Term
|
Definition
Changes in one variable are directly responsible for changes in another
True experiments attempt to establish cause and effect relationships by:
1. Manipulation
2. Measurement
3. Comparison (Statistics)
4. Control |
|
|
Term
| Independent Variable (IV) |
|
Definition
The variable being manipulated by the researcher
Treatment Conditions- Situation or variable characterized by one specific value of the manipulated variable
Levels- The different values of the IV selected to create and define the treatment conditions |
|
|
Term
| Cause and Effect Relationships |
|
Definition
Causation and the third-variable problem
The summertime murder rates/ ice cream sales study
Is there a confounding variable?
Causation and the directionality problem
Which one causes the other?
Controlling Nature
Create an unnatural situation where variables exist in isolation |
|
|
Term
| Distinguishing Elements of an Experiment |
|
Definition
General Goal: Establish cause-and-effect
Demonstrate that changes in one variable causes changes in the other
Rule out the possibility of a third variable causing changes
Tools utilized to reach goal
Manipulation
Control |
|
|
Term
| Dealing with Extraneous Variables |
|
Definition
Control prevents and extraneous variable from becoming a confounding variable
Confounds have two important characteristics:
Influences the DV
Vary systematically with the IV
Types of extraneous variables:
Environmental variables- Different rooms, weather, temperature, lighting
Participant variables- Gender, age, IQ, family structure
Time-Related variables- Fatigue, timing of study |
|
|
Term
|
Definition
Posttest-Only Designs
Pretest-Posttest Designs
Assigning Participants to Conditions
Independent Groups Design |
|
|
Term
|
Definition
Two equivalent groups of participants
Introduction of the IV
Measurement of the effect of the IV on the DV |
|
|
Term
|
Definition
A pretest is given to each group prior to introduction of the experimental manipulation
Assures that groups are equivalent at the beginning of the experiment
Can quickly measure changes that occur from the pretest to the posttest |
|
|
Term
| Advtanges/ Disadvantages Pretest- Posttest Design |
|
Definition
Advantages
Can evaluate Attrition/ Mortality ( dropout factor)
Assess equivalency of groups with small sample size
Can be used to sleet participants for the experiment
Disadvantages
Time consuming and awkward to administer
Sensitizes participants to what is being studied
Demand characteristics
Reduces external validity |
|
|
Term
| The Experimental Research Strategy |
|
Definition
Goal: Establish a casue and effect relationship
Four characteristics:
1. Manipulation
2. Measurment
3. Comparison
4. Control |
|
|
Term
|
Definition
| Different groups of scores are all obtained from the same groups of participants |
|
|
Term
|
Definition
| Differnt groups of scores all obtained from seperate groups of participants |
|
|
Term
| Controlling Extraneous Variables |
|
Definition
Two active control techniques:
1. Holding a variable constant
Standardized environment and precedes, specific sample
May limit external validity
2. Matching values across treatment conditions
Same number (each gender/age) in each condition
Same average score in each condition
Vary order of treatments
Time consuming, tedious
One passive control technique
3. Randomization
Simpler technique
Using a random process to disrupt the natural/ systematic relationship between two variables |
|
|
Term
| If you are investigating test performance and self-esteem: |
|
Definition
Manipulate IV
Administer a general knowledge test- one with obviously difficult questions and one with obviously easy questions
Extraneous Variables
Gender, confusion, depression, hair color
What might be considered confounds?
What? Why not?
