Reporting With Your Own Data: What Mona Chalabi's Ten Lives Teaches Student Journalists

How can student journalists combine personal data with reporting to make abstract issues concrete? Reporting With Your Own Data offers one answer, modeled on the Guardian's review of Mona Chalabi's first book Ten Lives, which examines her finances alongside those of ten others from across the economic spectrum and argues that our collective ignorance is sustaining an unfair system. The review describes the result as a mix of journalism, data visualisation and memoir aimed at our silence around income. We suggest four exercises, each a hypothesis to test, for building the same habit at student scale.

Stage One: Write a definition for every number you own

The Guardian's review of Ten Lives describes Mona Chalabi, a former Guardian data journalist and illustrator, putting her own finances under a microscope alongside those of ten others from across the economic spectrum, and quotes her argument that 'our collective ignorance is sustaining an unfair system.' The review frames the book as a mix of journalism, data visualisation and memoir, aimed at our silence and furtiveness around individual income and finances. How a single figure survives that kind of scrutiny is not something the review's summary shows, so we offer an interpretation for students to test: a number you want to publish, even one from your own bank statement, needs a written definition before anyone else can examine it. Stage One therefore starts before reporting, with the numbers you already own.

Our suggested exercise runs for one week: log every personal figure you might cite in a story, such as monthly rent, transit spending, income or savings, and for each entry record where it came from, what time period it covers, and a one-sentence definition of what counts toward it. Two constraints apply. Use only figures you can trace to a document, not estimates from memory, and keep each definition to a single sentence, because a definition that needs qualifying paragraphs is not yet a definition. Then swap sheets with a partner, whose only task is to re-derive each figure from your definition alone. We offer this as a hypothesis to test: figures with written definitions should survive partner re-derivation more often than figures without. Run both conditions on ten numbers each and compare the pass rates before deciding whether the discipline is worth keeping.

Stage Two: Predict your rank before you see the spread

The review of Ten Lives describes Mona Chalabi, a former Guardian data journalist and illustrator, putting her own finances under a microscope alongside those of ten others from across the economic spectrum, in a book it calls a mix of journalism, data visualisation and memoir. The excerpt does not show how that comparison is conducted, so what we suggest students borrow is the structure alone: one writer's own figure placed inside a set of others. Our interpretation, offered as a hypothesis to test, is that writers misjudge how typical their own situation is, and that this misjudgment can be measured. Ask a writer to predict where their own value will rank before seeing anyone else's number, then compare the prediction with the actual distribution; the gap between the two is the observable evidence.

Our suggested exercise scales the structure to a classroom. Recruit six to ten consenting classmates or neighbors from different circles, agree on one identically defined metric such as weekly food spending, and collect anonymized values. Three constraints apply. Every participant gives explicit consent before any number is collected; values circulate only in anonymized form; and the metric is defined identically for everyone, since values gathered under different definitions would not be comparable. Before seeing anyone else's number, each writer records where they predict their own value ranks, then compares prediction against the actual distribution. To evaluate the result, list each participant's predicted and actual rank and note the size and direction of every gap; if predictions cluster near actual ranks across the group, the hypothesis fails for that group. The deliverable is the anonymized dataset plus a short paragraph stating only where your value sits.

Stage Three: One mark per case, caption second

The review describes Ten Lives as a mix of journalism, data visualisation and memoir, by Mona Chalabi, a former Guardian data journalist and illustrator, arguing that 'our collective ignorance is sustaining an unfair system.' The excerpt does not show how her charts are constructed, so the framing we suggest students borrow is our own proposal rather than one the review makes: treat a chart as a claim to be checked, not a decoration to be styled. Our interpretation, offered as a hypothesis to test, is that a caption can assert more than the plotted values show, and that the mismatch becomes detectable when a reader describes the bare chart before seeing the caption. If their one-sentence description reaches a different conclusion than the caption states, then, on this hypothesis, the caption is claiming something the marks do not.

Our suggested exercise works on the dataset from earlier stages, such as the anonymized spending values collected from consenting classmates. Draw a unit chart or dot plot by hand, one mark per participant, and add no color, labels or styling until the shape of the points is fixed; the constraint matters because styling can suggest a conclusion the raw positions do not. Write your caption underneath, then hand the bare chart, without the caption, to a partner and ask them to state in one sentence what it shows. To evaluate the result, place the two sentences side by side and mark each as matching or mismatching the plotted positions. If the partner's description disagrees with your caption, revise whichever sentence the marks do not support, and rerun the test with a second partner; a caption that survives two independent readings is the deliverable.

Stage Four: Run a claim ledger before you publish

The review quotes Chalabi arguing that unless we talk more openly about our personal wealth, we will never be able to tackle the economic problems causing insecurity and anxiety for us all, in a book it describes as a mix of journalism, data visualisation and memoir. The excerpt does not show which figures carry a claim that broad, so how such sentences are evidenced in Ten Lives stays invisible to us. What we can offer students is a hypothesis, not a settled rule: when memoir and numbers share a page, a general sentence can drift beyond the single case behind it, and neither the voice nor the chart alone may reveal that gap. Stage Four is a pre-publication check built on this hypothesis.

Our suggested exercise is a claim ledger run on the draft before publication. List every general claim it makes, such as students overspend on delivery or commuting is the largest hidden cost, including the headline and closing sentence. Beside each, cite the specific figure or figures that cover it, noting how many cases each figure represents. Two constraints apply: every cited figure must be traceable to the datasets built in earlier stages or another named source, and a claim resting on one anecdote is downgraded in scope, from students spend to I spent or the ten people surveyed spent, rather than deleted. To evaluate the result, count unsupported generalizations before and after the ledger pass; if the count does not drop, the method needs rethinking, not the prose. A second reader can then recheck each surviving claim against its cited figure.

Sources