Sampling Distribution and Statistical Errors

Sampling Distribution  

A sampling distribution is formed by repeatedly taking samples from a population, calculating a statistic (such as mean or proportion) for each sample, and then combining those results to create a distribution.  

- It helps understand how sample statistics vary from one sample to another.  

- The larger the number of samples, the more accurately the sampling distribution represents the population.  

Central Limit Theorem (CLT)  

The Central Limit Theorem predicts the shape of a sampling distribution based on the sample size.  

- As the sample size increases, the sampling distribution of the mean tends to become normal, regardless of the population’s original distribution.  

- This principle is fundamental in inferential statistics, allowing researchers to make predictions and test hypotheses using sample data.  

Types of Errors in Hypothesis Testing  

- Type I Error (False Positive):  

  Occurs when a true null hypothesis is rejected.  

  - Probability of committing this error = α (alpha).  

  - Example: Concluding a medicine works when it actually doesn’t.  

- Type II Error (False Negative):  

  Occurs when a false null hypothesis is accepted.  

  - Probability of committing this error = β (beta).  

  - Beta depends on sample size and variance - larger samples reduce the chance of this error.  

  - Example: Concluding a medicine doesn’t work when it actually does.  

- Rejecting a False Null Hypothesis:  

  The probability of correctly rejecting a false null hypothesis is 1 - β, known as the power of the test.  

  - A higher power means a greater ability to detect true effects.  

Insight:  

Understanding sampling distributions and statistical errors helps researchers design reliable experiments, interpret data correctly, and minimise false conclusions, forming the backbone of sound statistical analysis.