Introduction

Abstract

Numbers are everywhere: they tell us what date it is, how long I have to work, and how much money I have in my bank account. We rely on mathematics to build houses and bridges, to produce sweets and cast concrete, to calculate costs, to analyse economic trends, and to make estimations in many areas such as health care and education. Living our life without numbers, calculations, and estimations is nowadays almost impossible to conceive. But the grip they offer on “our reality” is also beset with a number of problems: more often than not they create an illusion of straightforwardness, of facts which cannot be doubted, of what cannot be denied, i.e., “what is the case.” Numbers have a seductive objectivity. We are likely to be forgetful of the presuppositions that are implied by certain uses of numbers. We tend to forget that everything starts with a choice about what one may want to focus on, i.e., a particular concept that is the beginning of our research. What is the case for things that can simply be counted gets even more complicated in statistics where various levels of measurement (nominal, ordinal, interval, and ratio scale) are involved. Analyzing the data according to the presuppositions of various scales and correlative statistical (often very complex) techniques for testing a hypothesis requires an in-depth understanding of probability (which itself requires mathematical insight and knowledge). The use of statistics in educational research is very widespread, as it is in many other fields of research. Sometimes, the outcomes of the analyses are critically discussed when presented in the popular media, but more often than not they are shaped as “facts”. Experts understand that things are not so simple. But where reports are broadcast on radio and television, or when figures or tables are printed in newspapers in a manner that lacks the sophistication and the nuance they require, we are on dangerous terrain: their distorting picture may be a travesty of reality. For the sake of clarity (in order to make it less complex), such uses could cultivate one-sightedness, prioritizing an image of reality at the cost of many possible other views. The lack of understanding of the purposes and limits of statistics, of the different ways they can be interpreted, is particularly damaging.