Confusion (Varies with condition, might effect DV)
Gender and depressions ( Might effect DV- no systematic variation on IV)
Hair Color (No effect on DV, no systematic variation on IV |
|
|
Term
| Comparison: Control Groups |
|
Definition
Experimental group: treatment condition
Control group: no-treatment condition
1. No treatment control group
Do not receive the treatment being evaluated
Creates a standard of normal behavior (baseline)
2. Placebo control group
Receive an inert/ innocuous medication instead of the actual treatment |
|
|
Term
|
Definition
An additional measure to assess how participants perceived and interpreted the manipulation and/or to assess the direct effect of the manipulation
1. Explicit measure of the IV (mood questionnaire)
2. Ask questions after participation ( participants)
Particularly useful when there are:
Participant manipulations
Subtle manipulations
Simulations
Placebo Controls |
|
|
Term
| Between- Subjects Experimental Designs |
|
Definition
A.K.A Independent measure Experimental Design
Requires a separate, independent group of individuals for each treatment condition
Data contains only one score per participant
Manipulation of an IV
Control of extraneous variables
Goal: to determine if differences exist between two or more treatment conditions |
|
|
Term
| Between-Subjects Design Adv/Disad |
|
Definition
Advantages
Each score is independent of other scores
Useful for any research comparing treatment groups
Disadvantages
Requires many participants
Individual differences (extraneous variables)
Can become confounding variables
Can result in highly variable scores |
|
|
Term
| Individual Differences as Confounds |
|
Definition
Groups should be equivalent except for the level of the IV
Assignment bias
Environmental variables
Equivalent groups are
Created equally
Treated equally
Composed of equivalent individuals |
|
|
Term
|
Definition
Random assignment
Restricted random assignment
Holding variables constant
Restricting range of variability
Matched groups (assignment) |
|
|
Term
| Other Threats to Internal Validity |
|
Definition
In addition to assignment bias (individual differences) and confounding from environmental variables)
Differential Attrition
Mortality differences between groups —> groups are no longer equal
Communication between groups
Diffusion
Compensatory equalization
Compensatory rivalry
Resentful demoralization |
|
|
Term
| Two-Group Mean Difference |
|
Definition
Simplest version of design
Single-factor two group design or two group design
Requirements
IV- Two groups (nominal)
DV- One ratio (numerical)
Analysis
Independent measures t-test
Question
Is there a significant difference between the means of each group?
Advantages
Easy interpretation
Can maximize differences between treatment conditions
Disadvantages
Provides relatively little information
Limits control group options |
|
|
Term
| Within-Subjects Experimental Designs |
|
Definition
A single group of participants undergoes all of the different treatment conditions
Each individual is tested/observed in all treatment conditions
This is why its called repeated-measures design |
|
|
Term
| Between Groups (Disadvantages) |
|
Definition
Requires many people
Individual differences between groups can become a confounding variable
Alternate explanation
Individual differences in each treatment can create high variance
Obscures difference |
|
|
Term
| Disadvantages of Within- Subjects Design |
|
Definition
Threats to internal validity
Confounding from environmental variables
Confounding from time-related factors
History/ Maturation/ Instrumentation/ Testing effects/ Regression
Participant Attrition |
|
|
Term
|
Definition
Not directly connected to participating in a previous treatment
History/ maturation |
|
|
Term
|
Definition
Directly related to experience obtained by participating in a previous treatment
Carryover Effect: Changes in behavior or performance caused by the lingering, after-effects of an earlier treatment condition
Progressive Error- Changes in behavior or performance related to general experience in a research study but not related to a specific treatment or treatments
Practice effect, Fatigue effect |
|
|
Term
| Order Effects as a Confounding Variable |
|
Definition
Hypothetical Example: The treatment produces no significant change (only measurement error)- but the order effect adds 5 points
Two important points:
1. Order effect varies systematically with the IV
2. Data incorrectly appear to indicate a significant increase in scores |
|
|
Term
|
Definition
Controlling Time
Shorten time between conditions
Increase likelihood of order effects
Switching to between-subjects
When strong order effects are expected to exist
Counterbalancing
Changing the order in which treatment conditions are administered from one participant to another so that the treatment conditions are matched with respect from time |
|
|
Term
| Counterbalancing and Order Effects |
|
Definition
