How can you handle missing data in information graphics?

1 Identify the source and type of missing data

Before you decide how to handle missing data, you need to understand why and how it is missing. Missing data can be caused by various factors, such as errors, gaps, outliers, or censorship in the data collection or processing. Missing data can also be classified into different types, such as missing completely at random (MCAR), missing at random (MAR), or missing not at random (MNAR), depending on how the missingness is related to the observed or unobserved data. Identifying the source and type of missing data can help you choose the most appropriate method to handle it.

2 Choose a method to handle missing data

There are three main methods to handle missing data: deletion, imputation, and visualization. Deletion means removing the cases or variables that have missing data from your analysis and visualization. This method is simple and straightforward, but it can reduce your sample size and introduce bias if the missing data is not MCAR. Imputation means replacing the missing data with estimated values based on the available data. This method can preserve your sample size and reduce bias, but it can introduce uncertainty and error if the imputation is not accurate or transparent. Visualization means showing the missing data as a separate category or value in your information graphic. This method can reveal the patterns and extent of missing data, but it can also complicate your graphic and distract your audience.

3 Apply the method to your information graphic

Depending on the type and purpose of your information graphic, you can apply different techniques to implement the method you have chosen. For example, if you are creating a bar chart or a pie chart, you can delete the missing data if it is a small proportion of your total data, or you can impute it using the mean, median, or mode of your data. Alternatively, you can visualize the missing data as a separate bar or slice with a different color or label. If you are creating a scatter plot or a line chart, you can delete the missing data if it is isolated and does not affect the trend or correlation of your data, or you can impute it using linear interpolation or regression. Alternatively, you can visualize the missing data as a gap or a dot with a different shape or size.

5 Evaluate the impact of missing data

Finally, you should evaluate the impact of missing data on your information graphic and your message. You can ask yourself some questions, such as: How does missing data affect the accuracy, clarity, and aesthetics of your information graphic? How does missing data influence the interpretation and inference of your information graphic? How does missing data affect the trust and engagement of your audience? By evaluating the impact of missing data, you can identify the strengths and weaknesses of your information graphic, and improve it if necessary.