final Flashcards
(160 cards)
1
Q
What measurement property does a nominal scale have?
A
Nominal scales are categorical, not numeric.
Example: Gender (Male = 1, Female = 2)
2
Q
What measurement properties does an ordinal scale have?
A
Ordinal scales have properties of difference and magnitude.
Example: Ranking in a race.
3
Q
What can you conclude based on an ordinal scale that you can’t conclude based on a nominal scale?
A
You can determine the order of items, but not the magnitude between them.
4
Q
What measurement properties do interval scales have?
A
Interval scales show differences, magnitude, and equal intervals.
Example: Temperature.
5
Q
What measurement properties do ratio scales have?
A
Ratio scales have an absolute zero and allow for comparison of absolute magnitudes.
Example: Length.
6
Q
What does it mean for an instrument to be reliable?
A
It means having a consistent way to measure the items at hand.
7
Q
What does it mean for an instrument to be valid?
A
It means that the instrument is the correct tool to measure the data.
8
Q
What is a reliability coefficient?
A
A type of correlation that indicates consistency.
9
Q
Why do we want the reliability coefficient to be strong and positive?
A
It indicates that the data is being measured correctly and consistently.
10
Q
Why must a measure be reliable in order to be valid?
A
A measure needs to be consistently correct to be valid.
11
Q
Can a measure be reliable but not valid? Give an example.
A
Yes, an example is a clock that consistently shows the same time but is incorrect.
12
Q
What is ‘test-retest reliability’?
A
The consistency of individual scores over time.
Example: IQ tests administered twice.
13
Q
What statistic is used to assess test-retest reliability?
A
Pearson correlation.
14
Q
What indicates high reliability in test-retest reliability?
A
+1.0 correlation.
15
Q
What is ‘equivalent forms reliability’?
A
The consistency of scores on two different tests measuring the same construct.
Example: Two anxiety tests.
16
Q
What is ‘internal consistency’?
A
The consistency with which items on a test measure a single construct.
17
Q
How is internal consistency assessed?
A
Using Cronbach’s alpha measure.
18
Q
What indicates good internal consistency?
A
A Cronbach’s alpha of +0.70 or higher.
19
Q
What is ‘interrater reliability’?
A
The agreement between two or more observers.
20
Q
What statistic is used to indicate interrater reliability?
A
Correlation of ratings.
21
Q
What is ‘interobserver agreement’?
A
The percentage of times different raters agree on nominal data.
22
Q
What are constructs?
A
Things that researchers study but cannot physically measure.
23
Q
What is construct validity?
A
The accuracy of operational definitions in measuring the intended construct.
24
Q
What is ‘content validity’?
A
Judgment by experts of how well test items represent the construct.
25
What is the difference between unidimensional and multidimensional constructs?
Unidimensional measures one trait; multidimensional encompasses multiple traits.
## Footnote
Example: Math anxiety (unidimensional) vs. intelligence (multidimensional).
26
What statistical technique is used to determine the number of dimensions in a construct?
Factor analysis.
27
What does 'homogeneity' mean in the context of constructs?
The degree to which a set of items measure a single construct.
28
What is a validity coefficient?
A type of correlation used in validation research.
29
What is 'criterion-related validity'?
How well a test relates to a criterion.
30
What is 'predictive validity'?
How well one test can predict future behavior or performance.
31
What is 'concurrent validity'?
How well scores obtained at one time relate to scores on a known criterion at the same time.
32
What is 'convergent validity'?
The extent to which test scores relate to other measures of the same construct.
33
What is 'discriminant validity'?
The extent to which test scores do not relate to measures of different constructs.
34
What is 'known groups validity evidence'?
Evidence that a test can distinguish between groups known to differ within that construct.
35
What is a 'sample'?
A small selection of participants being measured.
36
What is a 'population'?
The larger group researchers aim to study.
37
Why is it not normally important to have a random sample in an experiment?
Experiments focus on understanding behavior and causality, not generalization.
38
Why is it usually important to have a random sample in a survey?
To generalize findings to a larger population.
39
What is a 'representative sample'?
A sample that resembles the population.
40
What is the 'equal probability selection method (EPSEM)'?
A method ensuring everyone has an equal chance of being selected.
41
What is 'sampling error'?
The difference between a sample and the population.
42
What is a 'sampling frame'?
A list of everyone in a population.
43
What is 'simple random sampling'?
A method where each participant has an equal chance of selection.
## Footnote
Example: Randomized.org.
44
What is 'stratified random sampling'?
Selecting participants from mutually exclusive groups (strata).
## Footnote
Example: Employees in different departments.
45
What is 'proportional stratified sampling'?
Sampling that is proportional to the population size.
