By 
Search for numbers and labels disquantified and you will find a few blog posts that gesture at an idea without ever pinning it down. Some treat it as a philosophy against measurement. Others use it as a vague way of saying “be flexible with data.” Neither version gives you anything to use on a Monday morning.
This post does three things. It defines the phrase in plain terms, shows the real problem underneath it (a number turning into a label that follows a person around), and gives you a working framework for handling scores, ratings and categories more carefully.
One caveat first. This is not a standard term in statistics, sociology or data science. You won’t find it in a textbook. The problem it points at, however, is well documented under other names, and that research is what this article draws on.
What “Disquantified” Means in Practice
A working definition
For this article, numbers and labels disquantified means treating a number or category as one input to a decision instead of the decision itself. The number stays. Its authority gets reduced. A credit score, a test score, a performance rating or a BMI reading is still useful information, but it no longer gets to act as a final verdict on the person it describes.
Where the phrase comes from
The phrase shows up mostly on blogs. Some pages use it as a philosophy of rejecting data’s role in defining identity. Others use it as loose language about adaptability. I could not trace it to a peer-reviewed source or a named originator, so treat it as a handle for an idea rather than an established framework. contact disquantified
What it is not
- It is not a rejection of math or data.
- It is not a claim that grades, tests or metrics should disappear.
- It is not a license to replace measurements with gut feeling.
- It is not a movement with a verified institution behind it. Be skeptical of any page that cites a school or clinic using this approach without naming it.
Numbers vs. Labels: The Distinction Most Articles Skip
Most confusion starts here. Not every number measures something. Psychologist S. S. Stevens described four scales of measurement in 1946, and the difference between them explains why numbers and labels disquantified is a useful idea at all.
| Scale | What the number does | Everyday example | Can you average it? |
|---|---|---|---|
| Nominal | Identifies or names | Jersey number, ZIP code, employee ID | No |
| Ordinal | Ranks in order, with unequal gaps | Class rank, 1–5 star ratings, pain scale | Not reliably |
| Interval | Equal steps, no true zero | Temperature in Fahrenheit | Yes |
| Ratio | Equal steps, true zero | Income, weight, time, distance | Yes |
A player wearing number 10 is not twice the player wearing number 5. ZIP code 90210 is not “more” than 10001. These numbers are labels written in digits. Linguists even study this separately: in “player number five,” the numeral names an object instead of counting anything.
The trouble starts when a label-type number gets treated like a measurement. A five-star rating from reviewers who each use the scale differently gets averaged to 4.3 and treated as a precise quantity. A 1–10 pain scale gets treated as if the gap between 6 and 7 equals the gap between 2 and 3. A performance rating of “3” becomes a fact about a person instead of one manager’s judgment on one day.
How a Number Becomes a Label
A label rarely appears all at once. It builds through a predictable chain.
- Measure. Someone collects data: a test, a transaction history, a sales count.
- Compress. The data collapses into a single figure.
- Categorize. The figure gets sorted into a band: “good,” “high risk,” “below expectations.”
- Attach. The band gets recorded against a person’s name.
- Act. Decisions follow from the band, not from the underlying data.
By step 5, the original measurement is usually invisible. People see the label.
The threshold problem
FICO scores run from 300 to 850, and lenders commonly group them into bands such as poor (300–579), fair (580–669), good (670–739), very good (740–799) and exceptional (800–850). A person at 669 and a person at 670 have almost identical financial behavior, yet they land in different bands. One point changes the label, and the label can change an interest rate.
The stickiness problem
Labels outlive the numbers that created them. A student tagged as a “B student” in ninth grade carries that expectation through teachers who never saw the original work. An employee rated “meets expectations” once may be passed over a year later with no one checking whether the rating still fits.
Four US Examples
Credit scores
A credit score compresses years of borrowing behavior into three digits. It leaves out why a late payment happened, such as a medical emergency or a billing error. Some lenders have started adding cash-flow data, like rent and utility payments, to fill the gap, which is the kind of added context this approach calls for.
Body mass index
The formula behind BMI came from Belgian mathematician Adolphe Quetelet in the 1800s, and he built it to describe populations, not to diagnose individuals. Physiologist Ancel Keys popularized the name in 1972. The standard categories (for example, 25 to 29.9 labeled “overweight”) cannot distinguish muscle from fat or account for body composition differences between people. The American Medical Association adopted a policy in 2023 acknowledging these limitations and recommending BMI be used alongside other measures. That is a reasonable example of treating a number as input instead of verdict.
College rankings
In 2022, Columbia University fell from second to eighteenth in the US News rankings after mathematics professor Michael Thaddeus questioned the accuracy of data the school had submitted and the university acknowledged some of it was inaccurate. A single rank had been steering application decisions for thousands of students, and it rested on figures that did not hold up.
Sales targets at Wells Fargo
In 2016, Wells Fargo agreed to pay about $185 million in fines after regulators found that employees had opened large numbers of unauthorized customer accounts to meet aggressive cross-selling goals. The metric was real. The behavior it produced was not what anyone wanted measured.
