(Keynote) Telling Stories with Data: Communication in Data Science
Science Samuel Kamande Tweet this!
9:50 - 11:15 (85 mins each)
(BW) Introduction to Data Management in R Christopher Maronga Tweet this!
(IAW) Introduction to Package Development Balogun Stephen Taiye Tweet this!
11:15 - 11:20 (5 mins)
Break
11:20 - 11:40 (20 mins)
(Talk) #TidyTuesday John Mutiso & Maggie Wanjiru Tweet this!
11:45 - 13:10 (85 mins each)
(BW) Introduction to Mapping in R Anelda Van der Walt & Dr. Anne Treasure Tweet this!
(IAW) Modularising shiny apps Dr. Tor-G. Vågen Tweet this!
13:10 - 13:30 (20 mins)
Lunch
13:30 - 14:55 (85 mins each)
(BW) Making my first Shiny app Bryson Mwamburi Tweet this!
(IAW) Introduction to Tidy
Models in R Asmae Toumi Tweet this!
15:00 - 15:40 (40 mins)
(Talk) Create powerful dataviz with R Dr. Ilya Kashnitsky Tweet this!
15:40 - 15:45 (5 mins)
Break
15:45 - 16:30 (45 mins)
(Keynote) Working on projects in R Dr. Maëlle Salmon Tweet this!
16:35 - 18:00 (85 mins)
(Talk) Customizing RMarkdown Dr. Mine Çetinkaya-Rundel Tweet this!
18:00 - 18:10 (10 mins)
Closing Remarks satRday Nairobi Organisers
Team
The people putting it together
Faith Musili
Data Analyst
Christopher Maronga
R Programmer
Njoki Lucy
Biostatistician
John Mutiso
Statistician
Margaret Wanjiru
Risk Analyst
Kigen Tembu
Data Scientist
Shelmith Kariuki
Data Analytics Consultant
Diversity
satRday is dedicated to providing a harassment-free and inclusive conference experience for all in attendance regardless of, but not limited to, gender, sexual orientation, disabilities, physical attributes, age, ethnicity, social standing, religion or political affiliation.
We do not tolerate harassment of participants (including organisers and vendors) in any form. Sexual innuendos and imagery are not appropriate for any conference venue, including presentations.
Anyone violating these rules may be given warning or expelled from the conference (without a refund) at the discretion of the conference organisers.
Our code of conduct/anti-harassment policy can be found here.
Telling Stories with Data: Communication in Data Science
Data storytelling can be posited as the last ten feet of any data analysis and arguably the most important aspect. Your ability to break down your work articulately to various audiences who are looking for marginally or substantially differently packaged outputs for decision making is key. This is the only true value of data analysis to your audience. Based on his experience in both B2C (Agency) and internal analytics teams, Kamande highlights the importance of a stakeholder audience matrix and building up from this in data storytelling for maximum data science value creation.
Formerly at Ajua
Data Sientist
Kamande is a Data Scientist based in Nairobi. He most recently worked at Ajua, the Integrated Customer Experience Platform. In this role, he led a team that provided data support for various product streams and developed new products to drive consumer understanding through advanced analytics and machine learning. He has also been a Data Science curriculum consultant with Moringa School, where after helping develop the curriculum, he now helps to continuously improve and align to the changing African data science dynamics.Kamande previously worked at Nielsen as a Data Science Associate for East Africa. He supported the Commercial team to deliver Quality Retail Data to clients and delivered technical consultancy on Data Science related retail audit queries across FMCG categories. He was also part of the Data Science Global Emerging Leaders program. He holds a Masters’ degree in Statistics and a Masters’ degree in Computational Intelligence from The University of Nairobi. He has published papers on understanding the consumption and viewing habits of the Kenyan consumer and locally available data in peer reviewed journals and presented at various conferences in Africa. He is also NPS certified and facilitates Integrated CX masterclasses in Africa.
