In its most specific meaning, the term correlation is associated with statistics. This is a classical story we all know. However, it is still a story that needs to be told and criticized because it claims neutrality and objectivity while it is an act with a political meaning. What more needs to be told? As we are leaving traces behind, correlations are only one of the possible constructions that can be made on those traces. The correlated human is only a small part of the traceable human whereas the latter seems to be a more interesting and challenging one. (Maybe we should consider to rename our ’transversal theme’?) Generally speaking, traces as such are neither good nor bad but they can be used in different ways. One tends to have a rather negative view on the matter, however, we will look at the problem from the bright side. In a first example we shall elaborate on ’mathematical traces’. They lead to a better understanding of mathematics. Within the context of the classroom they will alert students of the relativity of mathematical truth and knowledge and it puts mathematics back in a humanised perspective. In a second example we want to plea for traceability in the political context. Within the scientific context, traceability is a precondition whereas in the political (and social - cultural) context it is rather seen as potential threat to human rights and privacy. Finally we shall elaborate on the eventual comparison between sciences and politics.
The pleasure to walk on earth, leaving your traces (behind) slide Karen Fran?ois In line with the talk of Hans, in this presentation I shall try to look at the correlated or traceable human from the bright side. But not before saying something about the classical interpretation of the correlated human. In this presentation I shall elaborate on: slide Correlation: from a mathematical matter of fact to a mathematical matter of concern The correlated human which is only a small part of the traceable human The bright side: Mathematical traces and the history of mathematics Traces in sciences and politics Let us start with the classical meaning of correlations. slide A correlation is often seen as a mathematical matter of fact Often used in psychology, social and political sciences To explain, to predict, to proof something … To me a correlation is a mathematical matter of concern. I will show you some maters of concern using statistical correlation. slide Correlation is a statistical technique which can show whether and how strongly pairs of variables are related. A correlation is a linear relation between the values of the variables. For example, height and weight of people are related. (photo) slide However some correlations have no meaning. Slide J For example, the size of shoes and mathematical understanding / mathematical problem solving Another example is the significant correlation between human birth rates and stork population sizes. (une cigogne) slide The main result of a correlation is called the correlation coefficient (or "r"). It ranges from -1.0 to +1.0. The closer r is to +1 or -1, the more closely the two variables are related. If r is close to 0, it means there is no relationship between the variables. If r is positive, it means that as one variable gets larger the other gets larger too. If r is negative it means that as one gets larger, the other gets smaller (often called an "inverse" correlation). In Dutch we call it a negative correlation. It is possible to calculate the correlation coefficient from whatsoever data, where we have at least two variables and a variety of soft-ware packets is available where you have only to ’double click’ to calculate correlations. slide On the other hand there can be a perfect functional association between variables although the correlation coefficient is == ZERO http://noppa5.pc.helsinki.fi/koe/flash/corr/corrx2-en.html The classical correlation technique does not work well with a non-linear or a curvilinear relationships (in which the relationship does not follow a straight line). An example of a curvilinear relationship is age and health care. They are related, but the relationship doesn’t follow a straight line. Young children and older people both tend to use much more health care than teenagers or young adults. However there are statistical procedures to transform a non-linear relation into a linear one. e.g. multiple regression Another mater of concern is the types of data. slide Like all statistical techniques, correlation is only appropriate for certain types of data. Correlation works for data in which numbers are meaningful, usually quantities of some sort. Stricto senso it cannot be used for purely categorical data, such as gender, political party, religion, nationality, IP number or favourite colour. In statistical terms, there is a difference between correlation and associations where the latter is used for ordinal (rating scales) and nominal scales. The term correlation is used for ratio scale and interval scale Rating scales are a controversial in-between case. The numbers in rating scales have meaning, but that meaning isn’t very precise. They are not like quantities (e.g. euros) where the difference between 1 and 2 is exactly the same as between 2 and 3. With a rating scale, that is not really the case. For example, the taste of coffee is disgusting, bad, normal, good or excellent. Or maybe one can say, it has ’no taste’. What does it mean? Most statisticians say one cannot use correlations with rating scales, because the mathematics of the technique assume a ratio scale. However, many survey researchers do use correlations with rating scales, moreover, nominal scales are often used. For example nationality, religion, mother tongues or gender. They only have to transform or rename the categories into numbers where for example man and woman are renamed as 1 (for man) and 2 (for woman). In the case of gender it is always this kind of translation (and is was Simone de Beauvoir who explained why women are Le deuxi?me sex). How to present data slide Voorbeeld van jean paul (inkomsten ratio) How to understand correlation coefficient slide While correlation coefficients are normally reported as r = (a value between -1 and +1), squaring them makes them easier to understand. The square of the coefficient (or r square) is equal to the percent of the variation in one variable that is related to the variation in the other. After squaring r, ignore the decimal point. An r of .2 means 4% of the variation is related (.2 squared =.4).An r of .5 means 25% of the variation is related (.5 squared =.25). An r value of .7 means 49% of the variance is related (.7 squared = .49). In social and political sciences, low correlation coefficient (0.2 - 0.3 e.g. are often used) to explain, to conclude … Normally, a correlation goes together with statistical significance. slide In spoken or in normal language, "significant" means important, while in Statistics "significant" means probably true, it means: it is not by coincidence, not due to chance. Significance level