You are here: Home Blogs Inductive Reasoning for Project Managers
Tuesday, 21 April 2009 01:44

Inductive Reasoning for Project Managers

Written by 
Rate this item
(1 Vote)
Project Managers often have to take critical decisions based on unverified information.  Being critical about the value of such information can contribute to more sucessful project delivery.

Project manager are often faced with the inductive thought process as we need to rely on information from internal and external sources, as well as lessons learned, for project planning purposes.  The problem with such information is the possibility of circumstantial variance.  The project manager then has to reduce the information to the most realistic and effective application for successful project delivery.

Reductionism, whereby testimonial justification is traced back to non-testimonial evidence and thereby reducing testimonial justification to non-testimonial justification, is often associated with the Scottish Enlightenment philosopher David Hume.  David Hume’s philosophy is characterised by his empiricism which led him to be sceptical about a lot of things his contemporaries took for granted1/92-93.

The problem of induction, raised by Hume was how we could be sure that the regularities that are observed within a representative sample should increase the likelihood of unrestricted generalisation that the observation is true1/114-5.
A variety of factors could influence the observations.  Someone could have been shown a picture of an ostrich and been told this is an emu.  This would create a Gettier case where such a person would observe with justified true belief that flightless emus exist in Southern Africa, where that person actually observed flightless ostriches in an area where emus never existed as a result of wrong externalistic epistemic norms.

Perceptual knowledge can also be a factor when the observer knows of only one large flightless bird – the emu, and is thereby deceived into unjustifiably believing that the ostrich is an emu.

But in the event that only emus were observed, the only defense to an unrestricted statement, that all emus are flightless birds, therefore, has to be supported by further inductive inferences.  According to Hume there can not be a non-circular way of justifying induction (Agrippa’s Trilemma – Coherentism), furthermore, there needs to be correlation between the observed regularities across sufficiently large and representative samples and the unrestricted regularity itself1/115.

As an answer to the problem there can be various responses:
  • That such a fundamental epistemic practice as induction does not require justification and can therefore be legitimately employed without concern about the availability of non-circular justification, which is intellectually unsatisfactory1/115.
  • To claim that as long as induction works, it does not matter whether there is non-circular justification available.  This would endorse epistemic externalism but will not appeal to everyone, especially not epistemic internalists who believe one must always be in possession of appropriate supporting grounds1/115-6.

Falsification
Karl Popper, however, advocated that one does not have to find a solution to the problem as it is rather one we can live with by claiming that science does not proceed in this inductive fashion but rather by making bold generalisations and then trying to falsify them (falsification)1/116-7.

Popper referred to his four stage model as follows:

   1. the old problem;
   2. formation of tentative theories;
   3. attempts at elimination through critical discussion, including experimental testing;
   4. the new problems that arise from the critical discussions of our theories2/14


Karl Popper was George Soros’ mentor and Soros has a considerable interest in philosophy3/45.   On 10 September 1992 Soros hedged US$ 10 billion against the pound sterling and 5 days later made a profit of US$ 2 billion3/171.  He started by questioning whether the pound would remain within the ERM margins at a time when the general opinion and political commitment was that it would remain within the ERM margin.  He falsified the generalisation and was right.
I suppose one can also question a situation where induction addressed the wrong question and may have required some form of falsification.  Before Coca Cola introduced “New Coke” they had 200 000 sample tastings done and virtually everyone (+/- 98%) preferred the taste of “New Coke”. 

The conclusion seemed infallible – introduce “New Coke”.  Falsification of the induction could have been to ask ‘if the taste is only of secondary concern, what would be the primary concern?’  As it turned out there was such uproar around the removal of the ‘old’ brand that Coke reintroduced “Classic Coke” within 3 months after launching “New Coke”.  They realised they were wrong in relying on taste alone and had never considered the fact that Coke (as it was) had actually become an American tradition4/142.

This leaves considerable room for falsification as deductive inference of a counterexample (as was the case with Soros).  I am of the opinion that applying falsification as a thought process can only contribute to more critical thinking and the possible discovery of more justifiably true beliefs (not necessarily infallible).

Pragmatism
Hans Reichenbach agrees with Hume that there is no justification for induction, but that it is nevertheless rational, at least in one sense of that term, to make inductive inferences.  The rationality of his idea lies in the fact that if we don’t employ induction, we are guaranteed to end up with very few true beliefs, whereas through induction we can form many more true beliefs about the world1/119.

The difference between Pascal’s wager and Reichenbach’s defence of induction is that Pascal did not set out to gain true belief about God’s existence, whereas Reichenbach’s defence of induction is aimed at eventually gaining true beliefs.  To gain true beliefs, we have to trust induction even though it lacks justification.  Such a belief in induction is not directly epistemically rational, rather indirectly; however, according to Reichenbach it is the sort of thing that an epistemically rational person should believe.  From this point of view it may be that Reichenbach’s way of dealing with the problem of induction is not as merely pragmatic as many have supposed it to be1/121.

We can not expect that all induction will end up in true belief; neither can we go around falsifying all beliefs nor regard all inferences as true belief.  Induction can also not be disregarded.  The problem of induction will however remain in that no matter how many observances have been made there may always be that one occurrence that changes it all.

The problem of circularity is that we can not predetermine the number of elements that should be in the circle as this will much depend on the question that is being addressed as well as the extent to which observations are based on existing beliefs and are limited to personal experience.  In the search for knowledge this may actually place doubt on the question that needs to be answered.

Inductive reasoning is supported by probalistic, rather than certain evidence as it will be limited to a plausible conclusion that it is true.  These, however, offers the opportunity to, through induction, address perceived impossibilities and is therefore not necessarily a process to find solutions but rather to search for answers and to question existing ‘knowledge’.  Induction requires open-mindedness as well as the realisation that we may never reach the end where factual findings and opinion meet one another.  An answer to this problem may be very difficult to find, however, the application of a critical thought process will make a considerable contribution to living with a positive approach to the problem.

By applying inductive reasoning during the initiation and planning phases of a project an improved state of understanding the objective will be achieved and reduce the number of adjustments (or changes) that will have to be made during implementation.


References: X/Y: X = publication & Y = page number/s Publications:

   1. Prichard D. What is this thing called knowledge. 2006. Routledge. London
   2. Popper K. All life is problem solving. 2006. Routledge. London
   3. Slater R. Soros – The Life, Times, & Trading Secrets of the World’s Greatest Investor. 1996. McGraw-Hill. London
   4. Keough D. Ten Commandments of Business Failure. 2008. Prentice Hall. New York
Read 10958 times Last modified on Friday, 24 April 2009 13:27
Login to post comments

News and Promotions

Keep up to date with the latest happenings by signing up for our newsletter. Subscribe below.

Twitter Update

Who's Online

We have 330 guests and no members online

Got something to say?