yellow shape
[EN] Innovation Hub Blog

Reducing bias in the tracking of social impact

On the 30th of June Partos hosted an Inspiring & Connecting session for members of the PMEL-community. The (aspiring) members of this community engaged in a lecture and workshop with The Bias Brothers. They presented the participants with common biases in programs and processes of investigating and measuring change. The Bias Brothers acted on behalf of the ‘Up to Data’-network which aims to have a societal impact by increasing awareness of, and skill in, handling data and technology in contemporary organisations. 

Author: The Bias Brothers 

09 augustus 2022

Every human comes equipped with a brain that is prone to cognitive biases. Even though we are seldom aware of them. These biases influence much of our perception and decision-making, which is mostly a good thing because it increases the efficiency of our brain. However, sometimes these phenomena also cause unwanted errors and problems. Especially, in situations that present us with an abundance of data and information. 

The role of PMEL professionals in organisations with a mission of social impact (PMEL stands for ‘plan, measure, evaluate, learn’) is becoming increasingly important and encompassing. These professionals are constantly looking to improve ways to measure and evaluate the impact that NGOs have. And the trend is to do this in a data-driven fashion. Specifically, these types of roles are bombarded with ever-increasing amounts of data. Next to the need of making difficult choices on how to measure and evaluate this data. Hence, the people fulfilling these roles are prone to bias.  

The Bias Brothers dove into the matter of bias in PMEL-cycles with the participants. Looking for observer-expectancy bias and illusion of causality bias in their own way of operating. Through peer-discussions, participants were able to learn from each other while on the hunt for specific fallacies in their thinking and data processing procedures. 

For example, when you are on the lookout for mere positive examples of the social impact of initiatives in your data you might experience observer-expectancy bias. Which could mean that in evaluating outcomes of data projects you only see a confirmation of your positive hypothesis. And not seeing room for improvement in the way you act also is enclosed in the data. Since your pattern of expectation makes you sort of blind to failures or setbacks within your dataset. Hence, you observe-what-you-expect. 

The lecture was based on the work of Daniel Kahneman, who has become famous for his work on cognitive bias in behavioural economics. And has coined the catch-phrase “what you see is all there is”. The participants have expanded their field of vision by participating in this event. With that, they have probably opened up a bit more for that what has previously not been seen (or measured and evaluated). 

If you want to know more about the work of the Bias Brothers, please check out their website here