Using data to inform air pollution policy in Ulaanbaatar

May 13th - May 17th 2019



Information

This event took place in the Library Building at the National University of Mongolia and was jointly organised with the University of Bath.

Room 502, Library Building,
National University of Mongolia,
Ikh Surguuliin gudamj -1,
Baga toiruu,
Sukhbaatar district,
Ulaanbaatar city
View on map >>>


Building on a 3 year collaboration between the Institute of Mathematics at the National University of Mongolia, and the Department of Mathematical Sciences and the Institute for Policy Research at the University of Bath, we hosted a workshop to explore how the issues of economic development, migration, and expansion of infrastructure in Mongolia relate to the ongoing challenges of air pollution.

The workshop has two goals:
  1. To collaborate with people from agencies across Mongolia to explore how using data science can help inform understanding and analyse the way in which air pollution relates to key sustainable development goals.
  2. To give participants the opportunity to take part in training in basic data science using data related to air pollution.

As an output from this workshop we hope to develop a series of questions that could be used to guide the collection and collation of a robust data set for future collaborative work and provide the basis for funding applications from UK and international agencies.

The workshop consisted of two interlocking parts which bring together the conceptual and the practical elements of data analysis for problem solving.

During the morning sessions we explored data collection methods, discussed how data can influence the way we think about policy questions and considered the design and evaluation of possible interventions. These group discussion sessions were hosted by a diverse academic panel, including leading economists, social scientists and policy researchers from the UK and included the chance to learn from global case studies.

The afternoon sessions were hosted by statistical experts from the University of Bath. Building on the morning discussions we worked on data related to air pollution from a variety of contexts. Some participants brought their own data to explore with the assistance of a statistician. These sessions covered the basic skills of data science using the free software package 'R' (including reading in data, manipulating and visualising data), through to more advanced skills in statistical inference such as some regression techniques.

Registration

To register for the data sprint, please complete the registration form available here.

Schedule

Friday Schedule

Installing the R software

A key part of the data sprint will be learning how to understand and apply statistical techniques to air pollution data. For this you must bring along a laptop with the R and RStudio software packages installed, instruction for how to do this can be found here.

It is important that you follow these instructions to install R before the statistics training as we will not have time during the session for this.

Course Material

Presentations

Welcome
Workshop Overview
Air Pollution Case Studies
Causation from a statistical perspective
Air Pollution Survey
Presenting Results to Policy Mkers
How to Write a Journal Article

Data Analysis Training

Monday R material
Tuesday R material
Wednesday R material
All R material (Mon-Thurs)

Survey Analysis R Material with solutions

Monday R solutions
Tuesday R solutions
Wednesday R solutions
All R solutions (Mon-Thurs)

Book Recommendations

R website recommended literature

Participant Survey

We would like to get your views about air pollution in Mongolia. Please complete this survey. Around 500 residents of UB have completed this survey in the last week and we will be looking at this data later in the week and discussing some of the results of the survey. Thank you!

Click here to access the participant survey

Survey Questionnaire

Mongolian Survey Questions

Give us feedback

We would appreciate it if you could provide some feedback on your experience here.

Please click here and enter the code 677395 to access the feedback form.

Contact Us

If you want to get in touch or have any questions regarding the data sprint you can email Susie Douglas.