Statistic Tests

Overview of Statistical Concepts

Introduction to Statistical Tests

  • Importance of statistics: Understanding and interpreting data.

  • Focus on the application of statistics in various fields, especially through the use of software like Microsoft Excel.

One-Sample Z Test

  • Definition: A statistical test used to determine if there is a significant difference between a sample mean and a known population mean.

  • Characteristics:

    • There is only one group being tested (the sample).

    • When to use: Comparing a sample mean to a population mean when the population variance is known.

Computing the Z Test Statistic

  • Purpose: To test whether the sample mean belongs to the population.

  • Formula for Z test: Z = \frac{\bar{X} - \mu}{\sigma/\sqrt{n}}

    • Where:

    • \bar{X} = sample mean

    • \mu = population mean

    • \sigma = population standard deviation

    • n = sample size

Components of the Z Test

  • Numerator: Difference between the sample mean and the population mean, which reflects the extent of deviation from the null hypothesis.

  • Denominator: Represents the standard error of the mean (SEM), which measures the dispersion of the sample mean around the population mean.

Standard Error of the Mean (SEM)

  • Significance: Reflects how much the sample mean varies from the population mean.

  • Formula for SEM: SEM = \frac{\sigma}{\sqrt{n}}

    • Where:

    • \sigma = population standard deviation

    • n = sample size

Important Points about SEM

  • SEM decreases as the sample size increases; larger samples provide more accurate estimates of the population mean.

Steps for Conducting a One-Sample Z Test

  • Famous 7 Steps to Conduct a Z Test:

    1. State the null hypothesis (H0): Assumes no effect or difference.

    2. Set the level of risk: Typical alpha levels are 0.05, 0.01, etc.

    3. Select the appropriate test statistic (Z test in this case).

    4. Compute the test statistic value (Z value) using the formula provided.

    5. Determine the critical value: This is the threshold that the computed Z value must exceed to reject the null hypothesis.

    6. Compare the obtained Z value to the critical value: If Z exceeds the critical value, reject H0.

    7. Decision: Conclude whether to reject or fail to reject the null hypothesis based on the comparison.

Example Illustration

Case Study: Dr. McDonald’s Earth Science Students

  • Data Table:

    • Sample Size: 36

    • Sample Mean: 100

    • Sample Standard Deviation: 5.0

    • Population Size: 1,000

    • Population Mean: 99

    • Population Standard Deviation: 2.5

Steps to Compute Z Test Statistic:

  1. State the null and research hypothesis (H0: no difference, H1: difference exists).

  2. Set the level of risk associated with the null hypothesis (e.g., alpha = 0.05).

  3. Select the Z test statistic for calculation.

  4. Compute the test statistic value using the Z test formula.

  5. Determine the value needed for rejecting the null hypothesis (e.g., critical Z value for alpha = 0.05).

  6. Compare the obtained Z value and the critical value.

  7. Make a decision: If obtained Z is more extreme than critical Z, reject H0; otherwise accept H0.

Interpretation of Results

  • Example Z value obtained: 2.38

  • p-value significance: p < 0.05

    • Z value: Represents the test statistic used.

    • 2.38: Indicates the obtained test statistic value.

    • p < 0.05: Suggests that the result is statistically significant, leading to rejection of H0.

What to Do When Population Variance is Unknown

Transition to One-Sample T-Test

  • Scenario for One-Sample T-Test Use:

    • Comparing a sample mean against a known mean when the population standard deviation is unknown.

    • Estimating population parameters when population standard deviation cannot be obtained.

Conducting a T-Test in Excel

  • Using the Data Analysis Toolpak: Provides straightforward methods for performing t-tests.

  • Reference to tutorial: https://youtu.be/v-ZcqrdTcIQ for practical guidance on conducting t-tests using Microsoft Excel.

Important Concepts in Estimation

Understanding Estimation

  • Definition: The process of making inferences about population parameters based on sample statistics.

  • Purpose: To determine how well our sample represents the population.

The Importance of Sample Quality

  • Evaluation of sample statistics (like sample mean) to ensure they act as good estimates for population parameters.

  • Example: Average bid for a hypothetical scenario (e.g., a game show) and assessing how confident we are in our estimates.

Point Estimate and Confidence Intervals

  • Point Estimate: A single value estimate of a population parameter. Example: Sampling 3 people’s estimates in a class as the point estimate of all viewers’ bids.

  • Confidence Intervals: Provides a range where the true population parameter is expected to fall.

    • Formula: 95\% \ CI = Sample Mean \pm 1.96 \times SE

    • Significance of confidence levels (e.g., 95% vs. 99.7%).

Calculating Confidence Intervals

  • General formula: Point estimate ± margin of error (which incorporates standard error).

  • Adjustment per confidence level (1.96 for 95%, and 3 for 99.7%).

Practical Example of Confidence Interval Calculation

  • Sample Mean: \bar{X} = 34.46

  • Known Population Standard Deviation: \sigma = 5.83

  • Objective: Calculate the point estimate and the 95% Confidence Interval based on sample data (216 serum albumin levels) with repetition of experiments for reliability.

Recap of Key Differences and Applications

  • Distinction between sample distribution and sampling distribution.

  • Differences between z-scores and Z-tests.

  • Appropriate conditions under which to apply a one-sample Z-test versus estimating confidence intervals with known population variance.

  • Conclusion on methodology and usage of statistical tests and estimates.