Week 1: First Quasi-Experimental Alternative
➔ Public health interventions are complex interventions:
➔ Complex interventions (Skivington et al, 2021):
- An intervention might be considered complex because of properties of the intervention
itself, such as:
- The number of components involved
- The range of behaviors targeted
- Expertise and skills required by those delivering and receiving the intervention
- The number of groups, settings, or levels targeted,
- Or the permitted level of flexibility of the intervention or its components
- Such interventions are delivered and evaluated at different levels, from individual to
societal levels (or system levels)
➔ Framework for developing and evaluating complex interventions:
➔ Hierarchy of Evidence:
,➔ Describing Complex Interventions: things that we need to know about the intervention:
1. Look at the Intervention aim:
The Programme theory:
- Describes how an intervention is expected to lead to its effects and under what
conditions
- Should be used to guide the identification of research questions
2. Intervention Logic Model:
- Diagrammatic representations of intervention theory of change and of logical
relationships.
3. Intervention mechanism of action (or effect)
- Assess the validity of the theory of change: Such analysis can explain why an
intervention is found to be effective or ineffective within an outcome evaluation
(Detels et al, 2021).
- Quantitative: Mediator analysis:
- To assess whether proximal outcomes appear to fully or partially account for
distal outcomes
- Proximal outcomes = intermediate outcome = mediator (NOT for this course)
- Qualitative: how pathways are perceived and described by those involved
4. Setting:
- Where the intervention is implemented
- Example: Community, school, worksite
5. Interventions Components: what is actually implemented such as:
- Intervention methods and strategies such as materials, activities, policies and
such. Example:guided yoga sessions, guided workout sessions
6. Deliverers or implementers:
- Those persons who deliver or implement the intervention components
➔ Types of designs that should match specific research questions:
- True-. Quasi-, and pre-experimental designs
- True experiment =RCT (randomization, control, pre and post test)
- Quasi-experiment = no randomization
- Pre-Experiment = no randomization, no control and/or no pre-test
- OXO-notation:
- R= Random assignment (to experimental or control condition)
, - O= observation or measurement (for example pre-test, post-test, follow- up)
there are 2 O’s, so one for before test and one for after test.
- X= exposure to an experimental variable or event (= the manipulation by the
researcher, the intervention)
- Randomization
- Threats to validity
➔ True Experimental designs
- Look at OXO notation
- If the sample size is big enough there is no need for a pre-test
- Combining the first 2 designs, interesting as it can tell us whether doing a pre-test can
have an outcome or not
➔ Quasi-Experimental designs
1. No randomisation - we can not be sure that the control group is equivalent
2. One group but several pre–tests and post-tests
➔ Public health interventions are complex interventions:
➔ Complex interventions (Skivington et al, 2021):
- An intervention might be considered complex because of properties of the intervention
itself, such as:
- The number of components involved
- The range of behaviors targeted
- Expertise and skills required by those delivering and receiving the intervention
- The number of groups, settings, or levels targeted,
- Or the permitted level of flexibility of the intervention or its components
- Such interventions are delivered and evaluated at different levels, from individual to
societal levels (or system levels)
➔ Framework for developing and evaluating complex interventions:
➔ Hierarchy of Evidence:
,➔ Describing Complex Interventions: things that we need to know about the intervention:
1. Look at the Intervention aim:
The Programme theory:
- Describes how an intervention is expected to lead to its effects and under what
conditions
- Should be used to guide the identification of research questions
2. Intervention Logic Model:
- Diagrammatic representations of intervention theory of change and of logical
relationships.
3. Intervention mechanism of action (or effect)
- Assess the validity of the theory of change: Such analysis can explain why an
intervention is found to be effective or ineffective within an outcome evaluation
(Detels et al, 2021).
- Quantitative: Mediator analysis:
- To assess whether proximal outcomes appear to fully or partially account for
distal outcomes
- Proximal outcomes = intermediate outcome = mediator (NOT for this course)
- Qualitative: how pathways are perceived and described by those involved
4. Setting:
- Where the intervention is implemented
- Example: Community, school, worksite
5. Interventions Components: what is actually implemented such as:
- Intervention methods and strategies such as materials, activities, policies and
such. Example:guided yoga sessions, guided workout sessions
6. Deliverers or implementers:
- Those persons who deliver or implement the intervention components
➔ Types of designs that should match specific research questions:
- True-. Quasi-, and pre-experimental designs
- True experiment =RCT (randomization, control, pre and post test)
- Quasi-experiment = no randomization
- Pre-Experiment = no randomization, no control and/or no pre-test
- OXO-notation:
- R= Random assignment (to experimental or control condition)
, - O= observation or measurement (for example pre-test, post-test, follow- up)
there are 2 O’s, so one for before test and one for after test.
- X= exposure to an experimental variable or event (= the manipulation by the
researcher, the intervention)
- Randomization
- Threats to validity
➔ True Experimental designs
- Look at OXO notation
- If the sample size is big enough there is no need for a pre-test
- Combining the first 2 designs, interesting as it can tell us whether doing a pre-test can
have an outcome or not
➔ Quasi-Experimental designs
1. No randomisation - we can not be sure that the control group is equivalent
2. One group but several pre–tests and post-tests