46
What is 'cluster random sampling'?
Randomly selecting groups (clusters) of individuals.
47
What are the three steps involved in systematic sampling?
1. Determine sampling interval (k). 2. Randomly select a starting point. 3. Include every k-th element.
48
What is 'convenience sampling'?
Using readily available participants.
## Footnote
Example: Students in a campus pool.
49
What is 'quota sampling'?
Meeting a specific number of participants from certain groups.
## Footnote
Example: 25 males, 25 females.
50
What is 'purposive sampling'?
Identifying individuals with specific characteristics.
## Footnote
Example: First-time parents with autistic children.
51
What is 'snowball sampling'?
Participants identify other participants.
## Footnote
Example: Studying drug users.
52
What is the goal of 'random selection'?
To obtain a representative sample of the population.
53
What is random assignment?
Randomly assigning participants to be part of the control/experimental group. It distributes extraneous variables and helps determine causality.
54
What are some ways researchers determine sample size?
Consider population size, statistical power, and whether N < 100.
55
What is statistical power?
The probability of avoiding a type 2 error.
56
Which measurement scale is accurately paired with an example: a) Interval—standings of major league baseball teams b) Ratio—telephone area codes c) Nominal—total number of points earned in a class d) Ordinal—rank in the military?
D
57
What differentiates interval from ordinal scales of measurement?
In an interval scale, equal distances represent equal distances on the dimension being measured.
58
Which reliability coefficient value indicates the most reliability: a) 0.35 b) 0.85 c) -0.85 d) 2.20?
B
59
Gerald is developing a measure of shyness and he determines that students scoring high on the measure also score high for introversion on a well-known introversion-extraversion scale. The outcome best illustrates a) Face validity. b) Concurrent validity. c) Predictive validity. d) Construct validity.
D
60
To study whether California high school students plan on attending college, a researcher randomly selects 5% of the state’s school districts and gives all the students in each district a survey designed to measure college plans What sampling procedure is being used here? a) Quota b) Stratified c) Cluster d) EPSEM
C
61
A researcher who selects a probability sample of college students that is70% undergraduate and 30% graduate student is most likely to be using what sampling. a) Cluster b) Stratified c) Convenience d) Quota
D
62
What is 'statistical conclusion validity'?
It asks if there is really an effect in the population.
63
What is the main threat to statistical conclusion validity?
Type 2 error due to small sample size.
64
What is construct validity in plain English?
The accuracy in your operational definitions.
65
What does 'reactivity to the experimental situation' refer to?
Participants' motives and perceptions of the experiment.
66
What are 'demand characteristics'?
Cues in the experiment that can influence participants' responses.
67
What is 'positive self-presentation'?
Changing behavior to be viewed positively by the researcher.
68
What should researchers do to rule out demand characteristics and positive self-presentation?
Hold participants' perceptions constant.
69
What is the 'experimenter expectancy effect'?
Ways the experimenter might bias the study.
70
What are the three categories of experimenter attributes that might bias a study?
1. Biosocial attributes 2. Psychosocial attributes 3. Situational factors.
71
What is internal validity?
The degree to which the study accurately reflects the causal relationship.
72
What is recording bias?
It refers to the influence of experimenters' expectancies on participant responses and can affect both human and animal subjects.
## Footnote
It can be mediated by handling in animal studies.
73
What are the three categories of experimenter attributes that might bias a study?
* Biosocial attributes (e.g., age, gender, race, religion)
* Psychosocial attributes (e.g., anxiety level, need for social approval, hostility)
* Situational factors (e.g., relationship between experimenter/participant, experience level of the experimenter)
74
What is internal validity?
It is defined as inferences about cause and effect.
75
What is the primary threat to internal validity?
Confounds.
76
What does 'THRIMMS' stand for?
* T - Testing
* H - History
* R - Regression
* I - Instrumentation
* M - Maturation
* M - Mortality (attrition)
* S - Selection
77
Define the 'history' threat to internal validity.
It refers to any external event that can impact the dependent variable (DV).
78
Define the 'maturation' threat to internal validity.
It refers to any internal change in the participant that can affect the DV.
79
Define the 'instrumentation' threat to internal validity.
It refers to changes over time in the measurement of the DV and is likely to occur in a pre-test, post-test model in observational research.
80
How might instrumentation threat be avoided?
By training observers.
81
Define the 'testing' threat to internal validity.
It refers to the changing of scores simply because participants have seen the test before.
82
What is 'statistical regression to the mean'?
It refers to the tendency for extreme scores to be closer to the mean on the posttest.
83
What is the 'attrition' (or mortality) threat to internal validity?
It refers to the dropout rate of participants.
84
Define the 'selection' threat to internal validity.
