MyLeoNes™

Missing data — Data, 14–17 years

Blank answers are part of a dataset, not an invisible detail. Understanding why values are missing helps you judge whether a conclusion is trustworthy.

Idea

Missing data means that a value was expected but was not recorded, was refused, or could not be measured. The blank itself carries information: it may show a skipped question, a broken sensor, or a person who could not be reached. Missing values should be counted and described, not silently ignored.

Why

Analysts began treating missing values as a serious problem because removing them can change the group being studied. If people with lower incomes are less likely to answer a money question, the remaining answers may look too comfortable. The key question is not only how much data is missing, but why it is missing.

Worked example

A survey asks 100 people whether they cycle to school. Eighty answer, and 48 say yes. Among respondents, the cycling rate is 48 ÷ 80 = 60%. We must not report 48% of all pupils unless we know the 20 non-respondents resemble the respondents; otherwise the estimate may be biased.

Common trap

A common choice is to replace every blank with zero. That feels tidy because a blank looks like “none”, but it changes “unknown” into a measured value. A missing temperature is not 0 °C, and a missing income is not no income; the replacement must have a justified reason.

Use

Missing-data checks are used in health records, school surveys, climate sensors and online forms. A hospital may discover that one clinic leaves a field blank far more often than others, revealing a training or software problem. Reporting the amount and pattern of missing data makes decisions safer.

Keep exploring

Other languages

Loading MyLeoNes™…