Introduction to Data Management in R
In almost every industry these days, huge volume of data is generated at a high velocity. Data is by without dispute, the main building block for scientific inquiry, decision-making and product design; and as such, efficient data management practices become as critical as the generating processes in order to benefit in getting useful insights. Data management is a combination of activities and processes applied to the raw data by means of processing, curating and archiving the final clean flat files for use in an analysis. In this talk, I am going to walk you through a step-by-step process in managing and cleaning your data using R programming language. I will also share some tricks on how to efficiently deal with huge volume of data in the context of data management.
University of Oxford, NDORMS
R Programmer
As an R programmer, Chris works with a team of talented health economist at Oxford University to develop and apply innovative algorithms and scientific methods to process & extract knowledge and insights from health care data. Prior to joining Oxford, he worked as a Clinical Data manager at KEMRI-Wellcome Trust for CHAIN Network study, a multi-country cohort with 9 sites across Africa and East Asia. Chris has a long-run experience in research data management and analysis with over 6 years handling clinical trials data as well epidemiological observational studies. His interests include visualization of spatial data, predictive modelling & data mining, joint modelling, survival analysis, time-series analysis and general statistical computing using R.
TidyTuesday
The #TidyTuesday community continues to grow daily and is open to anyone looking to improve their data wrangling and visualization skills. The project was borne out of the R4DS Online Learning Community and the R for Data Science textbook, with an emphasis placed on understanding how to summarize and arrange data to make meaningful charts with R and various packages within the tidyverse ecosystem. This workshop will cover the steps taken to participate and the benefits and goals achieved in the process. The workshop will also highlight the welcoming inclusive and online community with the help of the hashtag on Twitter with an emphasis on learning or expanding one’s skill set through different data projects posted weekly.
Financial Services
Risk Analyst
Margaret Wanjiru is a risk analyst in the insurance industry that helps the business determine the level of financial risks involved when considering the company’s investments and operational costs. She is also one of the R-Ladies Nairobi Co-organizers. Analytics aside, she is also a nature enthusiast and enjoys cycling and music.
Modularising shiny apps
In this workshop, we will explore how modules can help us better manage Shiny apps when they become a bit large and/or complex. We will develop a simple dashboard with maps, graphics and some simple statistical models/analysis (if time allows). We will be using data from the {rKenyaCensus} R library developed by Shel Kariuki, which can be installed by running the following code in R: devtools::install_github(‘Shelmith-Kariuki/rKenyaCensus’).The workshop will build on a number of R libraries such as {leaflet}, {sf}, {data.table}, {plotly}, {ggplot2} and of course {shiny}. We will use {shiny.react}, {shiny.router} and {shiny.fluent}, which are all developed by the amazing team at Appsilon (appsilon.com). For more information and installation instructions, have a look at https://github.com/Appsilon. The workshop requires some experience with the development of Shiny apps and general experience with R Statistics, including functions. We will develop a simple dashboard live as part of the session, so participants will need a stable internet connection to be able to follow along.
World Agroforestry (ICRAF)
Senior Scientist & Head of CIFOR-ICRAF Spatial Data Science and Applied Learning Lab (SPACIAL)
Tor-Gunnar Vågen is a Principal Scientist and head of the Spatial Data Science and Applied Learning Lab (SPACIAL) at CIFOR-ICRAF. He has a background in soil science, hydrology, landscape ecology, statistical modeling, computer programming, and remote sensing (earth observation). He currently leads research initiatives to establish global data-driven networks of sites for comparative assessments of ecosystem health, linking earth observation data and models to biophysical and socio-economic variables to understand drivers of deforestation, land use change and land degradation, and land restoration potential. He has also been leading several projects and initiatives that explore spatial data science and user-centered design principles in the co-design of interactive tools and platforms for evidence-based decision making.
Making my first Shiny app
Shiny offers data storytellers and end consumers the ability to publish and interact with insightful findings on the fly. In this session, we shall cover an introduction to Shiny alongside a few examples of Shiny Apps.
