YEAR 03 • CLINICAL TRANSITION

Research & Evidence-Based Medicine

20 learning units • 100 chapter sections • 200 explained questions. Build a question, judge the evidence and connect the answer with the person receiving care.

Original RAHBAR study material for the Year 3 learning pathway. Educational examples support supervised learning; institutional research and clinical requirements still apply. References & further reading ↓

MODULE 06 / 20

Descriptive Statistics & Graphs

Summarise distributions faithfully before testing hypotheses or interpreting associations.

Quick revision ↓ Medical Dictionary

01Centre and spread

The mean uses all numerical observations and is sensitive to extreme values. The median is the middle ordered value and is often more useful for a strongly skewed distribution. Standard deviation describes spread around the mean; the interquartile range spans the middle half of observations. Always pair a measure of centre with a suitable measure of spread and state units, sample size and missing observations.

02Read the distribution

Before choosing a summary, inspect the shape, outliers and possible data errors. A long right tail is common for lengths of stay and costs. Mean and median then describe different aspects of the distribution, and neither should be selected solely because it gives a more attractive number. An unusual value may be genuine; investigate the source rather than deleting it merely to obtain normality.

03Match graphs to data

Bar charts compare categories, histograms display the distribution of numerical data and scatterplots show the relationship between two numerical variables. Time-series plots preserve temporal order. Label axes, units, denominators and uncertainty clearly. A truncated bar-chart axis can exaggerate visual differences, while plotting a time trend without acknowledging changes in measurement or case mix can suggest a misleading improvement.

04Denominators and percentages

A proportion needs a numerator that belongs within its denominator and a clearly defined population. Forty improved patients out of fifty assessed is 80%, but if fifty more were lost to follow-up, that percentage does not describe the whole original cohort. Report counts alongside percentages. Differences in denominator, follow-up period or missingness can make apparently similar percentages represent very different evidence.

05Worked five-patient example

For lengths of stay of 1, 2, 2, 3 and 12 days, the median is 2 days and the mean is 4 days. The twelve-day admission raises the mean, while the median remains near the typical short stay. Both are correct summaries of different features. Describe the skew and the small sample rather than claiming the mean is mathematically wrong or treating the median as a complete description.

High-Yield Revision

SD describes individual spread; SE describes precision of an estimate.
Always state a percentage’s denominator.
Inspect the data before selecting a summary.

Common Pitfalls

Do not hide missing participants behind a favourable percentage.
Do not delete an outlier without checking its origin and documenting the decision.

References & Further Reading

Use the full methods and relevant chapters when appraising a paper. These sources support further study; the examples and questions above are original teaching exercises.

Reference review: 25 September 2026. Consult current institutional guidance for project approval and clinical decisions.