[X]

CSIR set to forecast US elections using world-class prediction model

Publication Date: 
Tuesday, November 3, 2020

The CSIR is putting together a forecast of the upcoming US presidential elections to conduct a successive test of the performance of its election prediction model into the US electoral system. This follows a successful trial during the US 2016 presidential elections.

The CSIR’s election prediction model relies on two core principles relating to voting behaviour of the electorates and the order in which voting results are declared on the night of the Election Day. These two principles combined allow the CSIR team to group voters (or rather voting districts) together based on their past voting behaviour (using a statistical clustering method) and to then expect that any changes to voting behaviour in the new election will be fairly similar within each group.

Contact Person

David Mandaha

+27 (12) 841 3654/072 126 8910

dmandaha@csir.co.za

The Council for Scientific and Industrial Research (CSIR) is putting together a forecast of the upcoming US presidential elections to conduct a successive test of the performance of its election prediction model into the US electoral system. This follows a successful trial during the US 2016 presidential elections.

The CSIR’s election prediction model relies on two core principles relating to voting behaviour of the electorates and the order in which voting results are declared on the night of the Election Day. These two principles combined allow the CSIR team to group voters (or rather voting districts) together based on their past voting behaviour (using a statistical clustering method) and to then expect that any changes to voting behaviour in the new election will be fairly similar within each group.

“We are demonstrating our capability as the CSIR to use science and technology to predict the outcome of the elections long before the results are officially declared. Voters do not randomly allocate their electoral preferences but are influenced by political, socio-economic and demographic factors, as well as past voting history. Further, the manner in which the results are reported also does not occur in a random fashion. The CSIR’s election prediction model provides the window of opportunity for the model to provide insightful information for political analysts and the public to ponder on while they await the final result,” says Dr. Paul Mokilane, CSIR Acting Research Group Leader: Data Science.

The original model was developed, by the CSIR, for the South African elections and has been applied during the 1999, 2004, 2009, 2014 general elections and the 2000, 2006, 2011, 2016 municipal elections. “The success of the model in predicting the final result with a small error margin (usually with a percentage point away from final result) for the South African elections, including the 2016 municipal elections led to an interest in testing the model outside South Africa, in different electoral systems,” says Dr Mokilane.

In 2016 the CSIR model correctly predicted that President Donald Trump would win and the results of 44 states out of 56 states, one federal state of Washington DC and major islands and territories.

For this year’s election the CSIR will provide a live forecasting of the US Presidential elections taking place on 3 November 2020. The forecast will be accessible on the CSIR website from as soon as the results become available on the Politico website on 3 November 2020.

Issued by the CSIR Strategic Communication unit

Enquiries:
David Mandaha: CSIR Media Relations Manager
Tel: 012 841 3654
Mobile: 072 126 8910
E-mail: dmandaha@csir.co.za
 

About the CSIR:
The CSIR, an entity of the Ministry of Higher Education, Science and Innovation, is one of the leading scientific and technology research, development and implementation organisations in Africa. Constituted by an Act of Parliament in 1945 as a science council, the CSIR undertakes directed and multidisciplinary research and technological innovation, as well as industrial and scientific development to improve the quality of life of all South Africans. For more information, visit www.csir.co.za.

Follow us on social media:
Twitter: @CSIR. Facebook: CSIRSouthAfrica. Instagram: CSIRSouthAfrica. LinkedIn: Council for Scientific and Industrial Research (CSIR). Youtube: CSIRNewMedia