Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
32 Cards in this Set
- Front
- Back
|
Applied Research
|
can solve real problems and contribute knowledge to the field of pyschology
|
|
|
4 Design Problems in Applied Research
|
Ethical Dilemmas
Reduction of Internal Validity and Increase in External Problems Unique to B/N Subjects Designs: nonequivalent groups > reduced internal validity > maturation or history > matching can lead to regression problems Problems Unique to W/N Subjects Designs: unable to counterbalance properly > uncontrolled sequence effects and/or attrition |
|
|
Conclusions that can be drawn from:
Manipulated Variables Subject Variables |
Can draw conclusions of causality.
Since all cannot be held constant, we can not eliminate alternative explanations: cannot determine causality. |
|
|
"True" Experiments include:
|
manipulated dependent variables
equivalent groups for b/n subjs or counterbalancing for w/n subjs |
|
|
Quasi-Experiments
|
when there is incomplete control over the variables in a study and casual conclusions cannot be drawn
can be used when random assignment is impossible |
|
|
Quasi-Experimental Designs
|
Single-factor nonequivalent designs, with two or more levels
Nonequivalent groups factorial design PxE factorial designs Correlational Research* Simulated Pretest-Posttest Control Group Nonequivalent Control Group with Pretest and Posttest Interrupted Time Series Design Nonequivalent Control Group Design with Posttest Only Multiple Time Series Design |
|
|
Simulated Pretest-Postest Control Group
|
Does military service improve self-esteem?
One pool of participants All participants are already being/going to be exposed to Treatment Obtain Subjects who are already signed up for T. (military service) Half are given pretest only Other half are given posttest only ^ ^ ^ made acceptable with random assignment & rules out testing confounds attrition is possible Choose to Use When You Have No Control Over Treatment |
|
|
Nonequivalent Control Group Design with Pretest and Posttest
|
8th Grade "Say No to Smoking" Campaign
Two different groups 1 Target and 1 Comparison Comparison group matches Target but does not receive Treatment Remember to describe how to measure (e.g. survey) Compare change between tests and groups |
|
|
Interrupted Time Series Design
|
Did the institution of capital punishment in 2000 deter violent crime in that state?
Pretest and Posttest observations taken Use Archival Data Measure number of incidences before and after |
|
|
Nonequivalent Control Group Design with Posttest Only
|
Choose to use when pretest is impossible
Used to study the impact of an event Need a matched comparison group Obtain data from archives |
|
|
Multiple Time Series Design
|
Target group and comparison group
Observations taken before and after |
|
|
Regression and Matching
|
An attempt to reduce inequivalency through matching can inhance the possibility of regression effects.
|
|
|
The main advantage of Interrupted Time Series Design
|
Trends: predictable patterns.d
|
|
|
Content Analysis
|
any systematic examination of qualitative information in terms of predefined categories
|
|
|
Program Evaluation
|
Includes:
procedures for determining a true need is the program running as planned methods for evaluating program outcomes cost analysis |
|
|
Needs Analysis
|
Does a population of sufficient size exists that would benefit from the proposed program?
|
|
|
Identifying Need
|
Census Data
Surveys of available resources Surveys of pontential users Key informants, focus groups, and community forums |
|
|
Formative Evaluation
|
monitoring the progress of a program while its in progress
|
|
|
Program Audit
|
how the program is being used
if the program is being implemented as planned how the program is being used (program improvement) |
|
|
Summative Evaluations
|
overall assessments of program effectiveness
|
|
|
Cost-Effectiveness Analysis
|
monitoring the actual costs of a program and relating those costs to the effectiveness of the program's outcome
|
|
|
Reasons for Small N Designs
|
Large N studies can fail to have individual-subject validity.
Potential participants are rare or hard to find. Skinner's argument: should be inductive from specific cases to general laws of behavior. |
|
|
Operant Conditioning
|
process in which the frequency of behavior is modified by consequences of behavior
|
|
|
Rate of Response
|
favored dependent variable of researchers working in the Skinnerian tradition; how frequently a behavior occurs per unit of time
|
|
|
Cumulative Recorder
|
apparatus for recording the subject's cumulative rate of response in operant conditioning studies
|
|
|
Elements of Single-Subject Designs
|
Since there is no control group, behavior must be shown as a result of treatment.
Target behavior must be operationally defined. Establishment of a baseline. Begin treatment and continue monitoring behavior. |
|
|
ABA Design
(heart of Single Subject Designs) |
Establish a steady baseline
Incorporate Treatment Withdraw Treatment |
|
|
ABAB Design
Weaknesses |
Same as ABA, but treatment is readministered.
Subject can become sensitized Experimenter bias Sequencing effects |
|
|
Multiple Baseline Design
|
For use when treatment is powerful enough to have lasting effects.
Involves either different: subjects, settings, or behaviors. |
|
|
Changing Criterion Design
|
Baseline and several criterions (e.g. increase of minutes of exercise)
|
|
|
Alternating Treatments Design
|
Baseline and two treatments (B and C) are repeatedly administered in random order.
|
|
|
Case Study
|
a detailed description and analysis of a single individual or individual event.
|