Kantar
Data Scientist
Bryson Mwamburi is a Data Scientist at Kantar. Kantar is a data analytics and brand consulting company, based in London, England that helps clients understand people and inspire growth. Bryson is a recipient of Facebook’s Secure & Private AI, Intel’s Edge AI scholarships and a Big Data Mastery Award from IBM, he contributes to the client and business success through analytics. Away from wrangling insights, Mwamburi enjoys motorcycling, rugby and swimming, having represented the country in the latter. An avid reader and self-proclaimed master chef, he pairs both with a good glass of wine and the occasional tinge of a sunset.
Introduction to Tidy Models in R
Asmae Toumi will provide an overview of the tidymodels ecosystem, a collection of R packages that have consistent API and are designed to work together in streamlining model building. R packages designed for pre-processing, sampling, model creation, validation and data visualization will be introduced and explored through an interactive end-to-end example.
PursueCare
Director of Analytics and Research
Asmae Toumi is the Director of Analytics and Research at PursueCare, a telehealth company specializing in treating substance use disorder. By night, she is the editor-in-chief of Hockey-Graphs, a hockey sports analytics blog.
Create powerful dataviz with R
Visualizing data is the necessary stage of data exploration that can routinely protect from embarrassing errors. As modern tools place data visualization in the focus of research workflows the need to visualize data is gaining increasing recognition among researchers. However, impressive and efficient graphics are quite rare in the ever-growing universe of dry academic texts. Many of the bad graphical design choices are justified by nothing but the suboptimal defaults and various software legacy issues and mostly outdated limitations imposed by traditional scientific publishers. In this talk I will focus on simple and powerful design choices that help uncover your data and tell your story.
CPop SDU
Postdoctoral Researcher
Ilya Kashnitsky is an Assistant Professor at The Interdisciplinary Centre on Population Dynamics, Odense DK. His current research is focused on the sustainability of pension systems under the changing mortality and the analysis of excess deaths due to the COVID-19 pandemic. Ilya received PhD in Demography from University of Groningen in 2020 defending a dissertation ‘Changing regional inequalities in ageing across Europe’. He is following demography, science of science, research ethics, and dataviz (which he adores, practices, develops, and teaches).
Working on projects in R
A good R project organization is useful even if you do not have collaborators as demanding as The Little Prince. But how to do that? Do you put the project in some sort of box? In this talk , Dr. Maëlle will present tools that are good to know: git, here, renv, targets, orderly… She’ll also report whether making an R package out of everything is always a good idea. After this presentation you will have learnt tips and tricks to apply as you like in your R projects. She hopes that every one of you will learn at least one new thing!
rOpenSci
Research Software Engineer
Dr. Maëlle Salmon is a R(esearch) Software Engineer, part-time with rOpenSci. She also enjoys blogging about R, in particular R package development (on her personal blog and R-hub blog). She is a volunteer software review editor for rOpenSci and the main manager of R-Ladies Global Twitter account; as well as a former R Weekly curator. She’s based in Nancy, France
Customizing RMarkdown
For data analysis output to be reproducible, the data and code should be assembled in a way such that results (e.g. tables and figures) can be re-created. In this workshop we will demonstrate a workflow for reproducible data science with R, R Markdown and discuss how R Markdown documents of various flavours and output types can be customized to achieve a customized look and style. In addition to customization, we will also touch on using RStudio’s visual editor for authoring in R Markdown.
Duke University and RStudio
Professor, Data Scientist and Professional Educator
Mine Çetinkaya-Rundel is Professor of the Practice at the Department of Statistical Science at Duke University and Data Scientist and Professional Educator at RStudio. Mine works on integrating computation into the undergraduate statistics curriculum, using reproducible research methodologies and analysis of real and complex datasets. She also organizes ASA DataFest, works on the OpenIntro project, teaches the popular Statistics with R MOOC on Coursera, and is the creator and maintainer of Data Science in a Box.
TidyTuesday
The #TidyTuesday community continues to grow daily and is open to anyone looking to improve their data wrangling and visualization skills. The project was borne out of the R4DS Online Learning Community and the R for Data Science textbook, with an emphasis placed on understanding how to summarize and arrange data to make meaningful charts with R and various packages within the tidyverse ecosystem. This workshop will cover the steps taken to participate and the benefits and goals achieved in the process. The workshop will also highlight the welcoming inclusive and online community with the help of the hashtag on Twitter with an emphasis on learning or expanding one’s skill set through different data projects posted weekly.
