Lesson 11 (slides 1-21) Flashcards
What is the plan called that establishes data collection procedures and hypothosis testing?
Evaluation design AKA research design
What provides structure for the major components of the study?
Evaluation design
- When, from whom, how, and what program procedures and measurements will be applied
Overall, what is evaluation design attempting to address?
CAUSALITY
- Did the intervention/program cause the effects?
What are the three design characteristics for improving causality?
- Comparison/control group similar to the experimental/exposed group
- Measurement of outcome variables before and after the intervention
- Minimal threats to internal and external validity
EVALUATION DESIGN NOTATION
R
Random assignment of an evaluation study participant (or unit) to a group
EVALUATION DESIGN NOTATION
E
Experimental (intervention or treatment) group
What does E1, E2, E3, …..En indicate?
Indicates planned exposure of the group to different intervention procedures.
EVALUATION DESIGN NOTATION
C
Control (equivalent) group established only by random assignment
- Indicates NO exposure to an intervention or exposure to a minimum or standard intervention procedures
EVALUATION DESIGN NOTATION
C
Comparison group established through any other method other than randomization
EVALUATION DESIGN NOTATION
X
Intervention procedures applied to an E (experimental) group
What does X1, X2, X3, …..Xn indicate?
Indicates an intervention consisting of multiple, different approaches
EVALUATION DESIGN NOTATION
O
An observation or measurement to collect data.
Including:
- Tests
- Interviews
- Visual or audio ratings, or
- Record interviews
What does O1, O2, O3,….. indicate?
Indicates multiple measurements at different times
EVALUATION DESIGN NOTATION
T
Time when an observation, assignment to a group, or application of intervention procedures has occured.
What does T1, T2,T3,…. Tn indicate?
Indicates repeated observations
EVALUATION DESIGN NOTATION
n
Number of participants (sample)
- Could be in the E, C, or C group
CATEGORIES OF EVALUATION DESIGN
Is random assignment used?
- YES
Experimental design
Is random assignment used?
- NO
Is there a comparison group or multiple observations?
YES?
NO?
YES
- Quasi-experiemental design
NO
- Non-experimental design
CATEGORIES OF EVALUATION DESIGN
Includes one experimental group (E) but no
control group (C) or comparison group (C)
Non-experiemental design
What are the 3 research designs?
Which one is the weakest and why?
The three research designs:
- Non-experimental design
- Weakest in regards to internal and external validity
- Quasi-experiemental design
- Experiemental design
CATEGORIES OF EVALUATION DESIGN
- Includes an experiemental group (E) and a comparison group (C) created by methods other than random assignment
- Includes observations of both groups, both prior to and after application of intervention (X) procedures
- Multiple waves of data collection
- Does not account for all biases of internal validity
Quasi-Experimental Design
CATEGORIES OF EVALUATION DESIGN
- Includes random assignment of experimental (E) group and a control (C) group
- Observations of both groups prior to and after application of the intervention (X) procedures
- Yields the most interpretable and defensible results of evidence of behavioral impact
Experimental Design
What is the ‘Gold Standard’ in evaluation design and why?
Experimental design
- It yields the most interpretable and defensible evidence of behavioral impact
- Asserts the greatest degree of control over biasing factors that compromise internal validity of results
What are the most commonly used evaluation (and research) designs?
- One-shot case study
- Retrospective Pretest
- One-group pretest-posttest design
- Time series design
- Pretest-posttest control-group design
- Case study design
Which design is this?
X——–O
(hint: How are paticipants faring after the services?)
One-shot case study
- (non-experimental)*
- Survey of program participants after participation in the program
What is the absolute most simplest form of outcome evaluation?
WebRep
currentVote
noRating
noWeight
One-shot Design
Ex: How many people who attended a smoking cessation program are not smoking a month later?
- How participants are faring after the services.
WebRep
currentVote
noRating
noWeight
What are the primary weaknesses of one-shot design?
Does not demonstrate change during or as a result of the program
- No comparison
- No multiple waves of data
When is it best to use a one-shot design?
Best if used to answer descriptive research questions
Why would it be difficult to establish observations are due to a disaster with a one-shot design?
Because there would be no comparison and no multiple waves of data collection
Which design is this?
X——–O
***respondents are asked to recall information before and after the intervention
Retrospective Pretest
- Survey of program participants after participation in the program
What are the primary weaknesses of the retrospective pretest?
Recall Bias
- Relies on memory
Which design is this?
O1——-X——–O2
One Group Pretest-Posttest Design
- Evaluator collects data before and after the program is conducted
What are the primary weaknesses of the one group prestest-posttest design?
- Potential selection bias
- Attrition (drop-out)
- Potential measurment bias
- Poor validity & reliability
What needs to be considered with internal validity?
Is the intervention the only (or most) plausible explanation for the changes?
What is internal validity?
Degree to which measured change can be attributed to the program
What is external validity?
Extent to which the program can produce similar effects in other populations
- Is the program generalizable?
What are the two types of threats to internal validity?
Two primary sources of bias:
- Non-program related sources of bias
- Program related sources of bias
What are the two sources of non-program related bias?
- Maturation
- Historical bias
What are the three sources of program related bias?
- Attrition
- Selection bias
- Measurement bias
Which sources of bias are uncontrollable and which are somewhat controllable?
Uncontrollable: non-program related
Controllable: program related