Learning outcomes from Analysis in Government Month 2023 live events

This Analysis in Government Month we will be sharing a weekly summary of the learning outcomes from our live events. Make sure to check this page regularly for your event reflections.

Retirement calculator – the power of Python  

  • Efficiency of Integrating Frontend and Backend Technologies: The combination of HTML/CSS/JavaScript for the frontend and Python for the backend provides a complete, efficient, and robust framework for web application development.
  • Improved Analysis Through Interactive Visualisation: JavaScript libraries (like React.js) can be used to create interactive and dynamic visualizations, which enhance the understanding and engagement of end users with the data analysis.
  • Data Manipulation and Analysis with Python: Python’s rich ecosystem of libraries such as Pandas and NumPy enable complex data manipulation and analysis.
  • Real-time Data Processing: The Python backend can process and update data in real time, which can be reflected immediately in the frontend, giving users the most up-to-date insights. 

Making charts accessible to all

If you enjoyed the session and want to find out more, or if you could not make the session, here are some suggestions for further learning and support:

The slides from the session (PPTX, 31MB) are available and we have made them as accessible as possible, but they do not contain all the information from the session. We recommend you look at our accessibility guidance and support for more information.

The team run drop-in clinics about communicating analysis every other Wednesday at 10:30am. Visit our events page to find out details about the next communicating analysis clinic.

Developing monitoring data collection systems and visualisation dashboards in the Department for Business, Energy and Industrial Strategy (BEIS)

Attendees learned about:

  • overcoming the challenges in developing and implementing a new digital data collection system, and of the importance of early engagement with a wide range of stakeholders
  • how the team developed informative visualisations in Power BI managed the messaging
  • how data was used to influence policy delivery in-flight, and to provide sound financial management of taxpayers’ money

The impact of government analysis: panel discussion

The panel featured:

  • Professor Sir Ian Diamond, Head of the Analysis Function and National Statistician
  • Jane Whittaker, Director Knowledge and Information at HM Revenue and Customs (HMRC)
  • Esther Sutherland, Head of AF Central Team and ex-COVID-19 Task Force at the Office for National Statistics (ONS)
  • Annette Cox, Head of Central Research, Evaluation and Monitoring at the Department for Business and Trade

The panel agreed that:

  • engaging with stakeholders is essential to help understand the problem you are trying to solve
  • impactful analysis always has a clear link to the direct impact on citizens
  • it is important to always communicate your analysis clearly and with context to the right person, in the right way, and at the right time

Influencing Influential Analysis

Delegates contributed to a lively, interactive discussion on how they influence analysis in their departments. The majority of attendees said that their department is supportive of evidence-based decisions.

Attendees heard case studies highlighting the work the Analysis Function are doing to influence influential analysis. Further updates on this work will be published on the Analysis Function website.

Devolution for analysts

The session:

  • increased understanding of the UK data landscape
  • highlighted the importance of understanding work being undertaken from across the UK on the statistical theme analysts are working on — read more about the coherence of statistics 
  • highlighted the importance of working with colleagues across the statistical system and embedding the wider narrative within statistical outputs — read more about exploring the user need for coherent data across the UK
  • demonstrated the structures and tools available to encourage positive working relationships and support cooperation between the UK Government and the devolved administrations — you can find out more about this in the Concordat on Statistics

How the Office for National Statistics developed a cost of living insights tool

The session demonstrated that:

  • the cost of living is a complex inflationary crisis
  • users need up to date, trustworthy data and insights
  • insights tools are more sustainable when automated
  • insights tools need continuous communications, promotion and horizon scanning
  • insights tools are the future of ONS content

You can find the Cost of living insights tool on the ONS website.

Create your own Census datasets

Attendees were shown how to create their own custom datasets using the web tool.

