Quantitative Analysis

Numbers seem to govern our lives these days. What percentage of people think this? What proportion have worked where? How many can deliver which skills? How fast is demand growing?  Will there be enough young graduates in certain subjects? The questions are endless and accurate answers are vital to effective decision-making.

We’ve been gathering quantitative data for decades – ever since the answers were in those strange things called books (actually they were detailed statistical annexes). Nowadays there are many sources of statistical data, and many of them are based on subtly different assumptions, the data gathered in different months and from slightly different populations.

Sampling is important, question design is important, and expert analysis is vital.

Sampling is not as easy as it looks – especially where the overall population is not well-understood. Achieving a representative sample is easy – if the budget is unlimited. Getting a representative sample cost-effectively requires experience and expertise. Questions must be clear, unambiguous, and not leading. Handling margins of error and significance testing is not for the faint-hearted.

Careful design and even ordering of questions can make a huge difference: people who are good enough to give of their time to respond to polls and surveys expect – and deserve – variety and interest in the survey. If that variety can be delivered alongside added-value to the client, all the better.

For example, Pye Tait developed – back in 2002 – an approach to skills analysis which – from very standard information – can tell a sector or a company which skills will merit the most training effort over a given span of years. Our skills-scoring system has now been used in dozens of studies.

Getting at the numbers is relatively straightforward – although there are always compromises to be made. The real benefit of employing experts lies in their being able to tell you what they mean.