Innovation Blog

By Shlomo Maital

Innovation, Absolute Truth and The Broken Workings of Scientific Research

  

 

After 40 years of academic research, writing and publishing close to 100 papers published in refereed journals, I believe I know something about the ‘scientific method’.  Here are two things I’ve learned – a) the ‘scientific method’ is seriously flawed, and b) it is not appropriate for research on companies, management, leadership and organizations.

    A recent article by Jonah Lehrer in The New Yorker supports my (a) claim.*  Lehrer documents how even rigidly-run scientific clinical studies that ‘proved’ a drug was effective, now show that similarly rigid studies show the drug is ineffective. This is happening increasingly often. And it is hard to explain. What is very easy to explain is the ‘paradigm bias’ Lehrer mentions, in refereed scientific journals, the journals whose ‘yes’ or ‘no’ is a matter of life and death for young researchers.  If you write an article attacking an ‘old’ scientific paradigm or theory, you increase your chances of getting it accepted.  But if you write an article attacking a relatively new paradigm, chances are it will be rejected.  And worst of all, researchers ignore negative results (“this drug doesn’t work”) and report only favorable results (smoking was  correlated with alcoholism, at a 95% significance level), often with correlations that are statistically significant but in practice meaningless.  (A large sample, which many journals favor, can generate “significant” correlation coefficients that are only 0.05).   So researchers dig, and dig, and dig, until they find something ‘significant’, and with enough digging you always find something…even though failure to report negative results indicates what you found is really just random noise (enough monkeys typing on typewriters will….)

    But what about (b)?  I have argued, after long experience, that survey data, statistics, correlations, all the social science voodoo methodology, are effective in obscuring the truth about organizations and management, rather than revealing it.  Rather than build hypotheses and models in advance, why not get out into the field, work with companies and managers, ask them a ton of questions, and only then, when you think you really know what is going on, sit down and build a theory or framework. This is known as grounded theory, and it is the only effective way to research issues that are highly complex, with interactions among the players.**  It is why business schools must liberate themselves from the terrible tyranny of the scientific method, and build their own integrated approach, which links consulting with research and teaching. (“I consulted in project management at Hi-Teck Inc., and here, students, is what I learned, in this and other companies, about the role of stage-gate planning in project management”).   Alas, qualitative research is very hard to publish, and very very few journals accept such work.  So, in future, I fear management faculty will continue to play by the rules of the scientific method, as set by the science faculties, and management research will continue to be of little value, for this reason.

 * Jonah Lehrer, “The Truth Wears Off”, The New Yorker, Dec. 13, 2010.

** Shlomo Maital, Srinivas Prakhya, and DVR Seshadri, “Bridging the Chasm between Management Education, Research, and Practice: Moving Towards the Grounded Theory Approach”.  Vikalpa-Indian Management Journal, Jan-March 2008.