Nearly every claim about diet and health rests on a measurement of what people ate. That measurement is far harder to obtain than the confident headlines suggest.
Recall is the practical default
The most common method is a structured interview covering everything consumed in the previous day, prompted systematically to surface forgotten items such as beverages and additions.
Trained interviewers use multiple passes and standardized portion aids, because unaided recall misses a large share of what was eaten and misjudges quantities.
A single day is a poor description of a person's usual diet, so studies repeat the interview on non-consecutive days to capture weekday and weekend variation.
Frequency questionnaires trade detail for reach
Large cohort studies often use questionnaires asking how often each of a list of foods was consumed over months. These are cheap enough to administer to tens of thousands of people.
The trade is precision. A fixed food list cannot represent every diet, and estimating average frequency over a long period is a demanding cognitive task.
Such instruments are better at ranking people relative to each other than at estimating anyone's absolute intake, and analyses are usually designed accordingly.
Reporting error is systematic, not random
Under-reporting of total intake is consistently observed, and it is not evenly distributed. It tends to be larger among people with higher body weight and among those aware of dietary expectations.
Foods carrying social judgment are under-reported more than others, while foods regarded as healthy are over-reported. That biases estimates of specific foods, not just totals.
Because the error correlates with the characteristics being studied, it cannot be dismissed as noise that averages out across a large sample.
Objective measures anchor the self-reports
Researchers use biological markers to check reporting. Nitrogen excreted in urine reflects protein intake, and certain compounds in blood or fat tissue track particular dietary sources.
Doubly labeled water measures total energy expenditure, which over a stable period approximates energy intake and exposes systematic under-reporting.
These methods are expensive and burdensome, so they are typically used in smaller validation studies whose results are then applied to correct larger ones.
What this implies for reading nutrition news
Findings from observational dietary research describe associations measured through imperfect instruments, which is one reason results shift as methods improve.
Anyone making dietary changes for a medical reason needs guidance from a physician or registered dietitian rather than from a single study summary.
Knowing how intake was measured is often the fastest way to judge how much weight a dietary claim can carry.