Missing Values
- Data records where a variable has no recorded value for a given observation.
- Can arise from non-response, missed measurements, or technical loss.
- Left unaddressed, they can bias results; common responses are imputation or excluding the missing entries.
Definition
Section titled “Definition”Missing values, also known as missing data or missing observations, refer to the lack of information or data for a specific variable in a dataset. This can occur for various reasons, such as when a survey respondent does not answer a question, when a measurement is not taken, or when data is lost due to technical errors.
Explanation
Section titled “Explanation”Missing values occur when one or more variables have no recorded value for some observations. Causes include respondents skipping questions, failure to take measurements, or technical failures that cause data loss. Because analyses typically assume available observations are representative, missing values can produce biased or incomplete results unless they are addressed. Common approaches mentioned in the source include imputing estimates for missing entries or excluding those observations from analysis.
Examples
Section titled “Examples”Survey on job satisfaction
Section titled “Survey on job satisfaction”In a survey where participants rate satisfaction on a scale of 1 to 5, some participants may choose not to answer or may accidentally skip the question. This results in missing values for the variable of job satisfaction.
Medical study on daily blood pressure readings
Section titled “Medical study on daily blood pressure readings”In a medical study measuring the effectiveness of a new drug, participants are asked to record daily blood pressure readings. Some participants may forget to take their readings on certain days or may not have access to the necessary equipment, resulting in missing values for the variable of daily blood pressure readings.
Notes or pitfalls
Section titled “Notes or pitfalls”- Missing values can lead to biased or incomplete results if not accounted for. For example, in the job satisfaction survey, overall satisfaction may be overestimated if unhappy participants disproportionately omit responses.
- Excluding observations with missing values reduces sample size and can reduce statistical power.
- Imputation replaces missing values with estimates based on existing data; for example, replacing missing job satisfaction responses with the average satisfaction of respondents who answered.
Related terms
Section titled “Related terms”- Missing data
- Missing observations
- Imputation