Skip to main content

Learning Path: Data Journalist

A six-stage structured path from spreadsheet fundamentals to statistical analysis — tailored for UK journalists working with ONS data, government open data, and UK public registers.

Last reviewed: Next review due:

Who this path is for

This learning path is designed for journalists who want to work with data — whether as a specialist data journalist, as an investigative reporter who uses data as one tool among many, or as a general reporter who wants to be able to interrogate datasets without relying on a specialist colleague.

The path is sequential: each stage builds on the previous one. Do not skip spreadsheets even if you have some SQL or Python experience — the conceptual foundations of data journalism (finding the story, cleaning the data, checking your working) are best learned with tools that expose the data directly.

The six stages

Stage 1

Spreadsheets

Master Excel and Google Sheets before anything else. Data journalism starts with structured data in rows and columns.

  • Sorting, filtering, and pivot tables to find patterns in datasets.
  • Formulas for calculating percentages, rates, and year-on-year change.
  • VLOOKUP and INDEX/MATCH for combining data from different sheets.
  • Cleaning dirty data: removing duplicates, standardising formats, splitting columns.
  • Understanding the difference between absolute and relative cell references.

The ONS publishes detailed datasets in Excel format. Practice by downloading a dataset from data.gov.uk and trying to find a story in it.

Stage 2

SQL

Structured Query Language lets you query databases that are too large for spreadsheets and is widely used in data-driven newsrooms.

  • SELECT, FROM, WHERE, GROUP BY, ORDER BY — the building blocks of SQL queries.
  • JOIN to combine data from multiple tables on a shared key.
  • COUNT, SUM, AVG and other aggregate functions for group-level analysis.
  • Filtering with HAVING for aggregate conditions.
  • Writing queries against public UK government databases and APIs.

SQLiteOnline.com allows you to practice SQL in a browser without installing anything. Start with a small, clean dataset from data.gov.uk.

Stage 3

Python for Journalism

Python is the most-used language for data journalism. It handles large datasets, web scraping, and statistical analysis.

  • pandas for loading, cleaning, and analysing structured data.
  • requests and BeautifulSoup for scraping public websites and APIs.
  • matplotlib and seaborn for exploratory data visualisation.
  • Jupyter Notebooks for reproducible, documented analysis.
  • Working with UK government APIs including Companies House, ONS, and data.gov.uk.

The Programming Historian publishes free, peer-reviewed Python tutorials for humanities and journalism researchers.

Stage 4

Data Visualisation

A chart that is wrong or misleading can harm your credibility as much as a factual error in text.

  • Choosing the right chart type: bar charts for comparisons, line charts for trends, maps for geography.
  • Avoiding chartjunk, dual axes, and truncated y-axes that distort comparisons.
  • Creating accessible charts: colour contrast, alt-text, pattern fills for colour blindness.
  • Datawrapper and Flourish for publication-ready interactive graphics.
  • Responsive chart design for mobile-first newsrooms.

Datawrapper Academy publishes free guides on chart selection, design principles, and common mistakes.

Stage 5

Mapping

Geographic data adds a dimension that text alone cannot convey. UK-specific geographic data is freely available.

  • ONS geographies: Output Areas, LSOAs, local authority boundaries — understanding the hierarchy.
  • QGIS for desktop geographic analysis — free and open-source.
  • Datawrapper and Flourish for publication-ready choropleth and point maps.
  • Geocoding addresses for mapping incidents or planning applications.
  • OS Boundary-Line for local authority, parliamentary, and ward boundaries.

The ONS Geoportal provides free UK boundary files for every geographic level. The Ordnance Survey provides OS Open Data under the Open Government Licence.

Stage 6

Statistics for Journalists

You do not need a statistics degree — but you do need to understand the concepts that underlie data journalism claims.

  • Mean, median, and mode — and when each is the right measure of average.
  • Percentage change vs. percentage points — a frequently confused distinction.
  • Correlation vs. causation: why a relationship in data does not prove one causes the other.
  • Statistical significance and sample size — when can you generalise from a survey?
  • ONS methodology notes: how to read and understand the methods behind a dataset.

The Royal Statistical Society publishes a free guide to statistics for journalists. The ONS Methodology Advisory Service provides guidance on interpreting official statistics.

Essential UK data sources

ONS
Population, economy, health, labour market, and regional statistics.
data.gov.uk
Aggregated open data from central and local government.
NHS Digital
Health and social care statistics, hospital activity, waiting times.
Companies House bulk data
Full company register download, officer data, PSC data.
Ministry of Justice statistics
Courts, sentencing, prisons, probation, legal aid.
Environment Agency
Flood risk, water quality, emissions, environmental permits.

Related guides

Data JournalismData Journalism HubRead →TrainingLearning Path: InvestigativeRead →TrainingNCTJ Readiness Self-TestRead →CareersNCTJ Diploma GuideRead →BlogJournalism Courses UK GuideRead →InvestigativeInvestigative Journalism HubRead →

Primary sources

Frequently asked questions

Do I need to learn to code to be a data journalist?
Not necessarily. Many data journalists work primarily with spreadsheets, OpenRefine, and tools like Datawrapper without writing any code. However, learning Python or R substantially expands what you can do — particularly for working with large datasets, web scraping, and automating repetitive analysis. Start with spreadsheets and learn scripting when you hit the limits of what manual tools can do.
What is the NCTJ data journalism module?
The NCTJ introduced a data journalism module as part of its Diploma in Journalism to reflect the growing importance of data skills in UK newsrooms. The module covers data literacy, working with spreadsheets and databases, understanding statistics, and data visualisation. Check the NCTJ website for current module specifications and accredited providers that offer it.
Which UK data sources are most useful for data journalists?
The Office for National Statistics (ONS) is the primary source for UK population, economic, and social data. data.gov.uk aggregates open datasets from hundreds of UK public bodies. NHS Digital publishes health statistics. The Ministry of Justice publishes courts and crime data. Companies House bulk data is useful for corporate investigations. The Environment Agency publishes flooding and environmental data.
What visualisation tools do UK newsrooms use?
Datawrapper is widely used in UK newsrooms for charts and maps because it produces clean, accessible, responsive graphics without coding. Flourish is popular for interactive narratives and map stories. The BBC uses its own Responsive News Charts tool internally, but many BBC journalists also use Datawrapper for quick turnaround. Google Charts and Highcharts are used in some legacy systems.
How do I verify data before publishing?
Data verification involves checking the source (official vs. third-party aggregator), understanding the methodology (how was it collected?), checking for known errors or revisions, running sanity checks on totals and percentages, and where possible reproducing key calculations independently. The ONS publishes methodology notes for all its datasets — always read these before drawing conclusions.