Which of the following is an example of data cleaning?

• A. Removing data points affected by fatigue in reaction time tasks
• B. Filling missing values with the mean
• C. Transforming data to a log scale
• D. Collecting more data

Answer: (A)

Data cleaning means removing or correcting data points that are inaccurate or unreliable so the dataset better reflects true patterns. In reaction time tasks, data from trials affected by fatigue tend to be unusually slow or noisy, distorting averages and variability. Excluding those fatigued trials cleans the data by reducing error and bias, making the results more trustworthy. Filling missing values with the mean is data imputation and addresses gaps rather than cleaning faulty measurements. Transforming data to a log scale is a transformation to meet assumptions or stabilize variance, not a cleaning step. Collecting more data expands the dataset rather than addressing problems in the existing data. So removing fatigued data points best exemplifies data cleaning.