An introduction to spatio-temporal modelling of small-area data in R - Online

Organised by

The University of Glasgow


Professor Duncan Lee


23/06/2021 - 24/06/2021




View in Google Maps  (M139PL)


Claire Spencer -


Spatially referenced data come in many different forms, as observations could relate to either a single geographical location or a predefined geographical areal unit such as a local authority. Examples of the latter include the number of hospitalisations due to respiratory disease in Intermediate Zones in Scotland, or the percentage of children getting 5 or more GCSE grades at A - C in each local authority in England. These data are known as areal unit data, and are found in many fields in social science and beyond. There are many motivations for modelling such data, including identifying areas that are hotspots with high data values (e.g. identifying Intermediate Zones with a high risk of respiratory related hospitalisation), and estimating the impact of covariate factors on the data (e.g. what factors cause local authorities to exhibit poor educational attainment).

Researchers may have access to these data for a single time point, or may have repeated measurements over multiple years. The key challenge when modelling these data is spatial or spatio-temporal autocorrelation, whereby observations relating to nearby areal units or nearby time periods are likely to exhibit similar values. This autocorrelation violates the assumption of independence commonly made by linear regression models, making them an inappropriate tool for data analysis. As such, more complex statistical models are required that allow for such correlations. This course gives a two-day introduction to modelling spatio-temporal areal unit data.

Learning Outcomes

  • Reading in and visualising spatio-temporal data.
  • Quantifying whether a data set exhibits spatio-temporal autocorrelation that needs to be modelled.
  • Modelling spatio-temporal autocorrelation in data and drawing conclusions from the models.



The course will use R and Rstudio for the analysis, and a basic knowledge of R is required (e.g. reading in data, basic plotting, etc). However, full R code will be provided to run the exemplar analyses. The R packages needed include:

CARBayes package version 5.2.3

CARBayesST package version 3.1.1

ggplot2 package version 3.3.2

leaflet package version 2.0.3

maptools package version 1.0-2

RColorBrewer package version 1.1-2

rgdal package version 1.5-16

rgeos package version 0.5-5

sp package version 1.4-2

spdep package version 1.1-5

tidyr package version 1.1.2



Entry (no or almost no prior knowledge)


The fee per teaching day is: • £30 per day for UK/EU registered students • £60 per day for staff at UK/EU academic institutions, UK/EU Research Councils researchers, UK/EU public sector staff, staff at UK/EU registered charity organisations and recognised UK/EU research institutions. • £100 per day for all other participants In the event of cancellation by the delegate a full refund of the course fee is available up to two weeks prior to the course. No refunds are available after this date. If it is no longer possible to run a course due to circumstances beyond its control, NCRM reserves the right to cancel the course at its sole discretion at any time prior to the event. In this event every effort will be made to reschedule the course. If this is not possible or the new date is inconvenient a full refund of the course fee will be given. NCRM shall not be liable for any costs, losses or expenses that may be incurred as a result of the cancellation of a course. The University of Southampton’s Online Store T&Cs also continue to apply.

Website and registration




Spatial Data Analysis, Spatial Data Analysis , , , , Spatio-temporal analysis , , , , Area-based analysis, , , , , Bayesian methods , , , , Exploratory data analysis

Related publications and presentations

Spatial Data Analysis

Back to archive...