Skip to main content

Music and Science

18 Research questions

18.1 What is a research question?

A research question is, simply put, a particular question about a particular aspect of a research topic. There will typically be many relevant research questions within a particular research topic. Ideally, as time goes on, historic research questions will be answered by new empirical studies, and new research questions will be postulated to replace them. This incremental answering and renewal of research questions through empirical studies is the basic mechanism by which the frontier of scientific knowledge is expanded.

18.2 What makes a good research question?

It’s useful to think about three criteria when evaluating potential research questions.

  • How unknown is the answer to the question? If we know the answer already, then there’s not much scientific value in conducting the study. Of course, there are some nuances to consider here:

    • Knowledge is a continuum. We rarely can say that we ever know anything for certain. However, if the research community is 99.5% certain on something, that does limit the value in testing that research question.

    • There are different kinds of knowledge. In music psychology it’s fairly easy to come up with research questions that already have clear answers outside the scientific domain (e.g. from music theory, or even from intuition), but that have yet to be resolved scientifically. In these cases, it’s important to motivate why you think that the non-scientific consensus might be wrong, and why a scientific study might provide an important alternative perspective.

  • How impactful would it be to answer the question? It is only worth investing the time and resources to answer a research question if we believe that answering this question will have valuable consequences. This could mean improving our understanding of the world (fundamental research), or solving practical problems in the real world (see 17.2.1).

  • How practical would it be to answer the question? Different research questions vary enormously in their practical demands. It’s worth examining the potential cost involved in conducting the required study (see 17.2.5). It’s also worth thinking about the risk involved: are we confident that the experiment will work as expected and deliver a meaningful answer, or is it likely that something will go wrong and the results will be inconclusive?

It should generally be possible to express your research question in a single sentence. If your draft research question is longer than that, try to think about ways you might shorten it. It might feel like shortening a research question means losing important detail, but in practice there are often upsides of this shortening process: it forces you to identify the real essence of what you want to find out, and to express it in a suitably general (i.e. not over-specific) way. Here are some examples of research questions in music psychology:

  • Does musical training increase a child’s visuospatial reasoning skills?

  • To what extent is consonance perception driven by roughness versus harmonicity?

  • Is octave equivalence a cognitive universal?

  • What environmental or genetic factors predispose a child to develop absolute pitch?

  • Why do people enjoy listening to sad music?

18.3 The role of literature reviews

As noted above, the relevance of a research question depends significantly on the prior literature. The prior literature tells us what is known already, and what remains to be found out; it also gives us useful methodological and theoretical precedents that should help us to conduct our own study. A detailed literature review is therefore an essential part of developing one’s own research questions.

There are many electronic databases available nowadays for sourcing literature. One particularly popular source is Google Scholar, which pulls articles from many different journals and databases; other popular sources include Scopus, Web of Science, and PubMed. You can use these databases very much like ordinary search engines: you type in certain keywords, and the software returns various candidate papers, ranked in order of relevance and popularity.

Evaluating literature is a tricky and time-consuming business, especially when one is starting off in the field. Papers can be quite long, and it can be time-consuming to go through in detail to verify the quality of the methods and analyses. To a certain extent, this process cannot be avoided; it is simply a necessary part of what it means to be an academic. However, there are certain external cues that can provide hints about a paper’s trustworthiness. Historically, a particularly highly valued feature is peer review. Peer review is an integral part of modern scientific research, whereby articles submitted to scientific journals are first scrutinised by a panel of (usually) anonymous reviewers, with the goal of identifying methodological and interpretative issues that potentially compromise a paper’s validity. Generally speaking, one can assume that papers appearing in academic journals have been peer-reviewed; in contrast, conference proceedings, books, and contributed book chapters are typically not peer-reviewed. Consequently, journal articles will generally form the bulk of a music psychology literature review, as this literature is considered the most trustworthy. It should be noted, however, that conventions about peer review differ substantially between fields; for example, while peer-review is rare for music psychology conferences, it’s very common for computer science conferences.

Once one identifies a particularly relevant article, there are several useful strategies available for finding more related articles. One approach is to look backwards in time by examining the papers that are cited by the current article. A second approach is to look forwards in time, examining the papers that cite the current article. Modern search engines are particularly helpful for facilitating the second process. Google Scholar, for example, provides a link below each article in the database taking the form ‘Cited by XXX’, where ‘XXX’ is the number of articles that cite the current article. Clicking on this link takes us to a page that lists these articles in descending order of popularity. This can be a very helpful way to work out how the field has progressed since the original paper was published.

This process is a relatively informal and fast way to orient oneself within a research field as a precursor to conducting an empirical study within that field. However, it is also possible to perform a systematic literature review, where the process of collecting, filtering, and summarising literature is highly codified. A related process is a meta-analysis, where one performs statistical analyses to summarise the evidence for a particular claim in the literature. Systematic reviews and meta-analyses can be time-consuming to perform, especially when the pool of literature is large, but they play a valuable role in simplifying busy research fields, and are consequently recognised in the literature as valid publishable studies in their own right.

18.4 Further reading

  • Chapter 1, ‘Research questions’ in Williamon et al. (2021) (eBook link).

References

Williamon, A., Ginsborg, J., Perkins, R., & Waddell, G. (2021). Performing music research: Methods in music education, psychology, and performance science. Oxford University Press.