Manipulating Power, Perception, and Legitimacy Through Social Media "Statistics"

Whenever I read social media, or watch the news on TV, I find myself questioning many of the claims made. As an example, there was a post on LinkedIn about how an anti-abortion rally turned into a "super-spreader" for Measles. It mentioned that the event "brought a flurry of measles diagnoses," with "multiple confirmed cases reported in the weeks that followed." What the post doesn't say is how many is "multiple," two is technically multiple. Nor does it show the nexus to the anti-abortion event, because it took weeks and no correlative data was presented. The photos to the post show a demonstration and two pictures, one with a baby appearing to have measles and one with a teenage or adult palm appearing to have measles. Yet, there were 34 comments and 22 reposts at the time I saw this post. Many of the comments had nothing to do with measles, but had more to do with political positioning of multiple agendas.

On the six o'clock news I see the weatherperson showing a map with the caption similar to "45 Million People At Risk By Winter Storm." Are all 45 million who live in those five states actually at risk? How do we know? What about the 300 million who are not at risk, is this something we'll need to consider? Was a winter storm in winter a predictable event? Did people get prepared for it?

Why Should We Care About This as Leaders and Followers?

Social media is especially ripe for statistical misrepresentation because it rewards speed, emotion, and simplicity over context and precision. Percentages, vague qualifiers (“multiple,” “surging,” “record-breaking”), and selectively framed data spread quickly because they trigger alarm (fear based) or validation while omitting base numbers, time frames, or uncertainty. Social media algorithms amplify the most emotionally engaging interpretations, not the most accurate ones, and character limits discourage nuance that would clarify scale or meaning. As a result, technically true statements are routinely presented in ways that distort reality, shaping public perception through implication rather than evidence.

Pulling a Rabbit From Statistics

Here are some examples of how information may be presented, why it is misleading and how to make a more truthful and clearer representation:

Vague Quantifiers (“Multiple,” “Several,” “Many”)

Words like multiple, several, or numerous sound large but often mean very little.

• “The outbreak involved multiple infected parties.” Reality: “Multiple” can legally mean two.

• Why it misleads: The listener imagines a widespread problem when the data may reflect a minimal occurrence.

Truthful but clearer version:

“Two individuals were infected.”

Percentages Without Base Numbers

Percentages exaggerate change when starting numbers are small.

• “Cases increased by 200%.” Reality: 2 cases became 6 cases.

• Why it misleads: “200%” sounds explosive, but the absolute change is only 4 people.

Truthful but clearer version:

“Cases increased from 2 to 6.”

Absolute Numbers Without Context

Big numbers feel alarming when scale is omitted.

• “10,000 new cases reported.” Reality: In a population of 50 million, that’s 0.02%.

• Why it misleads: Without population size or time frame, people can’t judge severity.

Truthful but clearer version:

“10,000 new cases, representing 0.02% of the population...”

Selective Time Frames

Choosing when to start and stop measurement can shape any story.

• “Crime doubled this month.” Reality: It rose from 1 incident last month to 2 this month.

• "Why it misleads: Short or abnormal time windows inflate perceived trends.

Truthful but clearer version:

“Crime increased by one incident compared to last month.”

Averages That Hide Variability

Averages smooth out extremes and can hide real conditions.

• “The average income is $90,000.” Reality: Most people earn $45,000; a few executives earn millions.

• Why it misleads: Averages imply typical experience when they may represent no one.

Truthful but clearer version:

“Median income is $45,000; incomes range widely.”

Correlation Framed as Causation

Two things moving together doesn’t mean one caused the other.

• “After policy X was introduced, incidents declined.” Reality: Other variables (seasonal change, enforcement, reporting shifts) weren’t examined.

• Why it misleads: It suggests certainty where only association exists.

Truthful but clearer version:

“Incidents declined during the same period; causation has not been established.”

Denominator Manipulation

Changing what you divide by changes the story.

• “50% of reported cases came from one group.” Reality: That group makes up 60% of the population.

• Why it misleads: It implies overrepresentation when there may be underrepresentation.

Omitting Uncertainty or Confidence Intervals

Point estimates sound precise, even when they aren’t.

• "The risk is 30%.” Reality: The true risk may lie between 15% and 45%.

• Why it misleads:Certainty is implied where variability exists.

Passive Language That Obscures Agency

Statistics are often paired with grammar that removes responsibility.

• “Errors were made in 12% of cases.” Reality: Someone made those errors.

• Why it misleads: Focus shifts from cause to abstraction.

Technically True, Practically Deceptive Framing

This is the most common tactic.

• “No evidence suggests a significant risk.” Reality: The study was underpowered or incomplete.

• Why it misleads: Absence of evidence is framed as evidence of absence.

The Impact on Power, Perception, and Legitimacy.

Social media amplifies power by allowing those who control the narrative, not necessarily the data, to shape what others believe is important, urgent, or true. When statistics are framed without context, they function as tools of influence rather than information, signaling authority through numbers while avoiding accountability for accuracy. In this environment, power is exercised less through formal position or expertise and more through visibility, repetition, and emotional resonance. The ability to present a claim as “data-driven,” even when it is incomplete or misleading, grants the speaker disproportionate persuasive force.

If information is a form of power, then misuse or misrepresentation is an abuse of power. 

These misrepresentations strongly affect perception because humans naturally rely on cognitive shortcuts when processing large volumes of information. Percentages without baselines, dramatic increases from small numbers, and vague qualifiers trigger intuitive judgments of risk, threat, or importance. Social media platforms reinforce this by prioritizing content that provokes reaction, not reflection, causing perception to harden before clarification ever arrives. Once an initial impression is formed, corrective data, no matter how accurate, struggles to dislodge the emotional narrative already in place.

Over time, repeated exposure to statistically framed claims creates legitimacy by association rather than by substance. Statements that “sound scientific” acquire credibility through familiarity, likes, and shares, not through methodological rigor. This erodes trust in genuine expertise while elevating those who can most effectively perform legitimacy through numbers and confident framing. In this way, social media does not merely spread misinformation; it restructures how legitimacy itself is granted, shifting it away from evidence and accountability and toward influence, amplification, and perceived certainty.

What Should We Do To Combat This?

Think critically and independently, don't get emotionally hijacked by numbers and statements that are unverified. Whenever you hear a quasi-statistic, ask yourself some critical questions, like:

• Out of how many…

• Compared to what…

• Over what time…

• According to whom…

If those answers aren’t immediately available, the statistic is incomplete and possibly misleading in an attempt to illicit an emotional reaction (different than a cognitive response).

Putting It All Together

Social media and main stream media turn statistics into instruments of power and influence by shaping perception and manufacturing legitimacy through repetition, emotional framing, and selective context rather than accuracy in data. Numbers presented without scale, in uncertainty, or without comparison appear authoritative while subtly and subconsciously directing belief and behavior, allowing influence to masquerade as evidence.

Over time, this dynamic conditions people to accept “statistical performance” as a substitute for sound reasoning, where confidence, repetition, and numerical framing are mistaken for accuracy. As distorted or incomplete data becomes commonplace, audiences grow less able, and less motivated, to distinguish between evidence and implication, leading to a gradual erosion of trust in genuine expertise that values nuance, uncertainty, and methodological restraint.

In this environment, data no longer functions as a tool for understanding but as a rhetorical device for persuasion, rewarding those who simplify or sensationalize while marginalizing careful analysis. Consequently, the true safeguard against manipulation is not the mere presence of numbers, charts, or studies, but the practiced ability to interrogate how data is framed, what context is omitted, and whose interests are served by the conclusion being advanced.