It refers to the production of non-equivalent groups due to differing selection processes.
85
How can selection threat be prevented?
Through the use of random assignment.
86
What is external validity?
It refers to the degree to which the study results can be generalized.
87
What is 'population validity'?
It asks if the results can be generalized from the sample to the target population.
88
What is 'ecological validity'?
It refers to the results being able to be generalized to different settings/environmental conditions.
89
What is temporal validity?
It refers to the generalization of results across time.
90
What is 'seasonal variation'?
It is a type of cyclical variation in which the values on the dependent variable vary by season.
91
What is 'cyclical variation'?
It refers to any systematic type of movement over time, such as circadian rhythms.
92
What is 'treatment variation validity'?
It talks about the results generalizing across variations in the treatment.
93
Give an example of treatment variation validity.
Cognitive-behavioral therapy reduces depression only if administered in a certain way.
94
What is 'outcome validity'?
It talks about generalizing the results to different but similar dependent variables.
95
Give an example of outcome validity.
Behavior management for kids with ADHD leads to more time on class work and less classroom distractions.
96
How is external validity established?
Through replication.
97
Describe the relationship between internal validity and external validity.
They are inversely related; increasing one often decreases the other.
98
Which is more important, internal validity or external validity?
Both are important; it depends on the goal of the study.
99
What are demand characteristics?
Participants' responses influenced by their perceptions of the experiment.
100
In the context of a psychological experiment, what is a target population?
The group of individuals to whom researchers want to generalize their findings.
101
What is random assignment?
The process where participants have the same chance of being in the control or experimental group.
102
What is the main advantage of random assignment?
It distributes extraneous variables equally across groups.
103
What is matching in experimental research?
The process of pairing participants based on specific characteristics.
104
What are the advantages of matching?
* Controls for matched variables
* Increases statistical power
105
What is a major disadvantage of matching?
Participants may not be equal on non-matched variables.
106
What is 'matching by holding variables constant'?
Selecting one variable to match groups, which can restrict population size.
107
What is 'matching by building the extraneous variable into the design'?
Using a confound as another independent variable to test its influence.
108
What is matching by yoked control?
Matching participants based on the temporal sequence of administering an event.
109
What is a within participants design?
A design where all participants receive all treatment conditions.
110
What are order effects?
Changes in behavior due to the sequence of treatments.
111
What are carryover effects?
Influence of previous treatments on current behavior.
112
What is counterbalancing?
Randomizing the sequence of conditions to minimize order effects.
113
What is the double blind placebo model?
A study design where both experimenter and participant are unaware of group assignments.
114
What does deception control for in experiments?
Participant effects.
115
What is a research design?
An outline/plan for an experiment.
116
What are the two purposes of a research design?
* Control for unwanted variation
* Suggest how new data will be analyzed
117
Why does the 'one-group posttest only design' not allow causal conclusions?
It lacks pre-test or control group to determine causality.
118
What is the improvement of the 'one-group pretest-posttest design' over the 'one-group posttest only design'?
It includes a pretest group.
119
What is the major weakness of the 'Pretest-posttest control-group design'?
Inability to generalize findings due to ecological validity.
120
What is a factorial design?
A design that has two independent variables.
121
What is a main effect?
The effect of one independent variable.
122
What is an interaction effect?
It indicates that the effect of one independent variable depends on the level of another.
123
What is a mixed model factorial design?
Involves one independent variable as within participants and another as between participants.
124
What is the independent variable in Dr. Benning's study?
drug (nicotine or placebo)
## Footnote
The independent variable is what is manipulated in the experiment.
125
What is the dependent variable in Dr. Benning's study?
reaction time
## Footnote
The dependent variable is what is measured in the experiment.
126
How did Dr. Benning operationally define "reaction time"?
as the amount of time between stimulus presentation and key depression
## Footnote
This operational definition specifies how the concept is measured.
127
What factor varied between participants in Dr. Benning's study?
type of drug
## Footnote
This refers to the different conditions each participant experienced.
128
What factor varied within participants in Dr. Benning's study?
there was no within-participants factor
## Footnote
This indicates that all participants experienced only one condition.
129
What method describes Dr. Benning's approach if an assistant decides participant drug assignments without her knowledge?
double blind
## Footnote
In a double-blind study, neither the participants nor the experimenter knows who receives the treatment.
130
How could Dr. Benning have controlled for the extraneous variable of age affecting reaction time?
by matching each participant in the experimental group with a participant of the same age in the control group
## Footnote
This approach ensures age equivalence between groups.
131
What technical advancement did Dr. Benning use to reduce experimenter expectancy effects?
she computerized the data collection
## Footnote
Computerized data collection reduces human error and bias.
