Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

Research Methods > exam three > Flashcards

exam three Flashcards

(71 cards)

Start Studying
1
Q

Weak experimental designs

A

0-1 comparisons
handpicked sample
2 groups or two treatments

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

Weak experimental design types

A
  • one group post test only
  • one group pretest posttest
  • post test only with nonequivalent groups
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

quasi experimental design

A

3+ comparisons
hand picked/cut off sample
pretest and post test tx conditions

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

quasi experimental design types

A

nonequivalent comparison groups
time series
regression discontinuity

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

strong experimental design

A

2+ comparisons
randomly assigned sample
multiple tx possibilities

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

strong experimental design types

A

between participants
within participants
factorial design

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

Quasi-Experiments

A 
Selecting subjects for conditions
Subject Variable
Observe categories of subjects
Moderate level of control
Open to alternative hypotheses
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

True Experiments

A 
Assigning subjects to conditions
Independent Variable
Manipulate variables for outcome
Highest level of control
Rules out alternative hypotheses
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Quasi-Experimental Designs

A

A research design in which an experimental procedure is applied but all extraneous variables are not controlled

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

Nonequivalent Comparison Group Design (NCGD)

A

Both the control and treatment conditions are given a pre and posttest for multiple comparisons
Goal is to rule out learning effects or natural growth effects
Similar to the Pretest-Posttest Control-Group Design that we saw in the between-subject type of strong experimental designs

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

Interrupted Time-Series Design

A

Giving the pretest and posttest many times to participants, in order to compare their results over time
Goal is to find patterns of change
Similar to the One-Group Pretest-Posttest Design that we saw in the within-subject type of strong experimental designs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Long-Term Research Studies

Longitudinal

A

These studies keep track of participants across the whole time span
Maturation and learning are better noted
Mortality can greatly decrease sample size over time

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

Long-Term Research Studies

Cross-sectional

A

These studies allow for a full study to take much less time by utilizing multiple comparison groups at different stages or ages
The groups are not as comparable as if they were the same participants
Save experimenters in time, energy, and money

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

Regression Discontinuity Design

A

Requiring participants to first take some measure of a trait you are looking for (the pretest measure), establish a cut-off point, and then only include participants that make your cut-off point in the treatment (whether you want below it or above it)
The ones who do not make your cut-off point become your control group
Then you posttest both and compare as normal

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

Program Evaluation types

A

Summative Evaluation

Formative Evaluation

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

Program Evaluation planning and procedures

A 
Identify the Stakeholders
Arrange Preliminary Meetings
Decide on Evaluative Need
Examine the Literature
Determine the Methodology
Present a Written Proposal
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

factorial designs

A

two or more IV’s (or factors) are studied to determine their separate and joint effects on the DV
Computes Main Effects & Interactional Effects

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

factorial design types

A

Between-Subjects Design
Within-Subjects Design
Mixed Model Design

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

3 Types of Interactions

Antagonistic Interaction

A

When the IVs perfectly reverse each other’s effects

There is an important interaction, but no real main effects for either; they only work well when used together

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

3 Types of Interactions

Synergistic Interaction

A

Where an IV increases the effect of the other

There will be larger differences in the first factor when adding in the other factor

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

3 Types of Interactions

Ceiling-Effect Interaction

A

Where an IV decreases the effect of the other

There will be smaller differences in the first factor when adding in the other factor

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Mixed Model Design

A

Combines between-subject and within-subject design components in a single research project

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Mixed Methods Research

A

Collects quantitative and qualitative types of data in a single research project and/or has both experimental & non-experimental components in a single research project

