What is the best way to handle missing survey responses?

2 Reasons for missing data

Missing data can occur for a variety of reasons, depending on the survey design and implementation. Non-response, when respondents don't complete the survey or answer some questions, is a common occurrence due to lack of interest, motivation, time, or access. Item non-response occurs when respondents answer some questions but not others, often because of confusion, fatigue, boredom, or difficulty. Data entry errors can also occur when responses are incorrectly recorded, coded, or transferred due to human or technical errors. Finally, data processing errors can happen when responses are incorrectly handled, manipulated, or transformed due to software bugs, algorithm errors, or inappropriate methods.

3 Methods to handle missing data

Missing data can have an adverse effect on the quality and validity of survey results, introducing bias, reducing precision, and limiting generalizability. To address this issue, it’s important to handle missing data appropriately and effectively. Common methods include deletion, imputation, and analysis. Deletion involves removing cases or variables with missing data, but this can reduce the sample size and introduce bias if the data is not MCAR. Imputation replaces the missing data with estimated values, such as mean imputation, but this can introduce error and uncertainty if the estimates are inaccurate or unreliable. Analysis adjusts the analysis methods to account for the missing data but can be complex to implement and interpret.

4 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?