The limitations of musical datasets

The value of statistical techniques in historical musicology depends on the quality of the available data. The extent and diversity of these sources is considerable, but it is important to remember that they can only ever illuminate a small proportion of the musical world.

A historical musical dataset can be thought of as a snapshot of part of the entirety of musical activity. Although we may be tempted to extrapolate our conclusions beyond the scope of the data, there are fundamental reasons why such extrapolations can only ever be valid within narrow limits.  Continue reading →

Triangulation

Surveyor 2Triangulation is a research technique that involves looking at the same thing from two different perspectives. In surveying, it enables positions and distances to be calculated by measuring angles from two locations. In the social sciences, it can increase the reliability of conclusions if they are found by two (or more) different methods. And in statistical historical musicology, looking for the same works or composers in two or more datasets can tell us a lot about the characteristics of the datasets, and about the works’ patterns of survival or dissemination. Continue reading →

Further thoughts on Composers’ Star Signs

Aquarius symbol 3Virgo symbol 3Following this previous article, a friend got in touch to thank me for disproving some astrological ‘nonsense’. I replied that I had not disproved anything – I had just failed to find supporting evidence – but it did get me wondering about the nature of the conclusions that can be drawn from this sort of analysis.

Suppose, for the sake of argument, that people born under Aquarius do show a significantly higher propensity to become composers than those born under Virgo. Consider these three possible explanations… Continue reading →

Pick a composer, any composer

jackmac34 / Pixabay

Often in statistical analysis we need to select things at random. For example, if it is impractical to work with a complete dataset, the only option might be to use a random sample. The science of statistics tells us how to analyse a sample in order to reach conclusions about the entire dataset, and gives us ways to calculate margins of error based on the size of the sample. But I digress.

So, how might we pick a random composer? Continue reading →

Women Composers: Sources and Bias

Louise Farrenc (1804-1875)

There is a lot of interest at the moment in women composers. Until recently, women were a small minority of the composing population, but in working with large datasets, I encounter a surprisingly large number of female names (although it is often frustratingly difficult to find out any details about them). In the nineteenth century, for example, perhaps 1-2% of published music was written by women.1 Whilst that is an embarrassingly small proportion, it still equates to a substantial body of music by many hundreds of women composers – most of whom have since sunk into obscurity. There are of course many more from the twentieth and twenty-first centuries.2 Continue reading →

Why Quantify Music History?

There is a shocking absence of statistics in books on music history. Generations of music historians have shown little interest in using statistical analysis to quantify their subject.

But why should it be considered outrageous that music historians have not embraced the tools and techniques that would enable them to quantify music history? After all, there are plenty of excellent accounts of the history of music, all based on thorough and rigorous scholarship and a deep knowledge of the subject. Is this not enough? Continue reading →