Sampling Methods and Central Limit Theorem Notes

Goals

• Explain why sampling is necessary to learn about a population.

• Describe various methods for selecting a sample.

• Define and construct a sampling distribution of the sample mean.

• Explain the Central Limit Theorem (CLT).

• Use CLT to find probabilities related to sample means from a population.

Why Sample the Population?

• Physical Impossibility: It's often impossible to check every item in a population.

• Cost and Time Efficiency: Conducting research on the entire population is usually costly and time-consuming.

• Adequate Sample Results: Typically, results obtained from a sample are sufficient for analysis.

• Destructive Testing: Some tests may destroy the item, necessitating a sample approach.

Probability Sampling

• Definition: A probability sample is selected so that each item or person in the population has a known chance of being included.

Methods of Probability Sampling

1. Simple Random Sample: Every member of the population has an equal chance of being selected.

2. Systematic Random Sampling: Members are arranged and a starting point is randomly chosen, then every k-th member is selected.

3. Stratified Random Sampling: The population is divided into subgroups (strata), and random samples are taken from each stratum.

4. Cluster Sampling: The population is divided into primary units; entire clusters are randomly selected.

Non-Probability Sampling

• In non-probability sampling, selection is based on the judgment of the sampler.

• Sampling Error: The difference between a sample statistic and the actual population parameter.

Sampling Distribution of the Sample Means

• Definition: A sampling distribution of the sample mean is the probability distribution that outlines all possible sample means derived from a specific sample size from the population.

Example: Tartus Industries

• Population: Seven employees with hourly earnings.

• Calculating Population Mean:
  $\mu = \frac{(7+7+8+8+9)}{7} = 7.71$

• Sampling Distribution for Sample Size 2: 21 possible samples calculated using the formula:
  $NCn = \frac{N!}{n!(N - n)!}$ where
  $N = 7, n = 2$.

Sample Means Table Calculation

• Average Calculation: Taken from each possible combination of two employees.

• Population mean and sample mean:

  • Mean of sample means will converge to the population mean as sample size increases.

Central Limit Theorem (CLT)

• For a population with mean $\mu$ and variance $\sigma^2$, the sample means from random samples will be approximately normally distributed if the sample size is large enough, regardless of the population distribution.

• The mean of the sample means will equal the population mean $\mu$, and the variance will equal $\frac{\sigma^2}{n}$.

Using Sampling Distribution of the Sample Mean (Sigma Known)

• If a population follows a normal distribution, the sampling distribution of the mean will also be normally distributed.

• Probabilities of sample means compared to the population can be calculated using:

  • For sample sizes of at least 30, sample means will follow a normal distribution even if the population does not.

Example: Cola, Inc. Quality Assurance

• Context: Examine cola amounts in bottles under normal distribution. Mean = 31.2 ounces, SD = 0.4 ounces. 16 bottles sampled had a mean of 31.38 ounces.

• Analysis Steps:

  1. Calculate z-value for sample mean.

  2. Determine probability corresponding to z-value.

Conclusion from Example

• It is unlikely to obtain a sample mean of 31.38 ounces from a normal distribution with the given parameters. This suggests that the filling process is currently putting too much cola into bottles identified as an operational issue.

Explain why sampling is necessary to learn about a population.

Sampling is essential in statistics as it provides insight into a larger population without the need for exhaustive data collection, which is often impractical or impossible. By examining a representative subset, researchers can make inferences about the entire population's characteristics, behaviors, or attitudes while saving time and resources.

Describe various methods for selecting a sample.

There are several methods of sampling, categorized into probability and non-probability techniques. Understanding these methods helps researchers choose the most effective sampling strategy based on their study's goals and constraints.

Define and construct a sampling distribution of the sample mean.

A sampling distribution of the sample mean is critical for understanding the variability of sample statistics. It provides a framework for evaluating how sample means behave across multiple samples drawn from the same population, ultimately facilitating hypothesis testing and confidence interval estimation.

Explain the Central Limit Theorem (CLT).

The Central Limit Theorem states that, regardless of the population's distribution, the distribution of sample means approaches a normal distribution as the sample size increases, typically n ≥ 30. This theorem underpins many statistical methods and enables researchers to make probability statements about sample means.

Use CLT to find probabilities related to sample means from a population.