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

Planning Mixed Methods Research

Time Order Dimension

A

Concurrent- quant and qual

Sequential

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Planning Mixed Methods Research | Paradigm Emphasis Dimension
Equal Status | Dominant Status
26
Types of Relational Research | Contingency Research
determines the if-then sequence of events between two or more variables
27
Types of Relational Research | Predictive Research
using the relationship between concepts to make predictions about additional persons or samples
28
Relational Research
Typically computed through correlational analyses. No intervention or controls are utilized in this research design. Gives evidence that two variables have a likelihood of changing together when under certain circumstances,
29
Causational Research
Typically computed through differential analyses. May have control groups or a pre/post-Intervention design. Gives evidence that one variable causes a change in another variable.
30
Contingency Research provides only
descriptive results of the data: | Percentages, Counts, frequencies
31
Contingency Research
Is typically between 2 discrete variables that each have multiple levels, so that a researcher can see all the pairwise combinations (or one continuous with a discreet) Often provided as an accompaniment to other relational research options Displayed through a contingency Table, which is computed through the non-Parametric test, Chi Square
32
Contingency Tables
D:A table describing the degree (or percent) of relationship between differing levels of the variables being studied Based on the number of levels for each variable, the table will be called a #x# Contingency Table Next slides display example Contingency Tables known as a frequency report or crosstabulation (“Crosstabs”) report based on type of variable(s) being represented
33
Correlational Research
Provides inferential results of the data Level of relationship between variables, significance, meaningfulness Is typically between 2 continuous variables where one is known to occur first chronologically (or one continuous with a discreet) Determines level of confidence in prediction between variables Computed through a Pearson’s r or Phi correlation Coefficient analysis (shown in matrix form if more than 2 variables)
34
Pearson’s r Correlation Coefficient
D: A relational analysis in which 2 continuous variables are studied to determine their level of connection with one another Assesses patterns in the variables in your sample to provide information on the strength and direction of the relationship between them Numerically ranges from +1 to -1, with numbers close to zero meaning no relationship between variables Also known as Pearson’s Product Moment Correlation Coefficient
35
Interpreting the Correlation Coefficient | strength
``` Very Strong Relationship 0.81 to 1.00 Strong Relationship 0.61 to 0.80 Moderate Relationship 0.41 to 0.60 Weak Relationship 0.21 to 0.40 Very Weak or No Relationship 0.00 to 0.20 ```
36
Interpreting the Correlation Coefficient | direction
DIRECT RELATIONSHIP Positive r As variable 1 increases, so does variable 2 As variable 1 decreases, so does variable 2 INDIRECT RELATIONSHIP Negative r As variable 1 increases, variable 2 decreases As variable 1 decreases, variable 2 increases
37
Predictive Research
Provides Predictive or Follow-up results of the data Predictability between variables, Ability to form new outcomes from predictors Is typically between 2 continuous variables (it must be known which would predict the other) Often use as a follow-up for Correlational Research Computed through a Regression analysis (Called Multiple Regression if there is more than one predictor)
38
Regression Analysis
How regression works: If X is the predictor and Y is the outcome, then we could predict a person’s Y once given their X We use our sample data to determine the values of a and b (through equation computation) and fill them in to the regression equation (seen above)
39
Types of Research
Nonexperimental Research Experimental Research Quasi-Experimental Research Mixed Methods Research
40
Types of Nonexperimental Research
``` Naturalistic Observation Research Case Study Research Survey Research Archival Research Relational Research Predictive Research ```
41
Naturalistic observation research
Observing and recording ongoing behavior without altering circumstances
42
Case Study Research
Exploring a single existing concept to create and test hypothesis
43
Survey Research
Collecting data by polling or interviewing participants
44
Archival Research
Examining existing records to test hypotheses
45
Relational research
determining the type and level of relationship between two or more constructs
46
Predictive research
using the relationship between concepts to make predictions about additional persons or samples
47
Major Methods of Data Collection
``` Test Questionnaires Interviews Focus Groups Observation Existing or Secondary Data ```
48
Terms for Nonexperimental Research
Qualitative Research | Correlational Research
49
Qualitative Research
Goal is to determine how people understand themselves and make meaning out of circumstances
50
Correlational Research
Goal is to measure 2+ variables and determine the degree of relationship between them
51
Examples of Qualitative Data
``` An audiotaped interview Photographs of behaviors Participant journals Examinations of rare possessions Autobiographies ```
52
What is Nonexperimental Research?
``` Studies real-world (naturally-occurring) situations Has a flexible design Aimed at insight, not inference Focuses on qualitative over quantitative data Can collect more complex information Attention to process, not product Forfeit a degree of control Researcher’s subjectivity is key Value creativity in analysis ```
53
Naturalistic Observation Research | 3 main types
Unobtrusive Research Participant-Observer Research Laboratory Observation Can be difficult to statistically analyze data as it will likely be primarily (or entirely) qualitative
54
Case Study Research | 3 types
Intrinsic Case Study Instrumental Case Study Collective Case Study
55
Intrinsic Case Study
When researcher is interested in a single participant (or small group)
56
Instrumental Case Study
When researcher studies the individual case to gain an understanding of something beyond that case
57
Collective Case Study
When researcher studies multiple cases for comparison
58
Survey Research is conducted through
Personal Interviews Phone/Internet Interviews Group Administration Self-Report Questionnaires
59
When seeking questionnaires for your own research projects:
Search online Some questionnaires are online for free See if the journal articles who used that questionnaire contain the desired questionnaire in full (either in an appendix or in a chart with the method or results sections of the article) Once you obtain a survey instrument, search for journal articles that report its reliability and validity properties If you can’t find the survey, contact the author or publisher of that survey instrument for permission and access If nothing exists for your area of research, construct your own
60
Questionnaire Construction-12
Write items to match research objectives Use language clear for target population Make questions simple and short Avoid loaded or leading questions Avoid double-barreled questions Avoid double negatives Determine whether to use closed- or open-ended questions Create response categories for closed-ended questions Consider different types of response categories Cover complex ideas over multiple questions Make sure questionnaire is user-friendly Pilot until perfected
61
Archival Research
``` Type of research in which new data is not being created, but is being compiled to prove hypothesis Examples: Public Records Published Writings Historical Records Journal Entries Photographs Government Documents ```
62
Phenomenology
describes how one or more participants experience a phenomenon
63
Ethnography
describes a cultural group ot a group of people with some common theme
64
Grounded Theory
collecting data with the sole purpose of creating a psychological theory
65
Single-Subject Designs
Studies one individual or one small group Single-Subject Design ≠ Single-Case Study Must use within-subject designs for comparisons Also known as Single-Case Designs
66
Single-Subject Designs Types:
ABA & ABAB (“Reversal”) Designs Multiple-Baseline Design Changing-Criterion Design Interaction (“Alternating Treatment”) Design
67
AB setup
A- measure at baseline -give tx B-measure after tx
68
ABA setup
``` A- measure at baseline -give tx B-measure after tx -w/draw tx A- measure at baseline ```
69
ABAB
``` A- measure at baseline -give tx B-measure after tx -w/draw tx A- measure at baseline -give tx B-measure after tx ```
70
Advantages of single subject designs
Focuses on Individual Performance Focuses on Big Effects Avoids Ethical & Practical Problems Flexible in Design
71
Disadvantages of single subject designs
Low Power Small Effects are Invisible Control is Difficult No Between-Subject Design Possibilities

Research Methods (3 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions