Community Medicine
Basic Statistics and Its Applications
Community medicine biostatistics for MBBS: data types, central tendency and dispersion, normal distribution, tests of significance, sampling and p-values, mapped to NMC codes CM6.1 to CM6.4.
MedNext Academy | 3 min read
Basic Statistics and Its Applications
Community medicine biostatistics for MBBS: data types, central tendency and dispersion, normal distribution, tests of significance, sampling and p-values, mapped to NMC codes CM6.1 to CM6.4.
This chapter introduces the biostatistics used throughout community medicine, from types of data and measures of central tendency and dispersion to the normal distribution and tests of significance. It covers sampling methods, confidence intervals and the interpretation of p-values that underpin evidence-based practice.
High-yield: Basic Statistics and Its Applications
- Data are classified as qualitative, meaning categorical, or quantitative, which may be discrete or continuous.
- Nominal and ordinal scales apply to qualitative data, while interval and ratio scales apply to quantitative data.
- The mean, median and mode are measures of central tendency, and the median is preferred for skewed data.
- The standard deviation is the most useful measure of dispersion, while the range and interquartile range are simpler alternatives.
- In a normal distribution the mean, median and mode coincide, and about 95 percent of values lie within two standard deviations of the mean.
- The standard error of the mean falls as sample size rises, which is why larger samples give more precise estimates.
- A confidence interval gives the range within which the true population value is likely to lie, usually at 95 percent confidence.
- The chi-square test compares proportions in categorical data, while the Student t-test compares means of continuous data.
- The null hypothesis states there is no difference, and a p-value below 0.05 is the conventional threshold for statistical significance.
- A type one error rejects a true null hypothesis, while a type two error fails to reject a false null hypothesis.
- Correlation measures the strength of a linear relationship, with the coefficient ranging from minus one to plus one.
- Random sampling methods include simple random, systematic, stratified, cluster and multistage sampling.
- Data can be presented as bar charts and pie charts for qualitative data, and histograms and line diagrams for quantitative data.
- The power of a study is the probability of detecting a true effect and equals one minus the type two error rate.
Choosing a test of significance
- **Chi-square test:** Compares proportions or associations in categorical (qualitative) data.
- **Student t-test:** Compares the means of continuous data between two groups.
- **p-value:** Below 0.05 is conventionally taken as statistically significant, rejecting the null hypothesis.
- **Errors:** Type one rejects a true null; type two misses a real difference. Power equals one minus type two error.
NMC competencies in this chapter
- **CM6.1:** Formulate a research question for a study
- **CM6.2:** Collection, classification, analysis and presentation of data
- **CM6.3:** Tests of significance and their application
- **CM6.4:** Sampling techniques, central tendency, dispersion and the normal distribution
Frequently Asked Questions
When do you use a chi-square test versus a t-test?
Use the chi-square test to compare proportions or test associations between categorical variables, and use the Student t-test to compare the means of continuous data between two groups.
What does a p-value of less than 0.05 mean?
It means there is less than a five percent probability that the observed difference arose by chance alone under the null hypothesis, which is the conventional threshold for calling a result statistically significant.
What is the difference between a type one and type two error?
A type one error rejects a true null hypothesis, declaring a difference that does not exist, while a type two error fails to detect a difference that is truly present. Study power is one minus the type two error rate.
Why does a larger sample give a more precise estimate?
Because the standard error of the mean decreases as the sample size increases, so estimates from larger samples cluster more tightly around the true population value and yield narrower confidence intervals.
Continue reading
MBBSAll Community Medicine chapters
Continue through the Community Medicine chapter map.
Continue studying Community Medicine
Explore clinician-written learning resources, structured revision and practice across the MedNext platform.
Open in the MedNext appSee plans