Formulating hypotheses

What is a hypothesis?

In this context, a hypothesis is a specification of the expected effect of the intervention. If you expect an effect on 'job satisfaction', the hypothesis should clearly define what is meant by job satisfaction and whether it is measurable. In addition to being measurable, a hypothesis should be verifiable, specific, and preferably supported by literature and/or practical knowledge obtained in the previous steps. One method to support this is the SMART approach (see Useful links).

The steps ‘Existing evidence’ and ‘Formulating hypotheses’ may be repeated. New insights from the literature may lead to the need to revise hypotheses.

Considerations

When formulating hypotheses based on the effect map, there is a risk of a large number of hypotheses. An intervention and the associated changes in activities often lead to multiple expected effects. For each effect, one or more hypotheses can be formulated. To deal with this in a practical way, you may consider the following:

• Involve the key stakeholders in selecting the hypotheses to be tested. This helps to ensure that the resulting research is supported by those involved.
• The hypotheses form the basis for the measurement plan. Consider for which hypotheses it is realistic to conduct research.
• Not every hypothesis needs to result in large-scale research. Even a limited amount of new knowledge may provide the missing insight. Therefore, weigh whether investigating a hypothesis is likely to generate sufficient new knowledge and whether additional research is needed.
• Various frameworks can support the prioritisation of hypotheses, such as MoSCoW or RICE (see Useful links).

Useful links

• The RICE scoring model. (z.d.). Open Practice Library.
• Wikipedia contributors. (2025, 3 april). MoSCoW method. Wikipedia.