The Weekly Reflektion 10/2025
Welcome to Reflekt’s Breakfast Seminar, Wednesday 26th March 2025 0800 to 1000 at the Quality Pond Hotel in Forus.
The main theme for Havtil in 2025 is Artificial Intelligence is also a risk factor. This theme and the depiction of the ‘black box’ in the Havtil film is the inspiration for our next breakfast seminar.
‘The black box’, can it be trusted?
Contact graeme.dick@reflekt.as, mike.pollard@reflekt.as, or post@reflekt.as if you would like to attend.
More details on our website www.reflekt.as
Breakfast will be served at 0730.
Causation is an important scientific principle and is fundamental to the understanding of cause and effect. Distinguishing between causation, correlation and coincidence is not always easy, andfailure to do this often leads to jumping to conclusions, in particular when these confirm a hypothesis that one would like to be true.
Do you behave differently if someone is watching you?
The Hawthorne effect refers to the phenomenon where individuals modify their behaviour in response to being observed. It originatedfrom a series of workplace studies conducted at the Hawthorne Works (a Western Electric factory in Illinois) in the 1920s and 1930s. Researchers were studying how different factors like,lighting, breaks, and work hours, affected worker productivity. Turning up the lighting in the factory for example, increased productivity and the researchers speculated over the reasons why this might be the case. Several other changes were made that led to increased productivity and some were less easy to explain. When the studies were completed, productivity returned to the same level as before. The increases were not sustainable. What was observed in the studies did not therefore represent “normal” behaviour, threatening the internal and external validity of the research.
Other examples of the Hawthorne effect that have been observed in research.
Healthcare and Medicine: Patients often report feeling better or adhering more strictly to treatment plans when they know doctors or researchers are closely monitoring their progress.
Education and Learning: Students perform better on tests or participate more actively in class when they know the teacher is paying close attention to them.
Fitness and Exercise: People exercising at the gym often push themselves harder when they know others are watching or when they’re wearing a fitness tracker that logs their progress.
Customer Service: Employees in retail or hospitality often provide better service when they know secret shoppers or customer reviews are evaluating their performance.
The fact that observation itself can influence behaviour, sometimes more than actual changes in environment or incentives, is important in designing experiments, managing workplaces, and understanding human motivation.
Another interesting aspect is related to the motivation for carrying out productivity studies. Many workplaces have experienced increase in productivity, including sustained increase, where the workers perceive that the management had a genuine interest in helping people with their challenges. When people felt that the management cared and were willing to make changes to improve the workplace then they responded positively. The opposite of course is also relevant if the people perceive the management intentions are not so honourable.
Colin Powel the former US Secretary of State recognised the Hawthorne Effect when he said, ‘Always do your best. Somebody is watching you’. It is of course important to get the right balance here and not create paranoia in the workplace.
Understanding the difference between causation and correlation is crucial in many areas. Correlation is when two variables change, and the changes are related. For example, one might decrease as the other increases or vice versa. Causation means one variable directly influences another, for instance, one variable increases, because the other decreases. Testing and analysis are required to confirm whether two variables are merely correlated or have a cause-and-effect relationship. We are unfortunately often influenced by confirmation bias in our analysis, and we can succumb to the desire to prove our hypothesis of a cause-and-effect relationship. We are often disappointed when the measures we put in place to improve do not give the desired result.