By applying the CLT, researchers can calculate probabilities associated with sample means and make informed decisions or predictions based on their findings, enhancing the reliability of their conclusions.

Why Sample the Population?

• Physical Impossibility: It is often rigorous or unfeasible to examine every element of a population due to vast sizes or inaccessibility.

• Cost and Time Efficiency: Researching an entire population tends to be prohibitively expensive and time-intensive, necessitating a focus on manageable sample sizes.

• Adequate Sample Results: In many cases, a properly chosen sample provides sufficiently accurate results that reflect the overall population, thereby supporting sound analysis without needing complete data.

• Destructive Testing: Certain assessments may damage or alter the items under study, necessitating a sample rather than full examination to preserve the remaining resources.

Probability Sampling

Definition: A probability sample is selected so that each item or person in the population has a known, non-zero chance of being included in the sample. This method ensures that the sample is representative of the population for more accurate conclusions.

Methods of Probability Sampling

1. Simple Random Sample: Every member of the population has an equal chance of being selected, promoting unbiased representation.

2. Systematic Random Sampling: Members are organized, and a random starting point is selected, with every k-th member chosen thereafter, allowing for structured sampling without bias.

3. Stratified Random Sampling: The population is divided into subgroups (strata) based on shared characteristics, and random samples are drawn from each group to ensure representation across diverse categories.

4. Cluster Sampling: The population is segmented into primary units or clusters, and entire clusters are randomly selected, which can simplify the selection process while still allowing for adequate representation.

Non-Probability Sampling

In non-probability sampling, selection is based on the judgment of the researcher rather than random choice. While more straightforward, this approach can introduce bias into the results and limit generalizability.

• Sampling Error: The difference between a sample statistic and the actual population parameter due to the sampling method used, which highlights the importance of proper sampling techniques to minimize error.

Sampling Distribution of the Sample Means

Definition: A sampling distribution of the sample mean refers to the probability distribution of all possible sample means of a given sample size derived from a population. Understanding this allows for the evaluation of sample accuracy in representing the overall population.

Example: Tartus Industries

Population: Seven employees with varying hourly earnings.

Calculating Population Mean:

$\mu = \frac{(7+7+8+8+9)}{7} = 7.71$

Sampling Distribution for Sample Size 2: There are 21 possible samples derived using the combination formula:

$N C n = \frac{N!}{n!(N - n)!}$ where
$N = 7, n = 2$.

Sample Means Table Calculation

Average Calculation: Averages calculated from all combinations of two employees.

Population mean and sample mean: With increased sample size, the mean of the sample means will converge towards the population mean, highlighting the balance achieved through proper sampling.

Central Limit Theorem (CLT)

The Central Limit Theorem asserts that for a population with mean $\mu$ and variance $\sigma^2$, the sample means from random samples will be approximately normally distributed if the sample size is sufficiently large. The principle holds true regardless of the population's initial distribution form.

The mean of the sample means will align with the population mean $\mu$, while the variance of the sample means will be equal to $\frac{\sigma^2}{n}$, emphasizing the importance of sample size in capturing population characteristics accurately.

Using Sampling Distribution of the Sample Mean (Sigma Known)

In scenarios where a population exhibits normal distribution, the sampling distribution of the mean will also be normally distributed. For sample sizes of at least 30, sample means will approximate normality even if the original population does not follow this pattern, making it possible to use familiar statistical methods.

Example: Cola, Inc. Quality Assurance

Context: Analyze cola amounts in bottles, which are expected to follow a normal distribution. The population mean = 31.2 ounces, with a standard deviation (SD) of 0.4 ounces. A sample of 16 bottles had a mean of 31.38 ounces.

Analysis Steps:

1. Calculate the z-value for the sample mean using the formula:

$z = \frac{(\bar{x} - \mu)}{\sigma / \sqrt{n}}$

2. Determine the probability corresponding to the z-value, which will inform us of how likely it is to obtain such a sample mean under the assumed distribution conditions.

Conclusion from Example

The findings suggest that obtaining a sample mean of 31.38 ounces from a normal distribution characterized by the given parameters is unlikely. This disparity alerts quality assurance teams to a possible operational issue in the filling process, which is urgently addressed to ensure product compliance and customer satisfaction.