Sample Surveys and Sampling Methods

Introduction to Sample Surveys

Sample surveys are essential tools in statistics for understanding the characteristics of large populations by examining a subset (sample) of individuals. Since it is often impractical or impossible to collect data from every member of a population, statisticians use samples to make inferences about the whole group.

• Population parameter: A numerical measurement describing a characteristic of a population (e.g., population mean, population proportion).

• Sample statistic: A numerical measurement describing a characteristic of a sample, used to estimate the corresponding population parameter.

In statistics, different notations are used for sample statistics and population parameters:

Additional info: Parameters are usually unknown and are estimated using statistics calculated from samples.

Population and Sample: The Relationship

The process of statistical inference involves two main steps: sampling from the population to obtain a sample, and then using the sample to make inferences about the population.

• Sampling: Selecting a subset of individuals from the population.

• Inference: Drawing conclusions about the population based on sample data.

Key Concepts in Sampling

Three Big Ideas of Sampling

• Bias: A systematic failure of the sampling method to represent the population. Bias can result from undercoverage, voluntary response, or nonresponse.

• Randomization: Protects against bias by giving each individual in the population a known chance of being selected.

• Sample Size: The fraction of the population sampled is less important than the absolute size of the sample. Larger samples generally yield more accurate estimates.

Types of Sampling Methods

1. Simple Random Sample (SRS): Every individual in the population has an equal chance of being selected. Steps:

   • Obtain a complete list of the population.

   • Assign numbers to each individual.

   • Select random numbers (e.g., using a table of random numbers) to choose the sample.

   Example: Selecting 10 students at random from a class of 100 by assigning each a number and using a random number table.

2. Stratified Sample: The population is divided into homogeneous subgroups (strata), and a random sample is taken from each stratum. Example: Dividing a university's students by year (freshman, sophomore, etc.) and randomly sampling from each group.

3. Cluster Sample: The population is divided into heterogeneous groups (clusters), and entire clusters are randomly selected. All individuals in chosen clusters are surveyed. Example: Randomly selecting several classrooms and surveying all students in those rooms.

   • Advantage: Useful for large, spread-out populations.

   • Disadvantage: Clusters may not be representative of the population.

4. Systematic Sample: Individuals are selected at regular intervals from an ordered list. Example: Selecting every 10th name from an alphabetical list of employees.

   • Note: Systematic sampling is only appropriate if there is no hidden pattern in the list.

5. Multistage Sampling: Combines several sampling methods. Example: Randomly selecting schools, then randomly selecting students within those schools.

6. Convenience Sample: Individuals are chosen based on ease of access. Example: Surveying people who happen to be in a particular location. Disadvantage: Often leads to bias and is not representative of the population.

Sources of Bias in Sample Surveys

• Nonresponse Bias: Occurs when selected individuals do not respond, potentially leading to unrepresentative results.

• Response Bias: Occurs when respondents answer inaccurately, often due to question wording or interviewer influence.

• Pilot Survey: A small-scale trial run of a survey to identify potential problems before the main survey is conducted.

Table of Random Numbers

Random numbers are used to ensure unbiased selection in random sampling. A table of random numbers provides a sequence of digits that can be used to select individuals or items at random.

Additional info: To use a table of random numbers, assign each member of the population a unique number, then use the table to select your sample.