Exam 2 Flashcards

1
Q

What are the four maxims that guide research participants’ interpretations when they are involved in an experiment?

A
  1. Relation - want to make contribution to research, take into account context of questions
  2. Quantity - will give only info researcher asks for, no more
  3. Manner - contribution clear, researcher had a purpose in designing Qs
  4. Quality - assume truthfulness, no deceit
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What were the key studies that prompted the development of ethical guidelines?

A

Milgram’s obedience study - extreme stress, psychological damage
Zimbardo’s prison study - extreme stress, psychological damage, danger
Micturation delay study (bathroom, personal space) - no consent
Humphrey’s study on gays - breach of privacy, no anonymity
*any deception is unethical - deontological position
*benefits must outweigh costs - utilitarian

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is debriefing?

A

Educate participant, completely explain purpose
Identify deception and explain
Query participants about responses
Address concerns

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is process debriefing?

A

More exact (ask them about their logical processes)
Major deceptions, unexpected behavior
Necessary for receipt of performance
May need to be iterative process

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are the defining features of the correlational research method?

A

Measuring 2 or more variables

Change in 1 variable associated with another - one variable assumed to be cause (correlation does not imply causation)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What are the different types of variables that can be tested using the correlational method?

A

Predictor: causal factor (like IV)
Criteria: outcome variable (like DV)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the difference between a categorical/continuous variable and a discrete variable?

A

Continuous: can take on any value between two values (1, 1.25, 3.4444…) ex. intelligence
Discrete: only integers (1, 2, 3…) ex. gender

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the causal inference problem?

A

statistical relationship could reflect multiple possible causal relationships
motivation -> performance, motivation performance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the third variable problem?

A

Another variable might be influencing your results
self-esteem -> performance (what else could be affecting it?)
ex. polio - thought to be caused by diet, really just more seasonal outbreaks

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is the importance of looking for non-linear relationships among variables?

A

correlation only examines linear relationship - can miss out on different relationships
theory may posit another type
formulas you use to test for other relationships
curvilinear relationships (ex. stress and arousal)

Leary:

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What are the guidelines for interpreting the size of correlations? What factors should be taken into consideration?

A
|.1| = small
|.3| = medium
|.5| = large
take into account:
size predicted by theory
were vars assessed by different methods?
sample size
small correlations can be meaningful
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is restriction of range?

A

Due to sample, one or both variables fail to show complete variation
ex. SAT & 1st year GPA - Clemson’s sample will only have medium to high range SAT scores based on acceptance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the difference between a mediator and a moderator?

A

Mediation: what accounts for the relationship between 2 variables; “responsible for”
Moderation: relationship between 2 variables influenced by 3rd variable; “strength of relationship”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What are the strengths of a longitudinal design?

A

Measure the same variables at two points in time; look at patterns of correlation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is the logic behind multiple regression?

A

use of 1+ variables to predict another variable

better prediction with predictors that account for unique variance (doesn’t share with other predictors)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are the four types of multiple regression?

A
  1. Standard: all variables go in at once, 3 or 4 max
  2. Stepwise: highest correlation goes in first (order, one-at-a-time, lots of variables)
  3. Hierarchical: order based on theoretical relevance
  4. Logistic: categorical (binary) outcome, focus on accuracy of predicting 2 groups (ex. graduated or not)
17
Q

What are the defining features of experimental research?

A

IV must be manipulated

Random assignment

18
Q

Why are control groups important?

A

Control group receives exact treatment as IV, but “zero level” of IV
Placebo effect: tendency for people to report a treatment has benefitted them regardless of whether or not treatment is present
Control group important so you have a baseline to compare experimental group to, placebo to see if effects are real or in their head

19
Q

Why are manipulation checks important?

A

increases validity
just because you manipulate the situation does not mean you are manipulating the IV
ensure manipulation of IV was effective
Usually come in the form of questions completed at the end of the study to assess effectiveness
(ex. to what extent were you thinking of yourself as a female during this imposter syndrome experiment”)

20
Q

What is error variance?

A

cannot be avoided

numerous influences on DV not related to TV (ex. mood, time of day, personality)

21
Q

What is a confound?

A

factor you unintentionally manipulate with your IV that may account for relationship between IV & DV
ex. time management training, control training -> how well you do in this class

22
Q

What is internal validity?

A

being confident that the differences in DV are because of IV

23
Q

What are some threats to internal validity?

A
  1. history - historical event influences effect
  2. differential attrition - different number of subjects in each group quit
  3. demand characteristics - subjects know what you’re testing for and give a response that is expected
  4. biased assignment - no random assignment
  5. differential treatment of subjects
24
Q

What is external validity?

A

will the relationship between the IV and DV be found in the “real world”

25
What are some problems with testing external validity?
effect size of tests in the lab and the field
26
How do you describe a design involving more than one IV?
factorial design | levels in IV1*IV2 = # conditions
27
What are the advantages and disadvantages of repeated measures?
``` + remove error of different people + decreases # participants needed - carryover effects: exposure affects responses to other levels - fatigue - habituation ```
28
What are the two types of conditions involving people in experimental conditions?
within-subjects | between-subjects
29
Why would a researcher want to measure the DV before manipulating the IV?
pre-test, post-test design, necessary when examining change | pretest might sensitize participants to purpose, shouldn't immediately precede experiment
30
What is matched random assignment?
divide subjects on attribute, randomly assign to conditions
31
What is a mixed-model design?
When you measure one variable and manipulate one | useful if interested in how personality interacts with manipulation
32
What are the two uses of an analysis of covariance?
Can be used to test why an IV is causing change in DV ex. nature sounds high math performance than control (relaxation?) Can be used to decrease error term in study - want your covariate to be related to DV, not IV ex. stereotype threat effects on math performance, math SAT as covariate
33
What is a quasi-experiment?
non-random assignment manipulation of IV not directly controlled by researcher overcome some limitations of correlational and experimental designs
34
What are the three types of Q-E?
1. non-equivalent control group design - ex. target school, control group; can't ensure comparison is the same 2. natural experiments - naturally occurring "manipulations" ex. hurricane, terrorist attack 3. time series designs - multiple pre- and post- measurements to study changes in trends, captures ongoing trends over time; ex. highly publicized suicides and single-car crashes
35
What are some challenges of conducting Q-E research?
Balance of internal and external validity issues | often difficult to get data
36
What are criticisms of single-case designs?
Results do not generalize to everyone, but to "group averages" Psychologists often interested in predicting individual behavior Effect must not be important if you can't see it in an individual Do not study variability within an individual person over time
37
What are advantages of single-case designs?
Able to show effect of IV in each person examined | Can replicate finding within the study of one individual
38
What are the types of single-case designs?
ABA: baseline, introduce IV, remove IV; record frequency ABAB: replicate the study within an individual AB1AB2: have two different levels of an IV
39
What are the problems with single-case designs?
Question of generalization; cannot examine moderators | Involves inspection of patterns rather than statistical tests