Evenly distributed order effects
Does not eliminate them! |
|
|
Term
| Limitations to Counterbalancing |
|
Definition
May distort treatment means
Variance
Changes within-treatment variance if order effects are present
Asymmetrical order effects
Order effects may not take place evenly
Number of treatments
Complete Counterbalancing- requires every possible permutation of treatment presentation
Partial Counterbalancing- uses enough different orderings to ensure each condition occurs first, then second |
|
|
Term
|
Definition
Simple and unbiased procedure
For 4 treatment conditions use a 4 X 4 matrix
Next row: move last letter in first line to beginning
Next row: ditto |
|
|
Term
|
Definition
Within-subjects designs are most preferred when:
1. The population has rare characteristics and are more difficult to recruit
Fewer subjects are required
2. The population is expected to exhibit a large amount of variability
Reduces or eliminates individual differences |
|
|
Term
| A note on individual differences |
|
Definition
While within-subject designs have fewer individual differences in comparison with between subject designs:
The design does not eliminate them
The design allows us to be able to measure and remove them with statistical analysis, but they do exist |
|
|
Term
| Two-Treatment Within-Subject Designs |
|
Definition
Advantages
Easy to conduct
Results are easily interpreted
Differences can be maximized
Easy to completely counterbalance
Disadvantages
Does not provide a complete picture of the relationships between variables
Two points of data
Analysis: Interval/ Ratio DV: Repeated measures t-test
Are treatment means significantly different? |
|
|
Term
| Multiple Treatment Designs |
|
Definition
Advantages
More likely to reveal functional relationship
More convincing cause-and effect claim
Disadvantages
Differences between groups may be too small to find significance
Increased likelihood of attrition
Difficult to completely counterbalance
Analysis: Interval/ Ratio DV: Repeated Measures ANOVA
Are any of the treatment means significantly difference from each other? If so, where? |
|
|
Term
| Within Subjects vs. Between Subjects Designs |
|
Definition
1. Individual differences
Problem: May become confounding variables or increase variance
Solution: Within-subjects design reduce this variance
2. Time-related factors and order effects
Problem: May distort results
Solution: Between-subjects designs only measure each individual once
3. Number of participants
Problem: Some populations are hard to recruit or costly
Solution: Within-subjects designs require fewer participants |
|
|
Term
|
Definition
Each individual in one group, is matched with a participant in each of the other groups, with respect to a variable considered to be relevant to the study
- Does not have identical (within) subjects, but equivalent
-No order or time effects
-Reduces some individual differences |
|
|
Term
| When Selecting Research Participants |
|
Definition
Samples may be drawn from the population using
Probability Sampling vs. non-probability sampling
Sampling must assure external validity to generalize to other populations
Determine the needed sample size
Power analysis (Ch 13- statistical analyses)
Larger samples provide more accurate estimates of population values |
|
|
Term
| Straightfoward Manipulation |
|
Definition
Most studies utilize straightforward manipulation, if any
Levels may be different types of stimuli or environment
Easy to interpret
Milgrams experiment: presence/ absence of authority figure |
|
|
Term
| Staged Manipulation ( A.K.A. Event Manipulation) |
|
Definition
Can get complicated
Use of a confederate or accomplice
Designed to get participants involved in “real” experience
May be difficult to replicate
Authority figure present and heart patient “learner” |
|
|
Term
| Manipulation Strength ( Example) |
|
Definition
Strong manipulation —> maximizes differences between groups
No treatments vs. full treatment
Study on Attitude Similarity and Liking
Do birds of a feather flock together?
IV: Similarity, Dv: liking
1,1= least similarity, least liking
10,10= most similarity, most liking
A strong manipulation would be comparing 1, 1, to 10,10 |
|
|
Term
|
Definition
Limited $$ equals
Limited use of equipment and supplies
Limited salaries for confederates
Limited incentives for participants
Limited space to conduct experiments
…..Which is why most use straightforward manipulation |
|
|
Term
| Self Report and Behavioral Measures |
|
Definition
Multiple measures of a variable provies a better picture than a single measure
Self Report
-Asking participants to report on themseleves
-Quesionnaire on activities of daily living
Behavioral
Direct observations of behaviors
Rate: how many times does it occur
Duration: how long does the behavior last?
Reaction Time: how quickly response occurs after stimulus |
|
|
Term
| Difference between Quasi Experiment and Experiment |
|
Definition
You are not manipulating!!! Ex. Gender!!!!
