Sources of Info, Variance, & Behavioral Science Research
Which of the following IS a scientific source?
A. Personal Experience
B. Journal articles
C. Authority
D. Intuition
B. Journal articles
What is the difference between alternative explanations and clarifying explanations?
Alternative explanations are a threat to the internal validity and original claim (usually due to confounds).
Clarifying explanations add to the original claim and do not threaten internal validity.
In replication, which of the following uses different operational definitions to test the original claim?
A. Replication-plus-extension
B. Direct replication
C. Conceptual replication
C. Conceptual replication
True or false: Multivariate designs in correlational research account for temporal precedence, but not internal validity.
False: Multivariate designs account for internal validity (i.e., can statistically remove third variables), but not temporal precedence.
Concurrent measures designs and Repeated measures designs are the two main types of what experimental research design?
Within-subjects designs
Effect size is essentially the proportion of _______ to _______ variance.
Modern ethical guidelines are shaped by what three unethical studies from the past?
1. Milgram's obedience study (1960s)
2. Stanford Prison study (1971)
3. Tuskegee Syphilllis study (1932- 1972)
When observers in observational research create a self-fulfilling prophecy by recording what they expect to see, it is called _________ ____.
+ 50 bonus points: How can we avoid this from happening?
Observer bias.
To avoid this, one should use multiple observers, practice coding beforehand, and have precise and concrete operational definitions.
Mediator
"Students will not spend more at the Cove when supplied pirate bucks."
What does it mean to say that research results are probabilistic?*
Research conclusions explain a certain proportion of possible cases, but may not explain it all.*
*from Textbook
For most experiments, trade-offs are made between validities because it is not possible to achieve all four at once. What is the most common trade off?*
A. Internal and External validity
B. Construct and Statistical validity
C. External and Statistical validity
A. Internal and External validity
*From textbook
Which of the following is a MUST in evaluating construct validity?
A. Convergent / discriminant validity
B. Content validity
C. Face validity
D. Criterion validity
E. None of the above
E. None of the above
Trick question: Rarely do objective standards exist, so assessing any one or more of these forms of validity contributes to the overall construct validity- all based on researchers' judgement.
The _____ __ ______ is the percentage of variability in one variable that is accounted for by the other variable (i.e., effect size in correlational research).
Coefficient of determination (r2)
How do we calculate the effect size in ANOVA (η2)?
η2 = SSbg / SStotal
Deduction (Theory --> Hypothesis --> Prediction) or Induction (Observation --> Hypothesis --> Prediction)
List and describe the big three principles of the Belmont Report (1979).
1. Respect for persons (Informed consent, avoid coercion, being careful around deception and vulnerable populations)
2. Beneficence (Protect participants from undue harm, right to privacy, include debriefing)
3. Justice (Participants should represent the population who will benefit from the study, just sampling practices)
If a researcher was in the early stages of learning about parental investment, what type of measurement approach would suit them best to gather data from parents about the topic?
Open-ended questions in a Self-Report.
What does the regression coefficient being significantly different from zero tell us about the relationship between the criterion and predictor variables?
That the relationship is still robust after controlling for other variables.
Write the five steps (including formulas) in conducting a paired-samples t-test.
1. Calculate the mean difference (Mdiff) for each participant or pair (usually given).
2. Calculate the standard error of the mean difference. SMdiff = √(s2diff/N)
3. Calculate your obtained t-statistic. tobt = (Mdiff - μdiff) / SMdiff
4. Find critical value(s) df = n1 + n2 - 2
5. Reject or fail to reject the null hypothesis
How can relying on intuition lead us to erroneous conclusions? List the six cognitive short-cuts
1. Good Story heuristic- if it sounds plausible, we believe it
2. Availability heuristic- judge the likelihood of an event by how easily we can think of examples
3. Confirmation bias- attend to evidence confirming our beliefs and ignore other evidence
4. Confirmatory hypothesis testing- structure investigation to favor our preconceptions
5. Overconfidence- stop seeking additional data when we think we know something
6. Bias blind spot- we tend to believe we are generally unbiased
Write the full formula for variance (s2)
S(X - X)² / n - 1 = Variance (s²) *
square root s² = Standard deviation (s)
*pretend S is the Sigma letter for Sum, it wouldn't let me include it
Regarding internal reliability, what exactly does Cronbach's alpha do?
Computes the average correlation among all the items in the test/measure.
Sketch out the four steps involved in testing for mediation.
1. Establish correlation c (relationship of interest)
2. Establish correlation a (relationship between predictor and mediator)
3. Establish correlation b (relationship between mediator and outcome)
4. Test whether c is still significant with the mediator present (c')
Write the five steps (including formulas) in conducting a independent-samples t-test.
1. Calculate the means of the two groups (usually given).
2. Calculate the standard error of the difference between the means. sp2 = (n1 - 1)s12 + (n2 - 1)s22 / n1 + n2 - 2
3. Calculate your obtained t-statistic. t = X̄1 - X̄2 / sp√ (1/n1 + 1/n2)
4. Find critical value(s). df = Ndifferences - 1
5. Reject or fail to reject the null hypothesis