What Research Says About Metrics Gone Wrong
The idea behind numbers and labels disquantified has deep roots. Several researchers described versions of it long before the phrase existed.
| Concept | Source | Core idea | Everyday example |
|---|---|---|---|
| Goodhart’s law | Economist Charles Goodhart (1975), popularly phrased by Marilyn Strathern | When a measure becomes a target, it stops being a reliable measure | Teachers narrowing lessons to what a test covers |
| Campbell’s law | Social scientist Donald Campbell (1979) | The more a quantitative indicator drives decisions, the more it invites distortion | Schools under high-stakes testing pressure |
| Trust in Numbers | Theodore Porter (1995) | Numbers gain authority because they look impartial, especially where trust in human judgment is low | Standardized scoring in public agencies |
| Commensuration | Wendy Espeland and Mitchell Stevens (1998) | Different things get reduced to one common metric so they can be compared | Ranking colleges with very different missions |
| Metric fixation | Jerry Muller, The Tyranny of Metrics (2018) | Organizations measure what is easy to count and neglect what matters | Hospitals tracking throughput over patient outcomes |
Read together, these works say the same thing from different angles: a number is a tool for a purpose, and when the purpose is forgotten, the number starts running the show.
A Practical Framework for Using Numbers Without Letting Them Become Verdicts
Here is a checklist you can apply to any score, rating or category you use or receive.
- Ask what the number leaves out. Every metric omits something. Name it before you act.
- Identify the scale. Is it a label, a rank or a true measurement? Do not average or compare in ways the scale does not support.
- Check the thresholds. If a one-point change flips the category, treat the cutoff as a rule of convenience, not a fact about the person.
- Pair it with context. Add one qualitative check: a work sample, a conversation, a history.
- Put an expiry date on the label. Decide how long a rating stays valid and when it gets re-examined.
- Separate measuring from rewarding. The tighter a metric is tied to pay or status, the more people will optimize the metric instead of the goal.
- Let the person respond. Anyone being scored should be able to see the number and challenge it.
The same decision with and without context
| Situation | Number alone | Number plus context |
|---|---|---|
| Hiring | GPA cutoff | GPA plus a work sample or project review |
| Healthcare | BMI category | BMI plus waist measurement, lab work and history |
| School | Letter grade | Grade plus a short teacher narrative |
| Lending | Credit score band | Score plus rent and cash-flow history |
| Employee review | Rating from 1 to 5 | Rating plus two specific examples |
When You Should Keep the Numbers
This is where many articles on the topic go wrong. Some things must be measured precisely, and no amount of nuance replaces that.
- Medication dosing
- Structural load calculations for bridges and buildings
- Financial reporting and audits
- Clinical trial results
- Aircraft maintenance intervals
The goal is not fewer numbers. It is the right relationship between the number and the decision. Where a measurement describes a physical quantity, trust it. Where a number summarizes a person, hold it more loosely.
How to Apply This in Your Own Work
If you manage people
- Replace single-number ratings with a rating plus written examples.
- Review any label (for instance, “high potential”) on a fixed schedule.
- Check whether your team is optimizing the metric or the outcome behind it.
If you teach
- Use grades as one record among several, alongside feedback and work samples.
- Explain what a score does and does not capture, so students do not read it as a ranking of themselves.
- Watch for the point where students stop asking “what did I learn” and start asking “what’s on the test.”
If you build data products
- Show the underlying inputs next to any composite score.
- Document the thresholds and why they sit where they do.
- Give users a way to flag a score that looks wrong.
Common Mistakes
- Treating the idea as permission to ignore data. Dropping measurement entirely usually replaces visible bias with invisible bias.
- Swapping numbers for vague labels. “Not a culture fit” is a label with no measurement behind it, and it is harder to challenge than a score.
- Overstating the evidence. Claiming an institution runs entirely without scores, with no way to check it, weakens your argument.
- Repeating the slogan without the method. Saying a system is “human-centered” does not change how a score is used. Process changes do.
Frequently Asked Questions
What does numbers and labels disquantified mean?
It means treating a score or category as one input to a decision, not the final word on a person or situation.
Is it a recognized academic theory?
No. The phrase appears mainly on blogs. The underlying ideas appear in work on Goodhart’s law, Campbell’s law and the sociology of quantification.
Does it mean we should stop using numbers?
No. It means using numbers with their limits in view and adding context where people are being judged.
What is the difference between a number and a label?
A measurement quantifies something, like weight or time. A label uses digits to name or rank, like a jersey number or a star rating.
What is Goodhart’s law in simple terms?
When a measure becomes a target that people are rewarded for hitting, it stops accurately reflecting what it was meant to track.
How can I avoid reducing people to scores?
Pair every score with a specific example, review labels on a schedule, and let the scored person see and respond to the number.
Conclusion
The useful core of numbers and labels disquantified is narrow and practical: know what a number actually measures, notice when it has hardened into a label, and keep a human check on any decision that affects a person. You do not need to abandon metrics to do that. You need to know which numbers describe the world, which ones only name things, and which ones were never meant to define anyone.
One Comment