You do not have FULL manipulation of independent variable
Self Selected |
|
|
Term
|
Definition
Galvanic Skin Response (GSR)
Electrical conductance of the skin, which changes when sweating occurs
Measures general emotional arousal and anxiety
Electromyogram ( EMG)
Muscle tension
Measures tension or stress
Electroencephalogram (EEG)
Electrical activity of brain cells
Measures brain arousal in response to different stimuli
Function MRI (fMRI)
Measures activation of brain regions in response to stimuli |
|
|
Term
| Sensitivity of the Dependent Variable |
|
Definition
The DV must be sensitive enough to detect differences between groups
Ceiling Effect- The independent variable appears to have no effect on the dependent measure because participants all reach the maximum performance level
Floor Effect- The opposite of the ceiling effect; the task is too difficult and all participants perform poorly |
|
|
Term
|
Definition
Paper and pencil surveys- inexpensive
Cost of reproduction and writing implements
Video-taped interviews- expensive
Cost of equipment
Cost of multiple raters to review and rate observations
fMRI- very expensive
Training for administrator, interpreter, observer
Facilities
Equipment |
|
|
Term
| Controlling for Participant Expectations |
|
Definition
Demand Characteristics
Using unrelated filler items on a questionnaire or otherwise “clouding” the true intention of the study/measurement
Placebo Effects
Using a placebo group to assure external validity is maintained
Participants in the experimental group should show greater performance than those in the placebo group
If placebo participants performs the same as experimental participants= evidence of placebo effect |
|
|
Term
| Controlling for Experimenter Expectations |
|
Definition
Expectancy effects: Clever Hans
Was he really that clever
Well, yes- he used eye cues to start and stop tapping
Solutions to the expectancy problem
Single blind experiment- participant unaware of placebo
Double blind experiment- participant and experimenter both unaware of condition |
|
|
Term
| Additional Considerations |
|
Definition
Research Proposals
Feedback on the quality of your research
Pilot Studies
Trial run of your study with small number of participants
Manipulation Checks
Direct measurement of whether the manipulation had intended effects on participants
Mood measure for mood lighting
Debriefing
Discussion of ethical and educational implications of the study |
|
|
Term
| Analyzing and Interpreting Results |
|
Definition
Statistical analysis of the data you collected
Examine and interpret the pattern of results
Revisit you hypothesis and determine the relationships between the IV and DV |
|
|
Term
|
Definition
Professional Meetings
APA Annual Meeting
APS Annual Metting
Local conferences
Brown bag lunch meetings
Journal Articles
Peer review
Almost 90% rejection rate |
|
|
Term
|
Definition
-Employs questionnaires and interviews to ask people to provide information about themselves
Attitudes and beliefs
Demographics
Past or intended future behaviors
-Rests on an assumption that people are willing and able to provide truthful and accurate answers |
|
|
Term
| Why Conduct Survey Research? |
|
Definition
Provides methodology for asking people to tell about themselves
To study relationships between/among variables
To study how attitudes and behaviors change over time
Provides useful information for making public policy decisions
Important complement to experimental research findings |
|
|
Term
|
Definition
A tendency to respond to all questions from a particular perspective rather than to provide answers that are directly related to the questions
-Social desirability, or “faking good”
-Most acute when questions concern a sensitive topic
Violence to aggressive behavior
Illicit behavior
Sexual practices
However, not everyone misrepresents themselves |
|
|
Term
| Defining the Research Objectives |
|
Definition
The survey questions must be tied to the research questions that are being addressed
Attitudes and beliefs
Questions focus on the ways that people evaluate and think about issues
Facts and demographics
Factual questions ask people to indicate things they know about themselves and their situation
Behaviors
Questions can focus on past behaviors or intended future behaviors |
|
|
Term
| Question Wording- Ensure Simplicitiy |
|
Definition
The questions asked should be relatively simple
People should be able to easily understand and respond to the questions |
|
|
Term
|
Definition
Jargon and technical terms people won’t understand
Double-barreled questions that ask two things at once
Loaded questions include emotionally charge words and may influence response
Negative wording |
|
|
Term
| Question Wording- Potential problems that stem from difficutly understanding the question |
|
Definition
Unfamiliar technical terminology
Vague or imprecise terms
Ungrammatical sentence structure
Phrasing that overloads working memory
Embedding the question with misleading information |
|
|
Term
| Closed vs. Open Ended Questions |
|
Definition
What is your ethnicity? _________ vs. What is your ethnicity( Check one) __white __Black
With closed-ended questions, there are a fixed number of response alternatives |
|
|
Term
|
Definition
Rating scales ask people to provide “how much” judgments on any number of dimensions
Graphic Rating Scale ( Mild, Moderate, Severe)
Semantic Differential Scale (1-10)
Non-verbal Scales for Children (Smiley/ Frowny faces) |
|
|
Term
| Finalizing the Questionnaire |
|
Definition
Formatting
Should appear attractive and professional
Neatly typed and free from errors
Use point scales consistently
Refining Questions
Proof questions with others
Pilot test the survey with a small group |
|
|
Term
|
Definition
Personal administration to groups or individuals
Presented in written format and respondents write their answers
Inexpensive and anonymous
Mail Surveys
Relatively inexpensive to administer
