Populations, Samples and Data Quality

A population is the complete group being studied; a sample is the group actually observed. Define both in context before interpreting a result. A discrete variable has countable possible values; a continuous variable measures quantities over an interval. **Decision** | **What to check** Random selection | Use a random mechanism so the intended population can be represented. A larger biased sample remains biased. Reliability | Check who collected the data, when, how and for which population. Missing or erroneous values | Investigate missing responses, units and recording errors; document corrections rather than silently deleting observations. Outliers | Flag values below Q_1-1.5IQR or above Q_3+1.5IQR. A flagged value may be genuine. Example: surveying only early arrivals to estimate all students’ travel times excludes later arrivals. Increasing the number surveyed at the same early time does not remove this bias.