Analysis: In a world saturated with numbers, the most important question may not be what does the data say, but how is it being used to make us feel and think
As an academic, I spend a good deal of my time working with numbers. I design surveys, analyse datasets and argue, often at great length, about what the results do and do not say. I also supervise students as they learn to do the same, guiding them through the mechanics of quantitative research and, just as importantly, through its limits.
One of the earliest lessons we try to instil in students of quantitative methods is a simple one, well established in the research on statistical literacy: numbers are not neutral. They are powerful, but they are not self-explanatory. Every dataset tells more than one story, and the story that reaches the outside world depends on how figures are selected, framed, and repeated.
This lesson matters far beyond the classroom. Any substantial survey will contain a range of findings. Some will reassure, showing broad agreement or stability. Others will alarm, highlighting minorities with extreme or unexpected views. Both are legitimate outputs of the same research exercise, but they do very different work once they escape the methods section and enter public discourse.
We need your consent to load this rte-player contentWe use rte-player to manage extra content that can set cookies on your device and collect data about your activity. Please review their details and accept them to load the content.Manage Preferences
From RTÉ Radio 1's Today with Claire Byrne, why do so many adults struggle with everyday maths?
In academic settings, we train students to present results proportionately. We teach them to show distributions rather than single point estimates, to contextualise outliers, and to explain uncertainty, sampling decisions and limitations. A single percentage, presented in isolation, would rarely pass muster in a dissertation or journal article. It would be challenged immediately. Compared to what? Based on whom? Over what timeframe?
Outside academia, those constraints fall away. It has been long shown that media systems reward salience rather than representativeness. Minority statistics shock. They provoke reaction. They travel faster. A large majority agreeing on something, by contrast, suggests continuity, and continuity rarely competes well for attention. Over time, this preference subtly reshapes how we understand society. What is statistically marginal can begin to feel socially dominant simply because it is the part of the data we encounter most often.
Social media accelerates this distortion dramatically. Research on digital platforms consistently shows that emotionally charged or surprising information spreads faster than contextualised or reassuring material. As numbers move from reports to headlines to timelines, they are compressed into ever smaller fragments. Caveats fall away. Denominators disappear. What remains is a single figure, stripped of the methodological context that gives it meaning.
We need your consent to load this rte-player contentWe use rte-player to manage extra content that can set cookies on your device and collect data about your activity. Please review their details and accept them to load the content.Manage Preferences
From RTÉ Radio 1's The History Show, how the CSO used data to show us what life was like for people in 1916
In this environment, a minority statistic mutates rather than merely informs. Reposted, screenshot and reframed, it begins to circulate as a symbol rather than a measurement. Scholars of framing theory have long warned that repetition creates salience. The more often a figure appears, the more important and widespread the phenomenon is assumed to be. Visibility becomes confused with prevalence.
I see the consequences of this dynamic regularly in my supervision work. Students often arrive with strong assumptions shaped by headlines rather than datasets. They expect the data to confirm a narrative of crisis or collapse because that is how numbers are presented to them in everyday discourse. Part of the supervisory task becomes gently slowing things down, returning to the full table, the full distribution, the uncomfortable fact that most social phenomena are messier and more moderate than the headlines suggest.
This is not an argument for ignoring uncomfortable findings. In both research and teaching, we emphasise that minority views matter. Outliers can be analytically important. Methodologists have long argued that edge cases are often where new questions emerge. But we are equally careful not to mistake the edge for the centre.
Context is everything. A figure of 9% means something very different when placed alongside the remaining 91%. Without that counterweight, readers are left with an exaggerated sense of breakdown or decline.
With it, the same number may instead point to a largely stable picture with specific areas requiring targeted attention. Both interpretations come from the same data. Only one respects its proportions.
The risk here is not deliberate misrepresentation, but is structural. When public conversation is driven by the most alarming slice of every dataset and when that slice is algorithmically amplified, societies can start to see themselves as more fragmented and polarised than they actually are. Research on agenda-setting shows that repeated emphasis shapes not what people think, but what they think about. Over time, this influences trust, attitudes and policy preferences.
This matters because numbers do not merely describe reality, they help construct it. If people are repeatedly told, through selective statistics, that norms are eroding or consensus has collapsed, confidence weakens. Institutions feel less legitimate. Extremes appear closer to the mainstream than they truly are. The story told by the numbers begins to shape the behaviour those numbers were meant to measure.
We need your consent to load this rte-player contentWe use rte-player to manage extra content that can set cookies on your device and collect data about your activity. Please review their details and accept them to load the content.Manage Preferences
From RTÉ Radio 1's Morning Ireland, George Lee, Environment Correspondent, analyses the 2022 Census figures showing that there were 5,149,139 people in the State on Census night.
In the classroom, we try to resist this. The research on statistical thinking is clear: good quantitative work is as much about restraint as discovery. Not every statistically significant result deserves equal prominence. Not every difference is substantively meaningful. Interpretation is not an optional add-on. It is the core of the task. That discipline often evaporates once data enters public circulation.
There is a quieter story that many datasets tell if we are willing to listen for it. Sociological research consistently shows that most societies are defined less by their extremes than by their broad middle. Shared assumptions, overlapping values and general agreement do not make for viral content, but they are the foundations on which democratic life rests. When those foundations are consistently overlooked, public debate tilts towards permanent crisis, even when the underlying evidence is more balanced.
Every dataset tells more than one story and the story that reaches the outside world depends on how figures are selected, framed and repeated
This is not an argument for complacency but for proportionality. Good use of numbers does not ask us to choose between reassurance and concern. It asks us to hold both at once. It asks us to acknowledge minorities without mistaking them for majorities, and to recognise consensus without assuming it is guaranteed.
I am increasingly convinced that statistical literacy is no longer just a technical skill. As several scholars have argued, it is a civic competence. In a world saturated with numbers, the most important question may not be what does the data say, but how is it being used to make us feel and think. That is a question we should be asking far more often.
Follow RTÉ Brainstorm on WhatsApp and Instagram for more stories and updates
The views expressed here are those of the author and do not represent or reflect the views of RTÉ
