Dr Gillyanne’s research on the female singing voice

Research on the female singing voice

Why a PhD?

I started doing my PhD at the age of 50, after almost 30 years away from academia (my first degree was a BA in music from York university). I’m often asked why I put myself through the experience of becoming a student again when I already had a great teaching career. The answer is always the same – I did not want to see the ‘top of my tree’ – I wanted to grow – and I felt that research into singing voice was the right direction for me.

Doing a PhD is not for the faint-hearted: it requires the ability to think analytically and strategically, to be single-minded and organised, and most of all it requires staying power! I chose the Institute of Education for my place of study because I had seen my main supervisor, Professor Graham Welch, present at a conference, and liked his approach. I knew that he had practical experience of singing and music, not just theoretical and I had two colleagues who were already working with him and who were happy. What I didn’t know was that I would have access to a second supervisor – none other than Professor Johan Sundberg from Stockholm – an amazing stroke of serendipity.

My research questions

My research on the female singing voice was totally empirical, which means that I didn’t set up an experiment (where you take a behaviour or situation and manipulate it), I set up situations in which I could observe, listen to and measure what my singers were doing in relation to my main research question. Building from that point, I asked more questions and created forms of measurement that seemed relevant to the data I’d gathered.

My main question was about how different music genres shape the way female singers use their voices. One of the biggest challenges of devising a research project is clarifying your research question: you think you know what it is you want to find out but you don’t always know what that’s based on.

Unless you are doing theoretical research, your work requires measuring things in relation to each other, demonstrating those relationships in some way and then interpreting what they might mean. That means a lot of reading around the context of your question and – in some cases – taking some kind of ‘baseline’ measure to demonstrate that what you want to look at is, in fact, real.

‘Method’

So as part of my ‘method’:

• I recorded the singers in range tests, in songs and in scale tasks
• I used nasopharyngoscopy to observe some of the singers internally
• I ran perceptual tests where listeners rated the singers according to genre (were they classical or musical theatre?) and also according to comfort (how comfortable did they sound when performing?)

For my analyses I used

• acoustic analysis – spectral and frequency
• interview coding
• observation (of the singers’ larynges)
• also statistical analysis.

What’s the question?

My formal overall question therefore became “How does genre shape the vocal behaviour of female singers?” with the subtitle Empirical studies of professional female singing in Western Lyric and Contemporary Commercial Music Genres.
From my main question I generated four sub-questions:

1. Do singers use their voices similarly or differently in different musical genres?
2. Do singers exhibit a comfortable vocal range?
3. Are some musical genres perceived as more, or less comfortable than others?
4. What is the relationship between comfort and tessiturae?

And if all of that seems wordy, there’s a good reason for it. If you are going to defend your doctoral thesis (and you will have to do that in your viva exam) then you’d better have pinned your questions and sub-questions down really neatly.

Did I find anything out I hear you asking? Yes, I did and I’m going to be doing a series of short articles about each of the questions and what I found out, over the next few weeks.

Was it worthwhile and did I enjoy it? Definitely, and most of the time. Did Jeremy enjoy it? You’ll have to ask him…