To control for this, the researchers measure the temperature of the room during every study session to control for its influence in their analyses. Which of the following best describes this approach to control?
Statistical control
Define statistical power
Statistical power refers to the likelihood (in %) of a hypothesis test detecting a true effect when there is one.
Data that appear to be multimodal when examined in a histogram plot in theory reflect….
That two distinct populations were sampled from instead of one