Making your mind up: how the order of performances impact Eurovision outcomes

And now for something completely different…

It is nearly time for the annual Eurovision Song Contest (ESC)! Will it be “Euphoria” or “Only Teardrops” for the UK’s entry Mae Muller? Raych George analyses the data to discover whether the order of performances could hold the answer, showing how analysis has an impact on all areas of life.

The basic concept of Eurovision has remained the same for many years. Each participating country sings a song, live, and other countries give the song points. But the rules of the competition have varied across the years. There have been changes to the rules around the languages that can be used for songs and changes to the voting system. But does the order in which countries perform impact the results?

What we mean by “order effects”

Order effects refer to the phenomenon that the “order” in which songs are performed may influence the way people respond to them. This may impact the outcome of the competition. Examples of the order effects phenomenon in everyday life include things like:

• how the order of interview candidates may affect who is offered a job
• how the order of questions in a survey can affect responses

How order effects may influence Eurovision scores

Marco Haan, S. Gerhard Dijkstra and Peter Dijkstra concluded in 2005 that performance order has had an effect on the final outcome of Eurovision contests over the years. Artists that performed later in the contest generally received higher scores.

This corresponds with other research on performance evaluation, such as the research from Lionel Page and Katie Page in 2010. But the advancement of data availability has allowed Evgeny Antipov and Elena Pokryshevskaya to investigate order effects using a more comprehensive data set.

Good Data Management has meant that it has been possible to analyse data by breaking down the public and jury votes for all countries. In the past data was unavailable for songs outside the top 10. This means all other countries were given a score of “0”, but this does not mean all these songs were equally rated.

Data was used from between 2009 and 2012. This is because the producers decided the running order of the contest after 2012, rather than using a random ballot.

The analysis by Antipoc and Pokryshevskaya showed there was some limited evidence of order effects in the ESC when it came to the professional jury. There was no obvious order effect when it came to the public vote.

While the significance of order effects at the ESC is unclear, the question remains whether you, as a viewer, prefer the songs performed towards the end rather than at the beginning? Do you think the order of the performances should be reversed between the “Jury Final” and the “Live Final” to help reduce any order effect for the public voting?

One thing is for certain: as Stephen Few (the Business Intelligence expert) states “numbers have an important story to tell and it is up to you to give them a voice”. I will be watching the ESC final with utter enthusiasm and will enjoy seeing the data flow in to help understand these phenomenons more.

References

Haan and others. ‘Expert Judgment Versus Public Opinion – Evidence from the Eurovision Song Contest‘ Journal of Cultural Economics 2005: volume 29, pages 59 to 78 (viewed on 24 April 2023)

Page and Page. ‘Last shall be first: A field study of biases in sequential performance evaluation on the Idol series‘ Journal of Economic Behavior and Organisation 2010: volume 73, issue 2, pages 186 to 198 (viewed on 24 April 2023)

Antipov and Pokryshevskaya. ‘Order effects in the results of song contests: Evidence from the Eurovision and the New Wave‘ Judgement and Decision Making 2017: volume 12, pages 415 to 419 (viewed on 24 April 2023)