SDG Dashboard: Clean Cooking & Child Mortality

The question

The SDG Hub@BUas asked us to use public Sustainable Development Goal data to investigate a data-driven research question and present the results in Power BI. I focused on access to clean cooking facilities and child mortality.

The project started from the observation that households without clean cooking facilities often rely on solid fuels, which can contribute to indoor air pollution. I wanted to investigate whether countries with lower access to clean cooking facilities also tended to show higher child mortality.

How I approached it

Preparing the data

I gathered public datasets covering access to clean cooking facilities, child deaths and mortality, indoor air pollution, pneumonia and related risk factors. Most of the variables covered many countries over roughly two decades, with several datasets spanning 2000 to 2020 or longer.

In Power Query, I removed empty or irrelevant values, corrected data types, renamed unclear columns, created additional fields where useful and merged datasets that could be analysed together. The resulting Power BI model connected multiple related datasets rather than relying on one prepared table.

Exploring the trends

I first compared how access to clean cooking facilities and child mortality changed over time. The dashboard used interactive filters and country selections so the same views could be explored for different parts of the data.

Measuring the relationship

The central analysis compared child mortality rates with access to clean cooking facilities using a scatter plot, correlation coefficient and a fitted regression line. The dashboard showed a correlation coefficient of -0.86, indicating a strong negative relationship in the selected data: countries with greater access to clean cooking facilities generally showed lower child mortality rates.

This analysis describes an association in the data. The correlation and regression line do not by themselves establish that one variable causes the other.

Investigating the relationship further

After finding the strong correlation, I explored related data to understand what could be behind the pattern. I looked at indoor air pollution from solid cooking fuels and at pneumonia , which the source material identified as a major cause of child mortality.

A second analysis compared clean cooking access with deaths attributed to indoor air pollution from solid cooking fuels. This view showed a correlation coefficient of -0.93. Together with the pneumonia and risk-factor data, this provided additional context for interpreting the original relationship, while still not proving causation on its own.

What I found

The dashboard showed a strong relationship between lower access to clean cooking facilities and higher child mortality in the selected data. The follow-up analysis also showed a strong relationship between clean cooking access and deaths attributed to indoor air pollution from solid cooking fuels.

The project led me to connect these findings with the wider evidence on indoor air pollution and pneumonia. In the original dashboard, I used that context to explain why clean cooking access could matter for child health. Looking back, I would distinguish more explicitly between a strong statistical relationship and evidence of causation.

Limitations and further analysis

I deliberately kept the dashboard focused rather than adding every possible factor. Variables such as GDP per capita and industrial air pollution were outside the scope, and some datasets described the total population rather than children specifically. I also analysed the general international relationship instead of focusing on one country or region.

For further analysis, I would include additional socioeconomic and environmental variables and test whether the relationships remain after controlling for those factors.

Presenting the work

The final deliverable was an interactive Power BI dashboard and a seven-minute live demonstration for the SDG Hub@BUas client, staff and fellow students. The dashboard was designed to guide the viewer from the research question and exploratory analysis through the main relationship, deeper investigation, discussion and conclusion.

Project resources

The Power BI dashboard and the main public data sources used in the analysis are available below.

The analysis used public data from Our World in Data, drawing on sources including the World Health Organization and United Nations datasets. Full references are included in the Power BI dashboard.