132
What is Dr. Benning's best option for ensuring that her participants are essentially equivalent between the two groups?
randomly assigning them to groups
## Footnote
Random assignment is crucial for ensuring group equivalence.
133
How could Dr. Benning control for the extraneous variable of smoking status?
all of the above
## Footnote
This includes using only non-smokers, making smoking status another independent variable, and matching participants.
134
Dr. Benning’s use of prisoners raises questions regarding what?
population validity
## Footnote
This concerns whether the findings can be generalized to the larger population.
135
How could Dr. Benning’s study be improved?
use non-prisoners
## Footnote
This could enhance the generalizability of the findings.
136
What type of design is described as a 2X2 between-participants factorial design?
2X2 between-participants factorial design
## Footnote
This indicates two independent variables with two levels each.
137
How many independent variables are in the described experiment?
two (time of day and dose of drug)
## Footnote
Each variable influences the outcome of the experiment.
138
If ten rats are tested in each group, how many rats are required in total?
60
## Footnote
This assumes there are 6 groups (2x3 design).
139
The dependent variable is operationally defined as what?
the movements counted by the computer
## Footnote
This specifies how the dependent variable is measured.
140
The dependent variable in this context is what type of variable?
quantitative variable
## Footnote
Quantitative variables are measurable and can be expressed numerically.
141
In order to make this experiment partly a blind procedure, who should not know the drug group of the rats?
the person handling and injecting the rats should not know which drug group the rats are in
## Footnote
This helps to reduce bias in treatment administration.
142
What do the results suggest in the table regarding the effects of the independent variables?
an interaction of time of day and drug dose
## Footnote
This indicates that the effect of one variable depends on the level of another.
143
Which of the following is an accurate interpretation of the outcome?
a) The drug has no effect on activity level of rats.
b) The time of day has no effect on the activity level of rats.
c) The effect of drug on activity level of rats depends upon the time of day.
d) The drug is a stimulant and therefore a good appetite suppressant.
C
## Footnote
This suggests a complex relationship between variables.
144
What differentiates quasi-experimental designs from true experimental designs?
There is no complete control
## Footnote
Quasi-experimental designs have partial control over extraneous variables.
145
What are the two strategies discussed to make causal inferences in quasi-experiments?
Investigating plausible threats to internal validity, Control by design
## Footnote
These strategies help improve the validity of findings.
146
What design feature does the “nonequivalent comparison group design” lack?
random assignment
## Footnote
This is critical for ensuring equivalent groups in true experiments.
147
What is the most likely threat to internal validity in a nonequivalent comparison group design?
the threat of History
## Footnote
Historical events may affect the groups differently.
148
Define the “selection-maturation” threat to internal validity.
inequivalent groups may develop different speeds
## Footnote
This can skew results if one group matures faster than another.
149
Define the “selection-history” threat to internal validity.
inequivalent groups will be impacted by external events differently
## Footnote
This means one group may experience effects that another does not.
150
Define the “selection-regression” threat to internal validity.
If the groups start off uneven, one group may regress to the mean more than the other; which possibly implies there is an effect when there isnt.
## Footnote
This can lead to misleading results if groups are not truly equivalent.
151
How is a treatment effect revealed in an “interrupted time series design”?
by a different pattern in the pre & post test design
## Footnote
This indicates a change due to the treatment.
152
What is the primary threat to internal validity in an “interrupted time series design”?
History
## Footnote
This refers to external events influencing the outcome.
153
What indicates NO treatment effect in the regression discontinuity design?
the line would be continuous and there would be no shift in the graph line
## Footnote
This suggests that the treatment did not have an effect.
154
What indicates a treatment effect in the regression discontinuity design?
there would be a break in the line on the graph
## Footnote
This suggests a significant change due to the treatment.
155
What are the two primary threats to internal validity in a regression discontinuity design?
differential history, differential attrition
## Footnote
These threats can affect the validity of the results.
156
When might one consider using quasi-experimental designs?
when one does not have the ability to randomly assign participants to groups
## Footnote
Quasi-experimental designs are often used when random assignment is not feasible.
157
What must Dr. Kaiser decide after administering the aerobic dancing program?
if there was a selection-history effect
## Footnote
This question addresses whether improvements were due to the program or other factors.
158
What experimental design is used by Dr. Torres to test an incentive program?
one group pretest-posttest
## Footnote
This design compares the same group before and after the intervention.
159
What is the 'interruption' in an interrupted time series design?
period of 2 weeks or more
## Footnote
This marks the point at which the treatment is applied.
160
What type of experimental design was employed in Vaughn et al. (2009) study?
regression discontinuity design
## Footnote
the group they were in determined whether they recieved the treatment or not