Internet Surveys
Easy to design and administer
Other Technologies
Computerized experience sampling via Palm Pilots first, now iPhones |
|
|
Term
|
Definition
Face to face interviews
Time, facilities, transportation costs
Telephone interviews
Data collected relatively quickly and at less cost
Focus group interviews
Expensive, but yields good information
Problem- Interviewer Bias |
|
|
Term
| Survey Designs to Study Changes Over Time |
|
Definition
Questions are the same each time
Track changes over time
Panel Study |
|
|
Term
| Sampling From a Population |
|
Definition
Confidence Intervals
Level of confidence that the true population value lies within an interval of the obtained sample
Provides info about the likely amount of sampling error, or margin of error
Sample Size
A larger sample size reduces the size of the confidence interval
Must consider the cost/ benefit of increasing sample size |
|
|
Term
|
Definition
-Simple random sampling
Each member of the population has an equal probability of being selected
-Stratified random sampling
Population divided into subgroups and random samples taken from each strata
-Cluster sampling
Identify clusters and sample from the ree clusters |
|
|
Term
| Non Proability Sampling (Unknown probability of any member being chosen) |
|
Definition
-Haphazard Sampling
Convenience sampling- a take em where you find em approach for obtaining participants
-Purposive Sampling
Sample meets predetermined criterion
-Quota Sampling
Sample reflects the numerical composition of various subgroups in the population |
|
|
Term
|
Definition
The actual population of individuals from which the sample is drawn
Rarely will this perfectly coincide with the population of interest |
|
|
Term
|
Definition
Percentage of respondents who complete the survey
If 1,000 questionnaires were mailed out and 500 are completed and returned, the response r rate was 50% |
|
|
Term
|
Definition
You are conducting a survey about peoples beliefs about the relationship, if any, between family support and success in college
Group 1: Write 5 open-ended questions
Group 2: Write 5 closed- ended questions
1. Is there a relationship between family support and success in college
2. What factors are important influences on a persons choice of movie?
3. In a heterosexual couple, does the male or female usually determine the events of an evening out |
|
|
Term
| Increasing the # of Levels of an IV |
|
Definition
Provides more information about the relationship than a two-level design
Can provide us with a curvilinear relationship
Comparing two or more groups
How dogs, cats and birds have beneficial effects on nursing home residents
` As opposed to just cats and dogs |
|
|
Term
|
Definition
| Tests a claim about a population mean with a sample mean (0 unknown) |
|
|
Term
| Independent-measure t-statistic- |
|
Definition
| Tests a claim about the difference between two population means by using two samples and evaluating their mean difference |
|
|
Term
| Rogers, Kuiper & Kirker (1977) |
|
Definition
An experiment demonstrating the effect of levels of processing (superficial to deep) on memorization of events
All participants were given a surprise memory test
They were not told beforehand that they needed to memorize the words on the list |
|
|
Term
|
Definition
Tests claims about mean differences between two or more populations by using two or more samples
The tested hypothesis is very similar to the independent measures t statistic but now can be applied when we have more than two groups |
|
|
Term
| A Typical Example of a Situation Requiring ANOVA |
|
Definition
T-tests only allow the comparison of 2 means at a time
This would require 3 separate t-tests with 3 separate a levels, which accumulate over a series of tests
ANOVA allows the evaluation of all three means at once |
|
|
Term
|
Definition
Independent Variable : Telephone conditions :Three treatment conditions are created by the researcher
Quasi- Independent Variable: Age: Non0manipulated variable used to create groups |
|
|
Term
| Statistical Hypotheses for ANOVA- Null |
|
Definition
All populations have the same mean
Telephone condition has no effect on driving performance |
|
|
Term
| Statistical Hypotheses for ANOVA- Alternative |
|
Definition
H1: There is at least one mean difference among populations
At least one of the population means is different from another ( no specific decision which or how they differ)
Not all treatment conditions are the same, there is a treatment effect somewhere |
|
|
Term
| Why Variance vs. Mean Differences? |
|
Definition
When we have more than two sample means, how do you evaluate a mean difference?
Its problematic
However we can evaluate the variance between these three means |
|
|
Term
| Type 1 Errors and Multiple-Hypothesis Tests (Will be asked) |
|
Definition
Why not just use multiple t-tests?
Each time we conduct a t-test, we risk a Type 1 error
The level of risk for making a Type 1 error is set by our a -level
For an IV (telephone conditions) with three levels ( none, hands free, hand held) we would need three separate t- tests
Each test compounds the chance of making an experiment- wide Type 1 error
With ANOVA we maintain an a=.05 |
|
|
Term
|
Definition
These scores are all different. Or, in statistical terms, these scores are all variable
Our goal is to measure the amount of variability to explain why the scores are different
Three Steps
1. Calculate between- treatments variance
Here we see that there is a large variance between sample means
2. Calculate within-treatment variance
There is some variability in scores within each sample
3. Calculate the F-ratio to compare the two variances |
|
|
Term
|
Definition
Between Treatments Variance “Good” Variance
-This is the variance we are interested in
Differences between samples
-Possible sources of variance
Differences caused by treatment
Random Chance
-How do participants in different conditions differ?
