November 24, 2025
Will AI make us smarter or dumber? The insights of Klöpping, Scherder and Online Dialogue
Reflection on Klöpping × Scherder by Simon Buil (Data Analyst at Online Dialogue)
Running an optimization program hinges on forming hypotheses. Always forming your hypothesis in the same way creates overlap between the various studies you conduct. This overlap not only ensures a higher return from your program but also safeguards knowledge within your organization. Hypotheses thus form the basis of a data-driven knowledge culture in which examining assumptions is central and decision-making takes place based on data-driven behavioral insights.
At Online Dialogue, psychologists form the hypotheses and ensure that the hypotheses of all the different studies overlap. Hypotheses are formed based on insights from data and previous studies (Do good research with our 6V model). Each new hypothesis takes into account the main line (main hypothesis). In this way, all individual studies contribute significantly to a bigger picture: fundamental data-driven behavioral insights. How we create these hypotheses and how you can ensure that a clear line emerges in your studies will be explained in this article.
To arrive at a hypothesis and investigate it, 8 steps are usually followed in science. Also at Online Dialogue we formulate our hypotheses this way (see article FACT&ACT).
A hypothesis predicts the relationship between two or more variables. However, a hypothesis consists of more than a prediction. For this reason, each hypothesis always begins with a research question that is then analyzed with background research. Only after the background research is completed does the researcher formulate a hypothesis. Your hypothesis should always be an educated prediction of what you expect to happen during your experiment.
Beware that your hypothesis is very unlikely to be correct. The hypothesis predicts what you, the researcher, expect to happen. The purpose of the study is to determine whether or not the prediction is wrong. There are always multiple variables that can explain the effect so there is a high probability that your hypothesis is incorrect.
Important here is that you carefully analyze which variables, which you did not include in your hypothesis, may have been influential. Even if your hypothesis turns out to be true. This analysis forms the basis for yet new hypotheses. By using steps 6 and 7 as input for steps 1 and 2 in this way, you create a clear red line in your research and thus your hypotheses.
Whereas a hypothesis is often still described as a question or an instinctive assumption, it is actually more specific. A hypothesis can be described as a substantiated assumption about a relationship between two or more variables. For example, ”Buying habits depends on positive feelings of certainty”
Before forming a hypothesis, take the following steps to ensure that your research hypothesis is a well-founded assumption.
One way to construct a research hypothesis is according to the following format:
“If [these changes are applied to the following “predictors”], then [an observable change will occur to a specific “outcome variable.””
Examples:
There is a lot to do about forming hypotheses. On paper, a hypothesis is an educated assumption about what you expect to happen. By working according to a set method and properly substantiating your hypothesis before you formulate them, a red line will emerge in your studies as well, your research ideas will be inexhaustible and the return on investment of your program will increase. Did this checklist help you and what are your experiences?
Nederlands Nederlands
English