Our team was privileged to be guests of the Rhode Island team this past Friday. Our goal was to help develop a quasi-experimental design to example the impacts of the CHWs throughout the state.

Hey, what’s a quasi-experimental design?

A quasi-experimental design is a type of research design similar to an experimental design, but with some important differences. In a true experimental design, researchers randomly assign participants to different groups (such as a treatment group and a control group) and then manipulate an independent variable to observe its effect on a dependent variable. In a quasi-experimental design, however, the researchers do not have complete control over the assignment of participants to groups.

Type of quasi-experimental designs

There are several different types of quasi-experimental designs, but some common examples include:

• Nonequivalent control group designs, in which participants are not randomly assigned to groups and there are pre-existing differences between the groups.
• Pretest-posttest designs, in which researchers measure the same dependent variable before and after an intervention, but do not randomly assign participants to the intervention or control group.
• Time-series designs, in which researchers measure the dependent variable multiple times over a period of time, but do not randomly assign participants to a treatment or control group.

Quasi-experimental designs can be useful when it is logistically or ethically infeasible to use a true experimental design, but they also have some limitations. Because the groups are not randomly assigned, it is more difficult to control for outside variables that may be affecting the dependent variable. Additionally, without random assignment, it is more difficult to establish causality between the independent and dependent variables.

The Coin of the Realm: Causality

Causality can be established if researchers can rule out alternative explanations of the observed results, by doing this they can increase the confidence that the relationship between the independent and dependent variables is causal.

Quasi-experimental designs can be helpful in situations where a true experimental design is not feasible or ethical. Still, it is important to be aware of the limitations of these designs and to interpret the results with caution.

So, our goal here is to see how different levels of CHW support contribute to building capacity and outcomes

Enjoying the Wifi on Amtrak

But in the meanwhile, Luisa and I got to enjoy some sweet Amtrak wifi. More to come on our work!