I typically peer review about two papers a year and I remember a few years ago, I conducted a review of a potentially helpful study on fishing boat air quality and the health of fishermen. It was a good study, with helpful findings. Unfortunately, the sample size was unclear and when I made inquiries, I found it was based on a sample size of one boat and crew. I recommended that this sample size be stated explicitly or the paper be rejected. I believe the paper was withdrawn because it never came back on my docket.
I have long been amazed at the tiny sample sizes used in many studies. This isn’t evil but the potential weaknesses of studies need to be more clearly stated. With the Reproducibility Project finding only 40% of psychology experiments as being verifiable we should all have a cynical eye on how science advances.
It is easy to overwhelm people with copious data and statistical analysis. The process of verifying someone else’s data has zero sex appeal. A.E. Housman opined: "Some individuals use statistics as a drunk man uses lamp posts - for support rather than illumination." I think he was right.
There exists a battle within science. Does is operate in an environment where scientists quickly abandon previous theory in the face of evidence, as Karl Popper would describe? Alternatively, is it founded on paradigms and group behavior, as Thomas Kuhn would assert?
Some philosophers (e.g., Kuhn and Feyerabend) have extended this inquiry into the limits of the scientific method. They argue that the outcomes of revolutionary times in science occur when traditional mental models or paradigms are being challenged. Scientists will not abandon a theory based on a single, falsification event. They seek (or await) a new paradigm that accommodates new data; however, they do not do so with alacrity.
Data usually drives paradigms, from ‘grounded theory’ approaches in the social sciences to inferential statistics. However, data, or in a general sense evidence, is a product of data acquisition methods. Evidence is also routed through biases connected with epistemology and context (such as time, environment or other circumstances). Moreover, evidence is routed through paradigms. All of these issues, whether data error or biases rooted in epistemic or context, are a threat to objectivity. The interface of evidence and bias can be summarized by Heisenberg’s assertion that “the world cannot be separated from our perception of it.”
Should designers care about scientific philosophy?
Design is an applied science and reflections on the capability and limitations of science are helpful for understanding potential benefits. We can recognize that not all technical advances are orderly and founded on historical developments. Some philosophers (e.g. Paul Feyerabend and Stanislav Grof) would consider the whimsy and chaos of individuals as productive and powerful agents of change. Others (e.g., Richard Rorty) considered some final truth to never be attainable or even desirable, which points back to the 18th century Scottish philosopher, David Hume’s realization of there being no end point in proving scientific theories. He argued that the contentious and muddy world of scientific (and what Kuhn would call pseudo-science) pursuit is good.
However, the necessity of creative forces in formulating new paradigms does not negate the power of the scientific method, from the practical inventions of Thomas Edison to Richard Feynman’s march through theoretical physics. Designers use abductive reasoning with its reliance upon a conclusion of an operating product rather than formal rules and preconditions – the ends justify the means. And they can throw away bad results/conclusions.
Design can seem to be arbitrary – a small whimsical thought can lead a team to pursue an approach that cannot easily be reversed. Is this idea superior or is there something much better that has not been conjured? Synthesis of ideation and ‘bounded rationality’ will lead to a final concept upon which copious time and talent will be spent.
Designers need science and engineering. Advances in material science, information technology and manufacturing techniques provide our fuel for dramatic change. In the end, we seek to design things that work. The word ‘things’ is used purposely. The 21st century designer is not just designing products — we are designing experiences, interactions and solving systemic problems. We run our hand along the guiderail of science but are not afraid to use our creativity and undefinable abilities in developing design solutions.