1
Q
Why Psychologists Conduct Experiments
A
- Test
- Hypotheses derived from theories
- Effectiveness of treatments and programs
- Third goal of psychological research
- Explanation
- -Examine the causes of behavior
- Multimethod approach
- Seek convergent validity for research findings across methods
2
Q
Experimental Research
A
An experiment must include
- Independent variable (IV)
- Dependent variable (DV)
3
Q
Independent Variable
A
- Manipulated (controlled) by experimenter
- At least 2 conditions (levels)
- “Treatment” and “Control”
4
Q
Dependent Variable
A
- Measured by experimenter
- Used to determine effect of IV
- Typically researchers measure several dependent variables to assess effect of IV.
5
Q
Internal Validity
A
Differences in performance (DV) can be attributed unambiguously to effect of independent variable (IV)
6
Q
3 Conditions for Causal Inference
A
- Covariation
- Time-order relationship
- Eliminate alternative causal explanations (confoundings)
7
Q
Confoundings
A
- IV covaries with a different, potential independent variable
- Alternative explanations for a study’s findings
- Experiment free of confoundings has internal validity
8
Q
Control Techniques
A
- Eliminate confoundings
- Hold conditions constant, balancing
- Holding conditions constant
- IV is only factor that differs systematically across groups
9
Q
Manipulation
A
- IV: participants in the conditions have different experiences
- Example: Barbie, Emme, or neutral images
10
Q
Balancing
A
- Random assignment to conditions balances subject characteristics, on average.
- Groups are equivalent prior to IV manipulation.
- All subject variables are balanced.
11
Q
Independent Groups Designs
A
Different individuals participate in each condition of the experiment.
-No overlap of participants across conditions
12
Q
Random Groups Designs
A
*Individuals are randomly assigned to conditions of the IV.
*Logic of causal inference
-If groups are equivalent at the beginning of an experiment (through balancing) and conditions are held constant,
any differences among groups on dependent variable are caused by the manipulated independent variable.
13
Q
Block Randomization
A
- Block: random order of all conditions in the experiment
- Randomly assign subjects one block at a time.
- Advantages
- Creates groups of equal size
- Controls for time-related events that occur during course of experiment
14
Q
Ability to make causal inferences is threatened when
A
- Intact groups of subjects are used
- Extraneous variables are not controlled
- Hold conditions constant
- Selective subject loss occurs
- Mechanical subject loss not a problem
- Demand characteristics and experimenter effects are not controlled
- Use placebo-control and double-blind procedures
15
Q
Use statistical analysis
A
- Claim IV produced an effect on DV
* Rule out the alternative explanation that chance produced any observed effect
16
Q
Replication
A
- Best way to determine whether findings are reliable
* Repeat experiment and see if same results are obtained
17
Q
Three Steps-Analysis of Experimental Designs
A
- Check the data
- Errors? outliers?
- Describe the results
- Descriptive statistics such as means, standard deviations, effect size
- Confirm what the data reveal
- Inferential statistics
18
Q
Internal validity
A
Degree to which differences in performance on a dependent variable can be attributed clearly and unambiguously to an effect of an independent variable, as opposed to some other uncontrolled variable.
-Threats to internal validity
19
Q
Intact groups
A
Are formed prior to the start of the experiment
20
Q
Mean
A
- Central Tendency
- Average score on DV, computed for each condition
- Not interested in each individual score, but how people responded on average in a condition
21
Q
Standard Deviation (Variability)
A
- Average distance of each score from the mean of a group
* Not everyone responds the same way to an experimental condition
22
Q
Effect size
A
- Measure of strength of relationship between the IV and DV
- Cohen’s d
- Difference between treatment and control means/average variability for all participants’ scores
23
Q
Guidelines for Interpreting Cohen’s d
A
- small effect of IV: d = .20
- medium effect of IV: d = .50
- large effect of IV: d = .80
24
Q
Meta-analysis
A
- Summarize effect sizes across many experiments that investigate same IV or DV.
- Choose experiments based on their internal validity and other criteria.
- Allows researchers to gain confidence in general psychological principles
25
Confirm what the data reveal
* Use inferential statistics to determine whether the IV produced a reliable effect on the DV.
* Rule out whether findings are due to chance (error variation).
* Two types of inferential statistics
- Null Hypothesis Significance Testing
- Confidence intervals
26
Null Hypothesis SIgnificance Testing
* Statistical procedure to determine whether mean difference between conditions is greater than what might be expected due to chance (error variation)
* Effect of an IV on the DV is statistically significant when the probability of the observed results being due to chance is low.