Phastar
Statistical Programmer
Mutiso is a creative and enthusiastic Statistician with experience in clinical health research and the pharma industry. He has experience in statistical data analysis, visualization and reporting using R & SAS. He possess excellent analytical and communication skills and a dedicated approach to working in a controlled environment.
Introduction to Mapping in R
This tutorial will provide an introduction to mapping and spatial data in R using African data. By the end of the tutorial, you should be able to load packages useful in making maps, have an understanding of the different spatial data types, and have plotted your first static and interactive maps with several data layers. We will focus on developing confidence in doing the basics really well in preference to straying too far into more advanced analyses. Our tutorials focus on flexible workflows that you can take away. The training material forms part of the afrimapr interactive learnr tutorials that we have created as part of the afrilearnr package (https://github.com/afrimapr/afrilearnr) described in this blog post : https://afrimapr.github.io/afrimapr.website/blog/2021/interactive-tutorials-for-african-maps/. Due to the time constraint the tutorial will not cover loading your own spatial data, or creating spatial data from data without spatial components. These topics are however covered in other afrimapr tutorials which are available online. There will be no need for learners to install R or RStudio. All learning material will be accessible via a web browser. Participants will need a stable internet connection for watching instructors' shared screens with video to be able to follow along.
Talarify
Anelda runs a small South African consultancy - Talarify - founded in 2014. She supports researchers and students in Africa across all disciplines to adopt aspects of open science and reproducible research to enhance the impact of their research.
Introduction to Mapping in R
This tutorial will provide an introduction to mapping and spatial data in R using African data. By the end of the tutorial, you should be able to load packages useful in making maps, have an understanding of the different spatial data types, and have plotted your first static and interactive maps with several data layers. We will focus on developing confidence in doing the basics really well in preference to straying too far into more advanced analyses. Our tutorials focus on flexible workflows that you can take away. The training material forms part of the afrimapr interactive learnr tutorials that we have created as part of the afrilearnr package (https://github.com/afrimapr/afrilearnr) described in this blog post : https://afrimapr.github.io/afrimapr.website/blog/2021/interactive-tutorials-for-african-maps/. Due to the time constraint the tutorial will not cover loading your own spatial data, or creating spatial data from data without spatial components. These topics are however covered in other afrimapr tutorials which are available online. There will be no need for learners to install R or RStudio. All learning material will be accessible via a web browser. Participants will need a stable internet connection for watching instructors' shared screens with video to be able to follow along.
Talarify
Anne is a research analyst and trainer at Talarify. She has a background in life sciences research and a strong interest in open science, open data and reproducible research.
Introduction to Package Development
The R software is very versatile. It comes with some in-built packages and several thousand additional packages in various repositories that expand its functionalities. Yet sometimes, you might not find the right function for the task you’d like to complete, or the process might be complex and you would want to simplify it with a function. R package allows you the opportunity to organize these functions, validate them, and make them available to others. The {tidyndr} package provides an example of such a specialized package, developed by the author for the Nigeria National (HIV) Data Repository. Recent developments in R have made the process of creating an R package very easy. In this workshop, we will walk through the step-by-step process of package development. We will be learning about: 1.) How to create a simple R package, 2.) How to deploy the package to Github, 3.) How to install the package from Github, 4.) The steps involved in deploying your package to CRAN
Team Lead, Epidemiology and Surveillance Caritas Nigeria CDC 4GATES Program
Stephen Balogun is a Medical Doctor and Public Health professional with over 10-year combined experience in clinical and public health programs. He currently works as the Surveillance Lead for the Caritas Nigeria PEPFAR-funded 4GATES program. In his role, he uses HIV program data to improve the quality of healthcare services in the country. Stephen is also a Fellow and a certified Quality Improvement Expert with the International Society for Quality Healthcare (ISQua). He is the author and maintainer of the {tidyndr} package, a package developed for the analysis of the Nigeria National (HIV) Data Repository (NDR).