Data visualisation in R using gg plot2

The session helped delegates understand:

  • how to structure basic ggplot2 code
  • the main components of ggplot2 code
  • the use of geoms and themes
  • how to specify graph labels, axes and titles
  • how to apply ONS accessibility guidance to your ggplot2 graphics

You may find the following resources useful to develop your knowledge:

Introduction to the data science masterclass

You can find more information about the Data Masterclass on the Data Science Campus website.

If you have any questions or if you are interested in hosting a cohort in your department, please email the team at Masterclass@ons.gov.uk. The team can connect you with your department champion and help you facilitate the course in the best way.

Disaster risk financing: how analysis can contribute to timely responses  

The session demonstrated that:

  • Pre-planned Disaster Risk Financing (DRF) can reduce the impact on people of disasters, and newer techniques like anticipatory action can mean actions can be taken before the disaster even hits. The cost of response can also be reduced by responding more quickly and in pre-planned ways aligned to the needs of those affected. This is supported by a range of analytical approaches including use of data science, satellite monitoring, catastrophe models and more. 
  • How impact can be made by using DRF to get the right resources at the right time to people who need it in the aftermath of a disaster  

The following resources can help you learn more about Disaster Risk Financing:

What makes a good prison? 

A true reflection of customer demand in DWP  

  • Stakeholder management: Engage early and engage often with an open mindset. This allows you to fully understand your stakeholder’s perspectives and needs before you start offering solutions. When you do offer solutions, help them understand the value and impact of your work from their perspective.
  • Data Visualisation: Use text narrative boxes to tell a story and chart annotation to guide the viewer. Think about your audience and their level of knowledge, avoiding technical jargon. Lots of helpful guidance on the Analysis Function webpage: Data visualisation – Government Analysis Function (civilservice.gov.uk)
  • Problem Structuring: understand who your stakeholders are (cast the net wide) and use them to check your understanding of the problem, wider context and your proposed solution. Useful tools from the Government systems thinking toolkit: Systems thinking for civil servants – GOV.UK (www.gov.uk) 

Error and fraud in the COVID-19 Job Retention Schemes 

  • Analysts on the team learnt to really interrogate the data. Analysts should find out as much as possible about their data from those that collected it (or as close as possible). In doing so, analysts will understand the caveats of how it was collected and the implications for which model to use, how to interpret results, and build wider trust in their analysis. 
  • Using a variety of different techniques that all came together was incredibly rewarding. Our analysts used techniques, some of which were devised in the team and others not widely used within HM Revenue and Customs, which are now being employed in other areas within the department. Our experience was a great testing ground for techniques which can bear fruitful results elsewhere within the department which were only possible through our work.  
  • Every analyst can play their part. There were many, deeply complex analytical components that analysts worked on and each component required enormous effort from analysts to complete, sometimes only constituting a small part of the estimate. The value of the end product is only as good as every component. 

UK climate change project: A cross-government initiative 

  • Participants learnt about the cross-Government Climate Change Project, including the UK Climate Change Statistics Portal
  • Participants learnt about the benefits of 5-star linked data and how we are using it in the Climate Change Statistics Portal. Google “csvcubed” to find documentation and training videos on how to create linked data CSV-Ws
  • For further information or feedback: Climate.Change@ons.gov.uk 

Hackathon 3.0 

  • Participants were challenged in groups with similar interests and skills, to put forward a research question related to the cost of living and look to carry out appropriate analysis to draw meaningful conclusions.
  • There was a huge range of topics, data, tools and techniques used over the course of the hackathon, many of which were new to them.
  • You can check out some of the datasets used in the Hackathon through this Cost of living latest insights article and the Open Geography Portal
  • The Data Science Campus website has lots of great resources if you want to learn more  

Analysis Skills for All Civil Servants 

If you are interested in increasing your analytical capability, you  can explore the following learning resources on the Analysis Function website:

Accessibility testing of data dashboards and interactive content 

If you enjoyed the session and want to find out more, or if you could not make the session, here are some suggestions for further learning and support: 

The team also run drop-in clinics about communicating analysis every other Wednesday at 10:30am. Visit our events page to find out details about the next communicating analysis clinic. 

