91Ó°ÊÓ

In statistics, the correlation coefficient is a numerical value that indicates the strength and direction of a linear relationship between two variables. It is often represented by the symbol "r" for sample data, and "\( \rho \)" for the population data. The correlation coefficient can range from -1 to 1, where:

• -1 indicates a perfect negative linear relationship,
• 0 indicates no linear relationship,
• 1 indicates a perfect positive linear relationship.

The closer the value is to 1 or -1, the stronger the linear relationship between the variables.
The sign of the correlation coefficient tells you the direction of the relationship as well. A positive "r" means that as one variable increases, so does the other.
In contrast, a negative "r" means that as one variable increases, the other decreases. In our example, the sample correlation coefficient was 0.78, suggesting a strong positive linear relationship between pump price and gallons sold.

The significance level, often represented by "\( \alpha \)", is a threshold set by the researcher to determine if a result is statistically significant. This value represents the probability of rejecting the null hypothesis when it is actually true (a Type I error).
Common significance levels are 0.05, 0.01, and 0.10. Smaller values demand stronger evidence to reject the null hypothesis.

• \( \alpha = 0.05 \): There is a 5% risk of concluding that a difference exists when there is no actual difference.
• \( \alpha = 0.01 \): Represents a 1% risk.

In the Pennsylvania Refining Company example, the significance level used is \( \alpha = 0.01 \).
This suggests a very stringent criterion for statistical significance, meaning the results obtained must be very strong to conclude that a true correlation indeed exists in the population.

The t-distribution, also known as Student's t-distribution, is crucial in hypothesis testing, especially when dealing with small sample sizes.It is used to estimate population parameters when the sample size is small and the standard deviation is unknown.

• It is similar to the normal distribution but has heavier tails.
• This means it is more prone to producing values that fall far from its mean.

As the sample size increases, the t-distribution approaches a normal distribution.
Calculating the test statistic involves using the t-distribution when the correlation is analyzed, especially under conditions of small sample sizes like in our example (\( n = 20 \)).
The study uses a calculated t-statistic to compare against the critical t-value from the t-distribution table, with \( n-2 \) degrees of freedom. This informs whether to reject the null hypothesis.

A one-tailed test is a statistical test used when the research hypothesis predicts a specific direction of the effect.It tests for the possibility of the relationship in one direction, either greater than or less than a certain value.In simple terms:

• A one-tailed test checks if a value is significantly greater than a specific number.
• Or if it is significantly less, depending on the hypothesis being tested.

In the context of our problem, the alternative hypothesis (\( H_a \)) proposes that the population correlation coefficient is greater than zero.
This makes use of a one-tailed test, since we are only interested in determining if the correlation is significantly positive.
The critical value for this test is determined accordingly, using \( \alpha = 0.01 \), making sure the test is sensitive enough to detect a positive correlation if it truly exists.