Within-Treatments Variance “Bad” Variance
-Variance due to chance and sampling error
“noise”
-How big are the differences between individuals when there is no treatment effect?
-How do the participants within the same condition differ? |
|
|
Term
| The F-ratio ANOVAS Test Statistic |
|
Definition
F= Variance between treatments/ variance within treatments
Treatment effect + differenced due to chance/ Differences due to chance
The denominator is our error term ( variance expected due to chance)
When the treatment effect is zero ( we cannot reject the null), the F-ratio approximates 1 |
|
|
Term
| Research Designs For ANOVA |
|
Definition
Independent -measures design
Uses a separate group of participants for each of the treatment conditions being compared
Repeated-measures design
Uses one group of participants for all of the treatment conditions
Two-factor ( or Factorial) design
2 or more IV’s |
|
|
Term
| Research Designs with Two-Factors |
|
Definition
Two IV’s, each their own levels
Creates a research design, which allows us to look at every possible condition
Summarized in a matrix, where we would require a separate sample for each cell
Factor A : Self Esteem - Low or High
Factor B : Audience Condition - Audience or No Audience
2 X 2 Anova |
|
|
Term
| A Two Factor Research Study |
|
Definition
For example, a researcher studying the effects of heat and humidity on performance could use the following experimental design
Factor A Humidity- Low/ High
Factor B: Temperature - 80,90,100
2 X 3 Anova |
|
|
Term
| What does the 2 factor ANOVA do? |
|
Definition
Evaluates three separate sets of mean differences
1. Mean difference between the two humidity levels
2. Mean differences between the three temperature levels
3. Any other mean differences that may result from unique combinations of a specific humidity level and a specific temperature level |
|
|
Term
| Interpretation of Factorial Designs |
|
Definition
Main Effects
The effect of each independent variable taken by itself
There is a “main effect” for each variable in your design
Interactions
Effect of one IV depends on the particular level of the other IV
There is only one interaction in a two-factor design |
|
|
Term
|
Definition
1. Main effect for Factor A (Gender: Quasi-IV)
2. Main effect for Factor B (Drug, IV)
3. Interaction between Factors A and B (The effect of the drug depends on gender)
-Main Effects- The mean differences among levels of one factor
-Interactions- When the mean differences between individual treatment conditions are different from what would be predicted from the overall main effects of the factors |
|
|
Term
| What Means Are We Evaluating? |
|
Definition
Main Effect for Factor A ( Gender)
Compares the means for males (60) and for females (80)
Main Effect for Factor B (Audience)
Compares the means for Drug A (70) and Drug B (70)
Interaction (Self-esteem x Audience)
Evaluates how the means for the levels for gender (Factor A) are influenced by drug (Factor B) |
|
|
Term
| Why are Interactions Important? |
|
Definition
If the DV is depression scores, consider what conclusions you might find considering main effects alone
Females have higher depression scores than men
The drug does not have any effect
However, if you consider the interaction between these two factors
The drug seems to be having a different effect for males than it does for females |
|
|
Term
|
Definition
F= Variance ( mean differences) not explained by main effects/ Variance (differences) expected by chance/error
-If the two factors are independent ( do not influence the effect of each other) there will be no interaction |
|
|
Term
| Graphical Illustration of an Interaction |
|
Definition
| When there is an interaction, plotting the means of the levels from the factors produces nonparallel lines |
|
|
Term
| Interactions and Moderator Variables |
|
Definition
A moderator variable influences the relationship between the two other variables
Influence of carrot consumption on the relationship between age and blood pressure |
|
|
Term
|
Definition
