91Ó°ÊÓ

Selection bias occurs when the sample chosen for a survey does not accurately represent the entire population. This results in outcomes that are skewed because certain members of the population are either underrepresented or overrepresented.
In our example, a manager selects the first 60 customers who enter the store on a Saturday morning. While this might seem like a simple way to gather data, it leads to selection bias.
The early Saturday morning shoppers might possess unique characteristics that distinguish them from those who shop at other times. Hence, their shopping habits may not fairly depict the behavior of the general customer pool.

An accurate sample representation is crucial to obtaining valid survey results. Ideally, every member of the target population should have an equal opportunity to be included in the sample.
A biased sample, like the one from our example, can lead to misleading conclusions. The chosen sample of customers who shop early on a Saturday might exclude students, professionals, or evening shoppers—each potentially displaying different shopping patterns.
Ensuring a well-represented sample can help capture a more accurate picture of the customer base and improve the reliability of the survey outcomes.
To achieve this, it's essential to consider various aspects such as:

• The times when different groups of people are likely to visit
• Days of the week that see diverse customer demographics

This approach increases the chance that the sample will reflect the full diversity of the customer population.

Random sampling is a fundamental technique to avoid bias and ensure that every individual in the population has an equal chance of being included in the sample.
In the context of our retail store survey, instead of selecting the first 60 customers on a specific morning, the manager should employ a random sampling method. This could involve:

• Randomly selecting customers throughout different hours and days
• Using a computer-generated random number to select customers
• Employing systematic sampling methods, such as picking every 10th customer

By using random sampling, the survey results will more likely reflect the actual preferences and habits of the entire customer base, leading to more accurate and actionable insights.
Random sampling helps to neutralize biases and provides a solid foundation for making generalizations about the population.