Spatial analysis is a useful tool in public health research with increasing amounts of spatial health data generated through population and health facility surveys, routine health management information systems, and clinical trials. While the focus of this workshop is on the application of spatial statistics on health-related data, these techniques also apply in various fields such as geography, spatial epidemiology and agriculture, and environmental sciences. The practical hands-on exercises for these three days’ workshops are tailored to expose participants to real-life examples and areas of application.


This course aims to provide a gentle introduction to spatial statistics and provide the necessary skills to manage and analyze health-related geographical data. 


The specific objectives are to introduce linear and generalized linear models with application to R software, provide an overview of spatial statistics and spatial prediction. The course will also introduce spatial analysis tools, such as Inkscape, ArcGIS/QGIS. 

Teaching methods

The content of this course is best learned through hands-on experience. Sessions will be taught through introductory lectures and presentations of an example dataset(s) related to a relevant public health problem used for the lectures and practical exercises. Participants will be encouraged to work together during practical exercises in a group of 3-4 students. The exercises will allow participants to independently reproduce the R code during lectures aided by hints and full solutions available as part of the course materials. Group presentations will enhance further discussions and clarifications to ensure participants are fully acquainted with course materials.

Learning Outcomes

After completing this course, participants will be able to:

  • Fit linear, generalized linear, and mixed models in R and interpret the outputs.
  • Describe spatial data using maps, perform analysis in R, and correctly interpret spatial analysis results.
  • Analyze patterns in point, area, and field data, learn to detect non-randomness, measure spatial autocorrelation, and create maps.

Course Syllabus

Note: All courses assume participants have basic skills using R software

Day 1: Introduction to linear models.Introduction to generalized linear models.Introduction to generalized linear mixed models.Practical.Day 2: Spatial Statistics: Overview of spatial statistics.Classification of spatial statistical methods.Methods for lattice or areal data.Geostatistics.Methods for point pattern analysis.Methods for lattice data.Geostatistics.Methods for point pattern analysis.Practical.
Day 3: Introduction to spatial prediction.Current approaches to spatial modeling.Useful tools in spatial statistics: Inkscape, ArcGIS/QGISPractical.

Who should take this course?

Postgraduate students (MSc, PhD, and Postdocs), GIS users, data managers/analysts, and researchers who need to create, use, and analyze maps of geographic data. The course is open to participants working in health-related NGOs, academics, ministries, research, and public health institutions. However, this workshop requires participants to have a background in basic regression analysis, particularly linear and generalized linear models. A previous experience using R will be an added advantage. Participants should take note that there will not be any introduction to R in this module.


Lead facilitator/instructor: Dr. Elphas Okango(PhD)

Dr. Okango is a statistician with a BSc. degree in Actuarial Science (First class honor) and an MSc. degree in Applied Statistics from the Jomo Kenyatta University of Agriculture and Technology, Kenya. He also holds a PhD in Statistics from the University of KwaZulu-Natal, South Africa. Dr. Okango has interests in Epidemiology, Biostatistics, Spatial Statistics, Machine Learning, and Artificial Intelligence. Specifically, his area of interest extends to, among others, data mining, predictive modeling, sentiment analysis, disease modeling and mapping, financial data analysis, and data science. In addition, he has vast experience in teaching, research, and coding. Dr. Okango is currently a visiting lecturer at Stellenbosch University, South Africa, and a Post-Doctoral fellow at African Health Research Institute.

Co-facilitator/instructor 1: Dr. Innocent B. Mboya (PhD)

Dr. Mboya is a Lecturer in epidemiology and applied biostatistics at the Institute of Public Health, KCMUCo, Moshi Tanzania. He earned MSc in Epidemiology and Applied Biostatistics at KCMUCo and PhD in Statistics from the University of Kwazulu-Natal, South Africa. He is a member and current president of the International Biometric Society (IBS), Tanzanian region. He has about ten-year teaching experience in addition to coordinating and conducting community-based surveys related to the nutritional status of children under five years of age, adolescent sexual and reproductive health, non-communicable diseases, and monitoring and evaluation of projects at local and national levels. He is experienced in project management from designing, implementation, monitoring, and evaluation. He possesses diverse data management and analysis skills using Stata, SPSS, R, and RDSAT (for data analysis from key populations). His interest is in research and interventions to reduce the burden of adverse pregnancy outcomes, particularly increasing child survival.


This course has been organized by the Department of Epidemiology and Biostatistics, Institute of Public Health, Kilimanjaro Christian Medical University College (KCMUCo). The course is made possible through the PhD funding scholarship for Dr. Ola Jahanpour (PhD candidate at KCMUCo) through the SSACAB scheme.

Date: 28-30 March 2022

Venue: Kilimanjaro Christian Medical University College

Course Fees: Free Course

Interested applicants should send their motivation letter and CV through: iph@kcmuco.ac.tz.