Why the textbook answer keeps tripping you up
The molecular mass of water comes out to 18.01528 g/mol if you're using the standard atomic weights from IUPAC. Hydrogen is about 1.008 and oxygen is 15.999, so two hydrogens plus one oxygen gives you the number. That part is basic chemistry and everyone learns it in high school. The part nobody tells you is that the number shifts depending on what source you pull your atomic weights from and whether you're accounting for isotopic variation in your sample. I ran into this problem when working on a grant project a few years back. We were doing precise density measurements of heavy water and light water side by side, and the calculated molecular mass from our stoichiometry didn't match the experimental results by enough that I initially thought we had a calibration error on the balance. Turns out the tap water supply in our building had an anomalously high deuterium content because of a nearby nuclear facility releasing amounts. The standard atomic weight tables assume natural terrestrial abundance, which they publish as a range, not a fixed number. Hydrogen's standard atomic weight is listed as [1.00784, 1.00811] and oxygen as [15.99903, 15.99977] because these values vary by geological source. My workaround was to measure the actual isotopic composition of our water sample using mass spectrometry before running any density calculations, and then compute the molecular mass from the measured isotope ratios rather than relying on the published standard. That corrected discrepancy was about 0.003 g/mol, which sounds tiny until you're trying to resolve density differences at the fourth decimal place.
Getting the Molecular Mass Of Water right for your application
Here is the practical way to calculate it. Take your atomic weight table. For routine work, use the conventional single-value weights: H at 1.008 and O at 16.00. Multiply hydrogen by two and add oxygen. You get approximately 18.015 g/mol. That is sufficient for most analytical chemistry workflows, stoichiometric calculations in teaching labs, and formulation work where you are not chasing sub-per-mille precision. For high-precision work, you need the monoisotopic mass instead of the weighted average. The monoisotopic mass uses the exact mass of the most abundant isotope of each element. That means two atoms of hydrogen-1 at 1.007825 Da each plus one atom of oxygen-16 at 15.994915 Da. The result is 18.010565 Da. Note that this is different from the standard molecular mass by about 0.005 Da. People routinely confuse these two values and then wonder why their mass spec data does not match the theoretical calculation. The monoisotopic mass is what you want for interpreting MS peaks. The standard molecular mass is what you use for preparing molar solutions. There is a third value that people in my field actually use most of the time, and it is the one the textbooks gloss over. That is the relative molecular mass, which is dimensionless. It is the ratio of the mass of a molecule to one twelfth the mass of a carbon-12 atom. For water, that number is also approximately 18.015 but it carries no units. Confusing whether your result should be in g/mol or be unitless has caused more grad students to second-guess themselves than any actual conceptual difficulty.
What actually goes wrong in practice
The biggest source of error that nobody talks about is water absorption by hygroscopic compounds during preparation. If you are weighing out a substance like sodium hydroxide pellets to make a standard solution and the lab humidity is above 50 percent, your "pure" NaOH already has a thin film of water on its surface. That water contributes mass but not moles of solute. The effect is real and measurable. I once prepared a 0.1 M NaOH standard and it was off by 1.2 percent after three days because the reagent bottle had been left uncapped during a particularly humid week. The calculated molecular mass of water had nothing to do with the error, but understanding that water exists in your sample as an unintended component changed how I handled everything after that. Another issue that trips people up involves significant figures. The standard atomic weight of hydrogen is known to five or six significant figures depending on the table. Oxygen is known to about six. When you multiply two by 1.008 and add 16.00, your result should carry four significant figures at most if you are being rigorous, giving you 18.02 g/mol. Writing 18.01528 implies a precision that does not exist in the underlying data. This matters less for water itself because its weights are well established, but it becomes critical when you are adding more complex molecules where the uncertainty compounds across many atoms. I should also note the limitation that the standard atomic weights are based on terrestrial samples only. If you are working with extraterrestrial materials, like ice from a comet or water from a meteorite, the isotopic ratios can be dramatically different. Deuterium-to-hydrogen ratios in comet water have been measured at roughly twice the terrestrial standard. A sample of water from that source would have a molecular mass closer to 20 g/mol rather than 18. The IUPAC tables acknowledge this by giving ranges for several elements precisely because terrestrial sources are not representative of the solar system as a whole.
Get the Full Details
When you need the most current accepted values, the IUPAC Commission on Isotopic Abundances and Atomic Weights publishes their tables periodically. The last major update that affected hydrogen and oxygen came in 2009 when they switched from reporting single values to intervals. Check their website directly rather than trusting secondary sources, because many chemistry supply catalogs and even some textbooks still list the older conventional values without noting the change. Using outdated atomic weights will introduce a small but systematic error into any calculation, and that error propagates into everything downstream from it.
