91Ó°ÊÓ

Population parameters are specific numerical values that describe characteristics of an entire population. These are the fixed values that we try to estimate or make decisions about using sample data. Common population parameters include:

• Mean (\( \mu \)
• Standard deviation (\( \sigma \)
• Proportion (\( p \)
• Variance (\( \sigma^2 \)

Knowing these parameters helps us understand the overall distribution of data in a population. Unlike sample statistics, these values are constant for a given population. They are not calculated from data like statistics but represent the entire group we aim to study.
For example, in Hypothesis A, we consider \( \sigma \) > 100. Here, \( \sigma \) is the population standard deviation, making it a legitimate hypothesis since we compare a population parameter to a specific value.
By forming hypotheses around population parameters, researchers can make inferences about the larger group based on data drawn from samples.

Sample statistics are numerical values calculated from data collected from a sample. A sample is a subset of a population and its statistics are used to make inferences about population parameters. Key sample statistics include:

• Sample mean (\( \bar{x} \)
• Sample standard deviation (\( s \)
• Sample proportion (\( \hat{p} \)
• Sample median (\( \tilde{x} \)

These values help us estimate the population parameters and make decisions about them. However, hypotheses should not be directly formed about sample statistics if we're interested in understanding the overall population.
For example, in the steps for evaluating Hypothesis B, \( \tilde{x} = 45 \), \( \tilde{x} \) denotes the sample median. Since it is not a population parameter, similar to Hypothesis E that involves sample means, such hypotheses are not legitimate. They do not focus on an entire population, which is the core of most statistical hypotheses.

Hypothesis testing is a statistical method used to make decisions about population parameters based on sample statistics. This approach involves forming two hypotheses:

• Null hypothesis (\( H_0 \)): The assumption that there is no effect or no difference between groups.
• Alternative hypothesis (\( H_1 \)): The hypothesis that we aim to support, indicating some effect or difference exists.

The goal of hypothesis testing is to determine if there is enough evidence in the sample data to reject the null hypothesis in favor of the alternative one. It’s a way to use sample data to draw conclusions about a broader population.
For instance, in Hypothesis D, \( \sigma_1 / \sigma_2 < 1 \) involves comparing the standard deviations of two populations. This hypothesis makes use of population parameters and is valid in exploring differences between groups using statistical tests.

A probability distribution describes how the probabilities of a random variable are distributed across its potential values. These distributions are foundational in understanding and analyzing data. Key types include:

• Discrete distributions (e.g., binomial, Poisson)
• Continuous distributions (e.g., normal, exponential)

Such distributions inform the probability of different outcomes occurring, enabling predictions and statistical inference. They outline how sample data probabilities can inform us about larger population parameters.
In the context of the exercise, Hypothesis F, \( \lambda \leq 0.01 \), uses \( \lambda \) as a parameter of an exponential distribution. This is a legitimate form of statistical hypothesis because it links a distribution parameter with the characteristics of component lifetimes.