Flashcards in PSY395 Exam 1 Deck (53):
Events have meaningful, systematic causes.
The method of making observations. (making observations is the best method).
If we have two competing theories, we should use the simpler or more frugal of the two.
You must be able to realistically test the theory (Validation, Falsification, Qualification)
Positive Test Bias (Confirmation Bias)
The tendency to seek out information that should confirm our theory.
And we do not seek out/ignore information that might falsify our theory.
Deductively Valid Arguments
Arguments where the truth of the premises guarantees the truth of the conclusion.
Affirming the antecedent.
-If my theory is true, then I should observe X.
-My theory is true.
-I observe X.
Denying the consequent.
-If my henry is true, then I should observe X.
-I did not observe X.
-Therefore, my theory is not true.
The line of demarcation between science and non-science.
We ca show that the consequences of a theory (or an idea) are not empirically supported. Modus Tollens.
US Department of Health & Human Services creates three ethic principals.
1. Respect for persons (autonomy): participants are treated as autonomous, capable of making deliberate decisions about participating in the research.
2. Beneficence: maximize benefits and minimizing any possible harmful side effects to the research participants.
3. Justice: ensure the fair distribution of costs and benefits to potential research participants - same cost/benefit
Tuskegee Syphilis Experiment
A study of black male sharecroppers in Alabama with Syphilis. Were given 'treatment' but not to cure - blood tests. Many had sex with spouses and had children born with congenital syphilis.
Participants should be provided with all information that might influence their decision to participate.
But not telling them the hypothesis.
-assures voluntary participation, withdrawal without penalty, confidentiality
Fear of retribution if choose not to participate.
Benefits are too great (undue influence).
Providing an incredibly large payment for participation.
Omission - withholding information
Commission - actively lying
APA = permissible under three conditions:
-research is important
-no foreseeable harm
-Dehoaxing - revealing study's purpose and hypothesis
-Desensitizing - reducing any stress caused
-participation can be seen as a learning experience
-participants respond more positively to research process
Loss of Privacy
Institutional Review Board (IRB) - criteria
-Beneficence (risks minimized)
-Risks reasonable in relation to benefits (beneficence)
-Selection of participants is equitable (justice)
-Informed consent sought (respect)
-Informed consent documented (respect)
IRB Review Procedures
Based on risk
Those exempt from review:
-anonymous surveys, questionnaires, educational tests, naturalistic observation, archival research
-minimal risk (routine risk)
-studies with common procedures
Full Board Review:
-greater than minimal risk
-all the rest
Non-specialists may pass judgment on procedures that they do not understand.
Overzealous in concern for risk of doing research (not enough focus on costs of not doing research).
Variability/inconsistency in decision-making.
APA Code of Ethics
5 general Principles that are an extension of Belmont Principles (includes same three) plus fidelity and responsibility, and integrity.
Fidelity and Responsibility
Establish trust: stay aware of responsibility to society and exemplify the highest standard of professional behavior in their roles as researchers. Use training responsibly, manage conflicts of interest appropriately.
(therapists must avoid dual-relationships with their clients.
Promote accuracy, honesty and truthfulness in the science, teaching and practice of psychology.
Compels researchers to be honest is all aspects of research.
Deliberately taking the ideas of someone else and claiming them as one's own.
Direct: copying words verbatim without citation
Accidental: paraphrasing someone else's idea without citation.
Failure to be honest in managing data
Forms of fabrication: manufacturing data/results
Forms of falsification: altering/omitting collected data, guessing values of missing data.
Non-Human Animal Subjects (Institutional Animal Care and Use Committee, IACUC)
Replacement: use alternatives to animal research if possible
Refinement: minimize/eliminate animal distress
Reduction: employ designs or procedures that require fewest number of animals possible
Theoretical constructs that are stated in terms of concrete, observable procedures (to connect unobservable traits or experiences)
Has discrete (mutually exclusive) categories - sex
Frequencies and chi-square
Many levels or values that have meaning
Nominal: umbers arenames only (no order) - frequencies and chi-square
Ordinal: numbers indicate order but distance between not equal - nonparametric statistics
Interval: distance between equally spaces but no true 0 point - ANOVA
Ratio: true 0 point - ANOVA
Each person in population has an equal likelihood of being in your study - generalizability
Each person in your sample has equal likelihood of being in any condition - causality
Law of Large Numbers
Sampling - the more people in your population, the better your generalizability
Assignment - the more people, the better for causality
Factors - when more than one, considered a 'factorial' design
Each participant is in one and only one condition.
Some (or all) participants undergo all treatment conditions.
Relationship between two or more variables
Measures of central tendency, variability
-mean, median, mode
-range, variance, standard deviation
Good for nominal variables or if need to know more frequent observation.
Quick and easy
Good for "bad" distributions, or those with arbitrary ceilings/floors
Most reliable, inference as well as description, best estimator (all data included).
Not for bad distributions because sensitive to every score or for categorical data.
How much the scores fluctuate.
-Range: difference between largest and smallest.
-Variance: average squared deviation of each score from the mean.
-Standard Deviation: square root of the variance.
Statement of how two or more things might be related.
Statement of how a cause predicts an effect (why related). Testable, falsifiable, parsimonious, fruitful.
Decide whether true/false based on observation.
"a scientific hypothesis must generate predictions"
Can it be disproved.
Choose simpler or more frugal theory to explain a phenomenon to test.
Will it lead to further research?
Type I Error
False positive - rejecting null when shouldn't.
Type II Error
False negative - failing to reject null when should.
Reduce observer bias
-Use good operational definitions
-Participant or subject sampling