* p < .05
27
Step 1 for Null Hypothesis Testing
(1) Assume the null hypothesis is true.
- The population means for groups in the experiment are equal.
- Example
- -The population mean for body dissatisfaction after viewing Barbie images is equal to the population means for Emme images and neutral images.
28
Step 2 for Null Hypothesis Testing
(2) Use sample means to estimate population means.
- Example
- mean body dissatisfaction for Barbie = -.76
- mean body dissatisfaction for Emme = 0.00
- mean body dissatisfaction for neutral = 0.00
- difference between Barbie and Emme/neutral = -.76
* Is the observed mean difference (-.76) greater than what is expected when we assume the null hypothesis is true (zero difference)?
29
Step 3 for Null Hypothesis Testing
* Compute the appropriate inferential statistic.
- t-test: test the difference between two sample means
- F-test (ANOVA): test the difference among three or more sample means
30
Step 4 for Null Hypothesis Testing
* Identify the probability associated with the inferential statistic
- p value is printed in computer output or can be found in statistical tables
31
Step 5 for Null Hypothesis Testing
* Compare the observed probability with the predetermined level of significance (alpha), which is usually p < .05
- If the observed p value is greater than .05, do not reject the null hypothesis of no difference
- Conclude IV did not produce a reliable effect
32
Confidence Intervals
* Sample means estimate population means
* Confidence interval for a mean provides the range of values that contains the true population mean.
- with some probability, usually .95
33
Compute confidence interval around sample mean in each condition.
* If confidence intervals do not overlap, we gain confidence that the population means for the conditions are different--that is, the IV has an effect.
* If confidence intervals overlap slightly, we are uncertain about the true mean difference.
* If intervals overlap such that the mean of one group lies within interval of another group, we conclude the population means do not differ
34
External Validity
* Extent to which findings from an experiment can be generalized to describe individuals, settings, and conditions beyond the scope of a specific experiment.
- Any single experiment has limited external validity
- External validity of findings increases when findings are replicated in a new experiment
35
Questions of external validity
* Would the same findings occur
- In different settings?
- In different conditions?
- With different participants?
* Research with college students is often criticized because of low external validity.
* Theory testing: Sample often doesn’t matter
36
Increasing External Validity
* Include characteristics of situations, settings, and population to which researchers seek to generalize
* Partial replications
* Field experiments
* Conceptual replications
37
Mechanical subject loss
* Occurs when a subject fails to complete the experiment because of an equipment failure
* Can occur if a computer crashes, or if the experimenter reads the wrong set of instructions, or if someone inadvertently interrupts an experimental session
* Less critical problem than selective subject loss because the loss is unlikely to be related to any characteristic of the subject
* Should not lead to systematic differences between the characteristics of the subjects who successfully complete the experiment in the different conditions of the experiment
* The result of chance events that should occur equally across groups
* It should be documented when it occurs
38
Selective subject loss
* When subjects are lost differentially across the conditions of the experiment
* When some characteristic of the subject is responsible for the loss
* When this subject characteristic is related to the dependent variable used to assess the outcome of the study
* Destroys the comparable groups that are essential to the logic of the random groups design and can thus render the experiment uninterpretable
39
Demand Characteristics
* Cues and other information that participants use to guide their behavior in a psychological study
* May behave consistent with these expectations rather than in response to the effects of the alcohol per se
40
Experimenter effects
Potential biases due to the expectations of the experimenters
*Source of confounding if experimenters treat subjects differently in the different groups of the experiment in ways other than those required to implement the independent variable.
41
Placebo Control Group
* Control demand characteristics
* Placebo: substance that looks like a drug or other active substance but is actually an inert, or inactive, substance.
* Any differences between the experimental groups and the placebo control group could legitimately be attributed to the actual effect of the drug taken by the experimental participants, and not the participants' expectations about receiving a drug.
42
Double-Blind Procedure
Both the participant and the observer are blind to (unaware of) what treatment is being administered.
43
Matched groups design
* Good alternative when neither the random groups design nor the repeated measures design can be used effectively.
* Instead of trusting random assignment to form comparable groups, the research makes the groups equivalent by matching subects
* Once comparable groups have been formed based on the matching, the logic of the matched groups design is the same as that for the random groups design.
44
Individual differences variables/Subject variables
Individual differences variable is a characteristic or trait that varies across individuals
45
Natural Groups Designs
Experiments involving independent variables whose levels are selected- like individual differences variables
*Is frequently used in situations in which ethical and practical constraints prevent us from directly manipulating independent variables.
