Some years ago, skeptical scientists began to question these methods, observing, for example, that cancer cells in a petri dish behave so differently from tumors in a human body as to cast doubt on much conventional research.
Gabriel Popkin, Cancer and the artillery of physics, Johns Hopkins Magazine, Spring 2018. Accessed March 22, 2018.
The tools we use to help us think—from language to smartphones—may be part of thought itself.
Larissa MacFarquhar, The Mind-Expanding Ideas of Andy Clark, New Yorker Magazine, April 2, 2018. Accessed May 11, 2018.
Scientists have confirmed a longstanding hypothesis that Earth’s orbit is warped by the gravitational pull of Jupiter and Venus in an epic cycle that repeats regularly every 405,000 years.
Peter Dockrill, Jupiter And Venus Are Warping Earth’s Orbit, and It’s Linked to Major Climate Events, ScienceAlert, May 8, 2018.
And including microbiome characteristics when predicting people’s traits, such as cholesterol levels or obesity, makes those estimates more accurate than only personal history, such as diet, age, gender, and quality of life, the study finds.
Jim Daley, Environment, Not Genetics, Primarily Shapes Microbiome Composition, The Scientist, February 28, 2018. Accessed May 11, 2018.
For another, researchers often focus their attention on just a few interesting microbes, “and people just don’t look at what the remaining things are,” Kowarsky said. “There probably are some interesting, novel things there, but it’s not relevant to the experiment people want to do at that time.”
More than 99 percent of the microbes inside us are unknown to science, ScienceDaily, August 23, 2017. Accessed May 11, 2018.
Increasingly, researchers across a variety of fields are discovering that better knowledge of our selves and our world is produced by making the units of analysis used in research more inclusive and more dynamic. The standard entities we use in everyday language (e.g., the body, the family, the city, the nation, the ecosystem) are often not the most productive units of analysis for scientific research. They externalize and hide much that is really part of the actual system of causation and they tend to leave the evolution of entities and boundaries over time out of consideration.
The implications of new understandings in the physical and biological sciences for the social sciences are significant. Studying kinship groups or cities or nations in a short time frame as though such entities and time frames enclose everything that is explanatorily relevant is less likely to produce durable explanations than we have thought. We will be better served, it appears, by embracing a more inclusive research perspective.
Most scientific research is done by starting from hypotheses with only two or a few variables, and then perhaps, adding in one or two variables at a time in a search for a still simple but sufficient explanation. In this approach, the field of inquiry is kept as limited as possible. The research path is from simplicity to complexity, with the assumption that a fairly simple explanation will be found.
This attempt at simplicity is not only spatial, it is also temporal. The explanations being sought must not only be simple, they must be time proof. This is another simplifying premise. There is no history that must be studied; measuring variables in one short period of time is assumed to suffice for confirming or disconfirming the full range of hypothesized explanatory relationships.
This approach is desirable because simple hypothesized explanations make possible low cost research and simple policy and treatment interventions. However, the approach brings with it a major source of confusion and controversy. Even for a system composed of only a few variable components, the number of possible two or three variable hypotheses is quite large. Add in a temporal dimension and the range of competing hypotheses grows even larger. For a real world research question, the wide range of explanatory hypotheses possible invites numerous competing and contradictory explanations. Moreover, given the evidence that research findings are quite often wrong, each researcher is also inclined to hold tight to their particular simple explanation as long as even one study seems to confirm it. Over the long run, the multiplication of attempts at simple explanations can run up quite a tab for research funders and yet produce very disappointing explanations.
The rise of complexity theories and the increasing use of dynamic systems thinking in research suggest an alternative research approach: starting with a unit of analysis that is as spatially inclusive as seems plausible and studying it over a significant period of time seems likely to be more fruitful that the current approach. In this approach, researchers would start with complexity and work toward simplicity, eliminating factors that can be shown to be causally inconsequential. This approach has four advantages. First, it aligns with the growing number of studies that show that the system totalities that matter are larger and more inclusive than we have thought. Second, it is more likely to define a common research orientation for the many research institutes and researchers studying the same topic. Third, the inclusion of time gives researchers a better chance to learn whether a discovered explanatory system is evolving over time or is stable. Finally, it aligns with the scientific principle that we can prove that a causal relationship doesn’t always hold, but we cannot prove that it does always hold.
For the study of workforce changes the Inclusive World Economy perspective that I have adopted (and which is derived from the World-Systems concept developed by Immanuel Wallerstein) provides the kind of system totality that probably encompasses all the possibilities for explaining changes in employment. It also makes it easier for many workforce change researchers to adopt the same research orienting perspective even while focusing on different hypotheses. We start with the grand hypothesis that policies, practices, and events in every part of the world and every part of nature have consequences for workforce changes in the U.S. We add to that the premise that explanatory constancy cannot be taken for granted; it must be demonstrated, not assumed. The shared research task is to work inward, throwing out factors that can be shown to be minimally relevant to the workforce topic being studied. We still make use of existing research findings, but instead of looking for research that shows which variables have explanatory efficacy, we look for research that shows which variables have been found in multiple studies to have little or no explanatory efficacy. A simple explanation is not the starting point in the search for a sufficient explanation; it is only a possible end to that search.
Widely adopting this approach would be a big shift in how we study workforce change, but it should be a fruitful shift. A growing record of explanatory controversies and failures in the social science fields begs for a new approach, and developments in the physical and biological sciences suggest that adequate explanations for workforce changes will involve more factors and be more complex than has been assumed. These things given, working from the inclusive and complex toward the simple should be at least as efficient in the expenditure of time and money as is the approach that now dominates the study of workforce changes and so often disappoints.