How can you do experimentation in behavioral science?
In this working paper by ETH Zürich, OECD - OCDE and Affective Advisory they highlight both observational and experimental methods.
1) What are Observational Methods:
Purpose: Explore human behavior in natural settings.
-Helps in understanding how environmental and socio-political factors influence behaviors.
-Often lays the groundwork for developing tailored policy solutions.
-Key for policymakers to discern drivers and barriers in specific contexts.
2) Experimental Methods:
Purpose: Establish a cause-and-effect relationship between two aspects.
-The first aspect pertains to introducing or adjusting policies.
-The second aspect observes the behavior and actions of individuals.
-Used to verify if changes in policy have a tangible effect on behavior.
-Ideal when policymakers seek to identify a causal relationship without necessarily understanding the underlying reasons.
Both methods have their advantages and disadvantages, discussed further in the working paper.
The method you use depends on the nature of the evidence you need, and while this is made for policymaking, it very well can be used for businesses.
While observational methods help in hypothesizing, experimental methods prove or disprove those hypotheses - it is best to combine them if possible.
They also created a wonderful map to help you, which you can see in the image, and some questions you can ask yourself.
The questions are
Q1: Do you want to test if a solution works? Q2: Do you want a solution to be applicable to global or local contexts? Q3: Do you want to compare changes across different groups?
Q4: Do you want to measure changes over time? Q5: Do you want to establish a causal relationship between a solution and its effects?
To use it, they recommend you to:
1) Start at the top left of the map and answer the five key questions (the 5 metro stations on the map) in order. The questions go from general to more specific and will help you narrow down the most appropriate method (one of the 7 metro lines on the map).
2) Select the most appropriate method by looking at the answers you have given to the questions (metro stops you want to go to).
Thank you Chiara Varazzani PhD Torben Emmerling Laura Castro Judith Wagner Stefano Brusoni Henrietta Tuomaila and Laura Fontanesi, Ph.D. for creating and sharing it - your work inspires us and makes us progress this great field!
I had a quick browse on the paper and will set time to read it in detail, I will probably add it as a reference on my Behavior Design for Business board under the testing part, as I find it very useful!
If you want to read the paper (which I recommend, it's 40 pages, but a great read) or download the map - I'll leave it in the comments