There is also a cross-government accessibility community you can join.  

Or you can join the team’s Basecamp projects for Making analysis accessible or for Presentation and Dissemination of data statistics and analysis. Email Analysis.Function@ons.gov.uk for an invite. 

Introduction to Theory of Change

Slides from the session Analysis in Government Month 2023_An introduction to Theory of Change presentation are now available.

Explore the following Analysis Function website pages to learn more about Theory of Change:

The team also hold Theory of Change support workshops if you require further support.

Introduction to the cross-government Data Science Community 

How can the Geography Profession help you? 

  • Geography underpins a lot of the work across the public sector and throughout the Analysis Function.   
  • Geographic principles can be found across a variety of guidance documents used by analysts such as the Code of Practice for Statistics, the Green Book, the Magenta Book and will in the future be part of the Aqua book.   
  • Many members of the Geography Profession are dual badged, with other analytical professions and have geography as a second profession. You can join the Government Geography Profession on the member site. 

Armchair Analysts: Using Analysis in Operational Delivery 

The session gave attendees a better understanding of how to present data in an effective and clear way. Learning what pitfalls to avoid when creating charts, graphs and data visuals. 

There is more information on the Analysis Function website if you can explore the topic further:  

Communicating Analysis Effectively – Three Perspectives 

Communicating analysis effectively three perspectives slides-to-share

Professional Development 

Useful links to further learning from this session:

There were several queries about getting access to the Online Skills Tool. Please email Analysis.Function@ons.gov.uk and we will give details of individual profession leads where we can. 

Own your development by:

  • Making sure your CV is up to date and representative
  • Focusing on your strengths and and areas for development
  • Considering what you can learn from job shadowing or mentoring
  • Talking to your line manager about your personal development plan
  • Exploring the Career Framework and Civil Service Jobs; think about where you are heading
  • Establishing positive routines and dedicated time for learning
  • Getting involved with networks and social learning
  • Thinking about what you can share with others

Improving forecasting accuracy: How analysis can add value to finance 

Attendees learned how: 

  • Systematic bias analysis was applied across the different Directorates General comprising the Department for Work and Pensions (DWP). 
  • The combination of a systematic bias score with additional statistical data on the size of the bias helped to identify areas of the business that required additional help with preparing accurate forecasts. 
  • Systematic bias analysis was also used to estimate the year-end financial position (over and underspend) of the department. 
  • The work analysts have done in DWP Finance Group has resulted in a step-change improvement in DWP’s financial forecasting ability. 

Lessons from global meltdowns and how to prevent them

Grant Fitzner, Chief Economist and Director in the Office for National Statistics hosted this event featuring author Linda Yueh talking about the topics in her book, The great crashes; Lessons from global meltdowns and how to prevent them.

The session is available to watch on YouTube until 7 July 2023.

In conversation with Sir Ian Diamond and Tamara Finkelstein 

  • Head of the Analysis Function, Sir Ian Diamond and Head of the Policy Profession, Tamara Finkelstein talked about the importance of bringing policy and analysis together 
  • Analysis has the biggest impact when we work together, Sir Ian said that “Communication in a critical way is essential to good partnership”. They reflected on their own careers and gave examples of how successful analysis can influence policy 
  • To connect with colleagues, you can sign up for monthly coffee connect, an informal networking session between policy professionals, analysts and members from Go-Science profession 
  • Complete the analytical literacy learning on the Policy Profession website for further learning
  • Explore the analytical awareness learning pathway for non-analysts which includes bitesize learning videos on core analytical concepts as well as a more comprehensive set of learning modules to build analytical literacy 

Machine learning for financial forecasting 

Achieving real impact with machine learning requires combining the statistical rigour and quality assurance expected of all analysts, with a real focus on explaining how our models work, and really taking time to understand our customer’s needs and workflows to ensure our analysis can integrated at the right point. 

The session demonstrated how to use the Random Forest algorithm.