Design and methodology
Design and methodology
Research integrity is all about doing research in a responsible way. Research wants to answer or reflect upon questions about the real world. Therefore, researchers set up studies or experiments, designed specifically to be able to answer their research questions. The choices in the study design are crucial, largely determining the quality of the research. Because of their importance, the methods, rules, models and principles used in research are a topic of (theoretically) study on their own, together called methodology.
It is essential for all researchers to be familiar with the good practices of methodology and design in their field of study.
ALLEA Code:
The ALLEA Code even states it as the first of four principles of research integrity:
- Reliability in ensuring the quality of research, reflected in the design, methodology, analysis, and use of resources.
Good academic practices on (different) methodologies
Methodologies differ largely among disciplines – literature researchers of course use other methods than molecular biologists. Courses in discipline specific methodology are an obligatory part of most university education programmes, as well as doctoral training programmes.
ALLEA Code:
- Research institutions and organisations ensure that researchers receive rigorous training in research design, methodology, analysis, dissemination, and communication.
It’s impossible to summarise all disciplines’ methodologies, but in most research fields, for instance, the following considerations are key:
- The variables measured (both influencing factors and outcome) should correspond to the question you want to answer, as directly as possible.
- The design of the experiment or study should enable you to answer the research question.
- Some form of randomisation might be advisable to eliminate the unwanted influence of confounding factors (like unknown or unmeasurable ones).
- The sample size should be appropriate: large enough to be able to statistically detect existing phenomena (according to a power calculation) – but, if applicable, not larger than ethically responsible.
Good academic practices on design and methodology
When you have carefully designed and planned your research actions and analysis, many research disciplines consider it a good practice to:
- Specify all choices for the study design and research activities in a protocol (that may even be published).
- Preregister the study and plan of analysis.
- Have your preregistered study and analysis plan peer reviewed, and in principle accepted for publication, before gathering data – so-called registered reports.
- These exemplary practices reduce the risks of unacceptable research practices like bad study design (because your study design will be reviewed before actually doing the research) or HARKing.
Cartoon by Patrick Hochstenbach under a Creative Commons CC BY-SA 4.0 license
If you perform a study or experiment together with colleagues from other disciplines or institutions, it is vital to share the same vision on the study. Take the time to make sure everyone is on board with all aspects of the plan, including why choices were made. Agree on the plan for data analysis with all partners, preferably before gathering the data.