Research Hypothesis- Misleading questions results in more errors in eyewitness testimony than do unbiased questions
Maybe the type of questioner (the moderator variable) influences this relationship |
|
|
Term
| Why We Love the two-factor ANOVA |
|
Definition
Allows us to examine three types of mean differences within one analysis
Three hypotheses, each with its separate F- ratio, which we are familiar with |
|
|
Term
| Independent Groups Design |
|
Definition
Different group of participants assigned to each of the different conditions
For a 2 X 3 design, we need 6 groups |
|
|
Term
|
Definition
The same individuals will participate in all conditions
For this 2 X 3 design, we would need one more group |
|
|
Term
|
Definition
Use both repeated measures groups and independent groups
For this example we would need 2 groups |
|
|
Term
| Increasing the # of our IV’s or Levels of IV’s |
|
Definition
2 X 2= Simplest design
2 IVs with 2 levels in each IV
We can increase the levels of an IV
2 X 3= Slightly more complex
2 IVs with 2 levels in one IV, 3 levels in the other IV
We can increase the number of IVs
2 X 2 X 2= Even more complex
3 IV’s with 2 levels in each IV
2 X 2 X 3=? |
|
|
Term
| Measuring Effect Size for ANOVA |
|
Definition
A significant difference indicates our difference is larger than expected by chance, but does not tell us how large this difference is
N2= The percentage of variance accounted for
N2= SSbetween/ SStotal |
|
|
Term
| What if I have a significant F? |
|
Definition
Reject the null hypothesis. There is a significant difference somewhere between at least two of thee groups (not all groups are equal)
But I don’t know where the difference is
If i have 3 groups: Group 1 could be different than B or Group B from C, Or Group A from C
So if there is a significant F-ratio, more tests must be done to in which groups are different |
|
|
Term
|
Definition
Additional hypothesis stets that are done after an ANOVA results in a significant F to determine exactly which mean differences are significant and which are not
-Finds where the significant differences are
-Developed to control the experiment-wise error rate |
|
|
Term
|
Definition
Tells us the minimum difference between treatment means that is necessary for significance
q: “Studentize range statistic” found in Table B.5
Must know k, df within treatments and set a
n: number of scores in each treatment (requires equal sample sizes in all groups) |
|
|
Term
|
Definition
More conservative than Turkeys HSD
Uses the ANOVA formula, but only compares two groups at a time |
|
|
Term
| Single-Case Experimental Design- (A.K.A. Single-subject Design) |
|
Definition
Use results from a single participant to establish the existence of cause and effect relationships
-The impact of an experimental manipulation on one participant
1. Behavior is measured over time during a baseline control period
2. Manipulation is introduced during treatment period
3. Change in behavior between baseline and treatment assessed
-Useful for applied research
Clinical, counseling, educational |
|
|
Term
|
Definition
Phase= A series of observations made under the same conditions
Minimum of 3 observations
Baseline phase (A)
A series of baseline observations (no treatment is being administered)
Treatment phase (B)
A series of treatment observations ( a treatment is being administered) |
|
|
Term
| Evaluating Graphs (Evaluating phase patterns) |
|
Definition
Level- Magnitude of participants responses
Trend- Increase/decrease in magnitude of behavior
Stability- The degree to which the level or trend is consistent throughout the phase |
|
|
Term
| How to deal with unstable data |
|
Definition
1. Wait- the pattern may stabilize
2. Average sets of observations
3. Look for patterns within the inconsistencies |
|
|
Term
|
Definition
Once a pattern is established
Change the conditions (usually by administering or stopping a treatment)
Expected to produce a noticeable change in behavior
Unique considerations:
Does the baseline show a trend?
Does the baseline indicate dangerous behavior?
