Uncertainty in chemistry is the range of possible values within which a measurement’s true value lies. Every measurement you take in a lab, whether you’re reading a thermometer, weighing a sample, or measuring a volume of liquid, carries some degree of uncertainty. No instrument is perfect, no technique is flawless, and no environment is perfectly stable. Uncertainty quantifies that imperfection as a number, giving anyone who reads your data a clear picture of how reliable it actually is.
Why Every Measurement Has Uncertainty
When you step on a bathroom scale and it reads 168 pounds, you’re not claiming your weight is exactly 168.000000 pounds. You’re saying it falls somewhere between about 167.5 and 168.5 pounds. That half-pound range is your uncertainty. The same logic applies to every measurement in chemistry, but the stakes are higher because small errors can cascade through calculations and change your final result.
Uncertainty comes from the physical limitations of your instruments, the conditions in the room, and the way you perform the measurement. An analytical balance can be thrown off by the tiniest air current or a vibration in the table. High humidity causes samples to absorb moisture from the air, while low humidity creates static that makes powders cling to surfaces. Even a thermometer gives uncertain readings if it doesn’t make good contact with the substance you’re measuring. These aren’t mistakes you can simply avoid by being more careful. They’re built into the process of measuring anything.
Random vs. Systematic Errors
The uncertainty in your measurements comes from two fundamentally different types of error, and understanding the distinction matters because you deal with each one differently.
Random errors are unpredictable fluctuations that change from one measurement to the next. Electronic noise in an instrument, slight variations in how you read a scale, or changing wind patterns affecting heat loss from a sample all introduce random error. If you measure the same thing ten times, random error is the reason you get ten slightly different numbers. You can reduce random error by repeating measurements and averaging the results, but you can never eliminate it entirely. Random error limits how precise your measurements can be.
Systematic errors push all your measurements in the same direction, either consistently too high or consistently too low. A scale that isn’t zeroed properly (called an offset error) will add the same extra amount to every reading. A pipette that consistently delivers slightly more liquid than its label says introduces a scale factor error. Systematic errors are sneaky because repeating your measurement won’t reveal them. You’ll get the same wrong answer every time, with great precision. Detecting systematic errors often requires calibrating your instruments against known standards or comparing your results with a different method entirely.
Accuracy and Precision Are Not the Same
These two words get used interchangeably in everyday conversation, but in chemistry they describe different things. Accuracy is how close your measurement is to the true value. Precision is how close your repeated measurements are to each other.
You can be precise without being accurate. Imagine a balance with an offset error that adds 0.5 grams to every reading. If you weigh the same object five times and get 12.5, 12.5, 12.5, 12.5, and 12.5 grams, your measurements are beautifully precise but consistently wrong if the object’s true mass is 12.0 grams. You can also be accurate but imprecise: five readings of 11.8, 12.3, 11.7, 12.2, and 12.0 grams scatter widely, but their average lands right on the true value. Ideally, you want both. Random errors erode precision, and systematic errors erode accuracy.
How Significant Figures Communicate Uncertainty
The number of digits you write down in a measurement is itself a statement about uncertainty. When you report a mass as 4.37 grams rather than 4.4 grams, you’re claiming your measurement is reliable down to the hundredths place. The last digit in any reported measurement is always understood to be an estimate, carrying uncertainty of roughly plus or minus one unit in that final position.
This is why significant figures matter so much in chemistry calculations. If your scale is only reliable to the nearest 0.1 grams, writing down 4.3742 grams implies a level of certainty your instrument can’t actually deliver. The extra digits are meaningless noise disguised as data. The convention is straightforward: your reported value should have no more digits than your uncertainty justifies. If your uncertainty is 0.5 pounds, reporting your weight as 168 (three significant figures) is appropriate. That corresponds to a relative uncertainty of about 0.3%.
How Uncertainty Carries Through Calculations
Here’s where uncertainty gets genuinely important for chemistry students. Rarely is a single measurement your final answer. You measure a mass, measure a volume, and then divide to get a density. You measure concentrations and volumes to calculate the number of moles in a reaction. Each input measurement carries its own uncertainty, and those uncertainties combine in the final result.
The rules for combining uncertainty depend on the math you’re doing. When you add or subtract measurements, the absolute uncertainties add together (technically, they combine as the square root of the sum of their squares, which is slightly smaller than just adding them). When you multiply or divide, the relative (percentage) uncertainties combine instead. This means the measurement with the largest relative uncertainty tends to dominate your final result. If you measure a mass to 0.1% precision but a volume to 5% precision, your calculated density inherits most of its uncertainty from that volume measurement, no matter how carefully you weighed the sample.
This principle, formally called the law of propagation of uncertainty, is the reason chemists care so much about identifying the weakest link in their measurement chain. Improving the precision of an already-precise measurement barely helps. Improving the least precise measurement makes the biggest difference in your final result.
Expressing Uncertainty in Your Results
The internationally accepted way to report a measurement with its uncertainty uses the plus-or-minus notation: you write the measured value, then the uncertainty, with units for both. For example, a mass reported as 10.058 ± 0.027 grams tells the reader both what you measured and how confident you are in that number.
A few conventions keep this clean. The uncertainty itself should be rounded to one or two significant digits. Your measurement should then be rounded to match: if your uncertainty is 0.027 grams, reporting the mass as 10.05762 grams would be inconsistent because those extra digits fall well within the uncertain range. You’d round to 10.058 grams so the last digit aligns with the scale of the uncertainty.
You can also express uncertainty as a relative value (a percentage of the measurement) when that’s more useful. Saying a concentration is known to within 2% gives an immediate sense of reliability without needing to know the actual number. Both absolute and relative uncertainty appear in professional chemistry reports, depending on context.
Reducing Uncertainty in the Lab
You can’t eliminate uncertainty, but you can shrink it. The strategies depend on the type of error you’re dealing with.
- Repeat measurements. Taking multiple readings and averaging them reduces the effect of random error. The more repetitions, the closer your average gets to the true value, though you hit diminishing returns fairly quickly.
- Calibrate instruments. Checking your equipment against known standards catches systematic errors before they corrupt your data. Glassware calibration, where you weigh the water delivered by a pipette, is a standard exercise in analytical chemistry courses for exactly this reason.
- Control environmental conditions. Using draft shields on balances, stabilizing room temperature, and managing humidity all reduce the random fluctuations that degrade your readings.
- Choose better equipment. A balance that reads to 0.0001 grams inherently has less uncertainty than one that reads to 0.01 grams. Higher-grade (Class A) glassware has tighter manufacturing tolerances than general-purpose glass.
- Improve your technique. Reading a burette at eye level to avoid parallax, allowing solutions to reach thermal equilibrium before measuring, and ensuring good contact between thermometers and samples all reduce errors that are technically avoidable.
The goal is never to reach zero uncertainty. It’s to make the uncertainty small enough that your result is meaningful for its intended purpose, and to honestly report whatever uncertainty remains so others can judge the quality of your work.