After you’ve collected and analyzed your data, and drawn some conclusions, it’s time to ask yourself: What’s next?
After you’ve finished your first study on Prolific (or elsewhere), analyzed the data, and drawn some conclusions, it’s time to ask yourself: What’s next? If it’s a well-powered, large study, then it may be informative enough to draw some conclusions. If you are a scientist, then you can probably write it up for publication in a scientific journal.
However, most of the time, one study won’t suffice to give you a definitive answer to your research question, so you’ll require converging evidence, or in other words, multiple lines of evidence. This is called triangulation, and constitutes a fundamental part of the scientific process. Whether you are a scientist in academia, a data scientist in a startup, or a market researcher in a company – this is useful to know for anyone doing any kind of research. In short, triangulation is
“the strategic use of multiple approaches to address one question. Each approach has its own unrelated assumptions, strengths and weaknesses. Results that agree across different methodologies are less likely to be artefacts.”(Munafò & Smith, 2018, p. 400).
The article by Munafò and Smith (2018) hyperlinked above provides a great introduction into this topic and will help you get started.
But methodological considerations aside, how do you know what further questions to ask in your research? How to decide what research questions are most important or pressing in your area of research or in your company/organization?
This is where theory building comes in. In any scientific discipline, it is critical that researchers cumulatively build knowledge and advance our understanding of how the world works. What makes for a good theory? Philosophers and scientists have debated this for a long time. Many books and articles have been written on this topic. We especially recommend this article by Klaus Fiedler, this article by William J. McGuire, this book by Bertram Gawronski and Galen V. Bodenhausen, and this book by Valerie Gray Hardcastle. Shout out to our Twitter friends, who’ve helped us identify these key readings. :)
There do not seem to be any widely accepted criteria for a good social science theory at the moment, but there appears to be more consensus around what a theory is not.
A theory is not:
- References (because they do not provide any causal explanation or logical arguments of a phenomenon).
- Data (because it only describes the behavioral patterns that were observed, without explanation. Thus, it should be used only to support or contradict a theory).
- Variables (because, despite being an important part of a theory, they are not sufficient to constitute it. Theory is needed to explain the connections between variables).
- Diagrams (because they simply depict data and thus do not form a theory for the same reason data alone does not make a theory).
- Hypotheses (because they bridge data and theory. Although they are derived from theory, their main purpose is to specify how variables are interrelated).
What, then, is a good and strong theory?
A good theory answers ‘why’ and ‘how’ and ‘when’ questions, such as: Why do certain phenomena occur? How are these phenomena interrelated? When (or, under which conditions) do these phenomena take place, and when do they not?
Good theory helps to systematically identify causal relationships and the timing of events.
As the underlying processes of a phenomenon are explained by a theory, it can be used to make accurate predictions.
Sutton, R. I., & Staw, B. M.(1995).What theory is not. Administrative science quarterly, 371-384.
For further reading on this topic, we recommend the following articles:
Markovsky, B.(2008).Graduate training in sociological theory and theory construction. Sociological Perspectives, 51(2), 423-445.
Kruglanski, A. W., & Higgins, E. T.(2016).Theory Construction in Social Personality Psychology: Personal Experiences and Lessons Learned: A Special Issue of personality and Social Psychology Review. Psychology Press.
Fiedler, K.(2017).What constitutes strong psychological science? The(neglected)role of diagnosticity and a priori theorizing. Perspectives on psychological science, 12(1), 46-61.