Does implementing the treatment severely change behavior? |
|
|
Term
| How to evaluate a change in pattern |
|
Definition
Change in average level
Immediate change in level
Change in trend
Latency of change |
|
|
Term
| Reversal Designs: The ABA Design |
|
Definition
Baseline (A)-> Treatment phase (B) -> Baseline phase (A)
Each B phase pattern is clearly different from each A phase pattern
A method to demonstrate the reversibility of the manipulation
A.K.A Withdrawal Design
Example- The use of parse as a treatment to measure the improvement of a child’s school performance
Measure test scores (A)
Give regimen of praise for correct homework problems (B)
Measure test scores (A) |
|
|
Term
| Reversal Designs: The ABAB Design |
|
Definition
Baseline (A)-> Treatment (B)-> baseline (A)-> treatment (B)
Changes from phase A->B and B-> A are consistent
An extension of the ABA design
Eliminates the possibility that change is due to chance function
Addresses ethical issue of ending a design with withdrawal of treatment |
|
|
Term
| Multiple Baseline Designs |
|
Definition
Change is observed under multiple circumstances
The manipulation is introduced at different times to determine that the manipulation caused change
-Two simultaneous baseline phases (A)-> Treatment phases (B) are implemented at different times for each baseline
-Only 1 phase change
1. Multiple baseline across subjects
Initial baseline phases correspond to two separate participants
2. Multiple baseline across behaviors
Initial baseline phases correspond to separate behaviors for the same participant
3. Multiple baseline across situations
Initial baseline phase correspond to two separate situations for the same participant |
|
|
Term
| Why use multiple baseline designs? |
|
Definition
B-> change is unethical
Shows behavior accompanies manipulation of the treatment
Replication of behavior change when implementing treatment |
|
|
Term
| Differences from Group Designs |
|
Definition
Conducted with only one participant
More flexible
Can be modified or changed
No standardization necessary
Requires continuous assessment |
|
|
Term
|
Definition
Strengths
Cause and effect relationships from one person
Not in case studies or quasi experimental designs
More generalizable to reality
Applied research
Extremely flexible
Weaknesses
Relationship only demonstrated for one person
Less generalizable to other individuals
Assessment procedures may influence behavior
Lack of statistical controls |
|
|
Term
|
Definition
Evaluation of programs implemented to achieve some positive effect on a group of individuals
Need Assessment-> Program Theory Assessment-> Process Evaluation -> Outcome Evaluation-> Efficiency Assessment |
|
|
Term
| Quasi Experimental Developmental Designs |
|
Definition
| Addresses the need to study the effect of an IV in settings in which the control cannot be established |
|
|
Term
| Non- and Quasi-experimental Designs |
|
Definition
Fail to meet at least one requirement of a true experimental design (ambiguous cause and effect)
Non experimental- Little or no attempt to minimize threats to internal validity
Quasi experimental- Some attempt to minimize threats to internal validity
-Both compare groups (levels of IV) that are not manipulated
Between-subjects: defined by a pre-existing participant variable
Within subjects: defined in terms of time |
|
|
Term
| One-Group Posttest-Only Design |
|
Definition
A.K.A. “One- shot case study”
Lacks a crucial element of a true experiment
A control group
We may have some implicit idea of how a control group would perform, but there is no way of confirming this |
|
|
Term
| Nonequivalent Control Group Design |
|
Definition
V does not permit control over assignment to groups
Schools with different programs, participants with different jobs, gender, warm vs. cold climates
Random assignment or matching is not possible
Groups are considered non equivalent |
|
|
Term
| One-Group Pretest-Posttest Design |
|
Definition
Comparison of measures before the manipulation ( a pretest) and again afterward ( a posttest)
Index of change is then computed to determine effect
Threats to internal validity
History, maturation, testing, instrument decay, regression toward the mean |
|
|
Term
| Nonequivalent Control Group Pretest-Posttest Design |
|
Definition
Uses preexisting groups: both are measured twice, only one receives the treatment
Reduces the threat of assignment bias
Reduces the threat of time-related factors
Quasi-experimental |
|
|
Term
| Interrupted Time Series Design |
|
Definition
Examines the dependent variable over an extended period of time, both before and after the IV is implemented
Interpretation problems: Possible regression to the mean |
|
|
Term
|
Definition
Improves interrupted time series design by finding an appropriate” control group”
Involves finding a similar population that did not receive a particular manipulation
Limited because this is not a true “control group” |
|
|
Term
|
Definition
Persons of different age groups measured at the same point in time
Cohort: A group of individuals of the same age at the same point in history
Advantages
Relatively quick
Relatively inexpensive
Gives insight into normative developmental change
Disadvantages
Cohort Effects: Experiences that happen at a certain point in time can cause an age group at one point in history to differ from the same age group at a different point in history
Does not provide information about individual differences |
|
|
Term
|
Definition
Some group (one cohort) is followed over time
Advantages
Limits unimportant variability
Gives insight into the role of the individual in behavior
Give insight into the cumulative effect of experiences over time
Gives insight into normative developmental change
Disadvantages
Time consuming
Expensive
Selective Attrition: You may lose more individuals of a certain type
Cannot assess cohort effects |
|
|
Term
|
Definition
Multiple cohorts are studied over time
Combines the advantages of both cross sectional and longitudinal designs
Gives insight into the role of the individual in behavior
Genes express themselves more actively with time
Gives insight into the cumulative effect of experience over time
Experiences in childhood can influence behaviors later in life more dramatically than they influence immediate behaviors
Gives insight into normative developmental change
Get initial data quickly and cheaply
Can analyze data to look for selective attrition |
|
|
