Monday, July 16, 2012

Theories need evidence, facts need proof.

Always nice to stumble upon an academic bun fight. For the very tip of a very large iceberg on the relative merits of quantitative versus qualitative analysis see here, here or here.

But I'm a PM not an academic so what's the angle? My qualitative answer would be that knowing the difference will sometime enable a project manager to make optimal decisions . My quantitative answer is about 45 degrees - which to be fair isn't much use in this context.

For reference.

Quantitative research consists of those studies in which the data concerned can be analysed in terms of numbers ... Research can also be qualitative, that is, it can describe events, persons and so forth scientifically without the use of numerical data ... Quantitative research is based more directly on its original plans and its results are more readily analysed and interpreted. Qualitative research is more open and responsive to its subject. Both types of research are valid and useful. They are not mutually exclusive. It is possible for a single investigation to use both methods. (Best and Khan, 1989: 89-90)

Qualitative research is harder, more stressful and more time-consuming than other types. If you want to get your MEd dissertation or whatever finished quickly and easily do a straightforward questionnaire study. Qualitative research is only suitable for people who care about it, take it seriously, and are prepared for commitment (Delamont, 1992: viii)

Both these excerpts are from "An introduction to the qualitative and quantitative divide"

Whether it be the generation of the initial business case, the management of risk, key design decisions or resource planning, the project manager is faced with a veritable zoo of decisions. Some are more critical than others and on a sliding (qualitative) scale we make decisions which if wrong have very limited consequences to those 'irreversible' decisions to which great heed must be paid.

Question; has anyone ever sat down with project stakeholders and asked them if they want quantitative risk management, qualitative risk management or both? Do you understand the question? Would your stakeholders? Does it matter?

Take the following examples. First, what I hope will look a fairly typical excerpt from a fairly typical quanititative risk log. All with me so far?

 Next, something from the qualitative end of the spectrum.

R1 - qualitative assessment

Failure to provide adequant fencing, or early warning mechanisms as appropriate may result in injury or death to Donald Duck.

(A little digression here on the the two approaches to risk - if you fail to communicate to your board the potential impact of a risk with numbers (quantitative) try words instead (qualitative). On more than one occasion I've managed to elicit a response by describing in detail the consequence of a risk occuring having failed by ascribing it an impact and probability.

The Beaufort Scale incidentally makes rather good use of both qualitative and quantitative approaches - that's meteorology for you.

The PERT weigted average here is purely quantitative.

Now I could rattle on at length here, but I don't know if a blog post is quite the place for it. So I'll leave you with a couple of summary points.

  1. Make a judgement of which type of data suits your needs and then go and get it
  2. Personally, I prefer numbers to adjectives (with the proviso that they're right)
  3. Look back at the quantitative risk assessment example above. Are your quantitative risk assessments based on analysis of the statistical likelihood of the event and the impact to cost and time should it occur? If not, then your quantitative analysis is in fact a qualitative analysis.

No comments:

Post a Comment