Intro to Marketing Research: Hypothesis Testing Test Statistics Z t F ChiSquare

What It Is

Test Statistics are mathematical methods used to analyze and interpret data in marketing research. A famous example is the Z-test used by Coca-Cola to determine the effectiveness of their advertising campaigns. By applying the Z-test, Coca-Cola researchers can determine whether the observed increase in sales is statistically significant, helping them make informed decisions about future marketing strategies.

Key Terms & Concepts

• Z-test: A statistical test used to determine whether a sample mean is significantly different from a known population mean. It is used to test hypotheses about a single population mean. (e.g., Z = (X̄ - μ) / (σ / √n), where X̄ is the sample mean, μ is the population mean, σ is the population standard deviation, and n is the sample size.)
• t-test: A statistical test used to compare the means of two groups. It is used to test hypotheses about the difference between two population means. (e.g., t = (X̄1 - X̄2) / (s / √n), where X̄1 and X̄2 are the sample means, s is the pooled standard deviation, and n is the sample size.)
• F-test: A statistical test used to compare the variances of two groups. It is used to test hypotheses about the equality of two population variances. (e.g., F = (s1^2) / (s2^2), where s1 and s2 are the sample standard deviations.)
• Chi-Square test: A statistical test used to determine whether there is a significant association between two categorical variables. It is used to test hypotheses about the independence of two variables. (e.g., χ^2 = Σ [(observed - expected)^2 / expected], where observed and expected are the observed and expected frequencies.)
• Null hypothesis: A statement of no effect or no difference, which is tested against an alternative hypothesis. (e.g., H0: μ = 0, where μ is the population mean.)
• Alternative hypothesis: A statement of an effect or a difference, which is tested against a null hypothesis. (e.g., H1: μ ≠ 0, where μ is the population mean.)
• Type I error: The probability of rejecting a true null hypothesis. (e.g., α = 0.05, where α is the significance level.)
• Type II error: The probability of failing to reject a false null hypothesis. (e.g., β = 0.20, where β is the probability of Type II error.)
• Power: The probability of rejecting a false null hypothesis. (e.g., 1 - β = 0.80, where β is the probability of Type II error.)
• Effect size: A measure of the magnitude of an effect. (e.g., d = (X̄1 - X̄2) / σ, where X̄1 and X̄2 are the sample means, and σ is the population standard deviation.)
• Confidence interval: A range of values within which a population parameter is likely to lie. (e.g., CI = X̄ ± (Z * (σ / √n)), where X̄ is the sample mean, Z is the Z-score, σ is the population standard deviation, and n is the sample size.)

Common Misunderstandings

• Misunderstanding: The Z-test is only used for large samples.
• Correction: The Z-test can be used for small samples if the population standard deviation is known.
• Misunderstanding: The t-test is only used for paired samples.
• Correction: The t-test can be used for independent samples as well.
• Misunderstanding: The F-test is only used for comparing means.
• Correction: The F-test can be used for comparing variances as well.

Quick Application / Identification

Scenario: A marketing researcher wants to determine whether the average purchase amount of customers who have seen a new ad is significantly higher than those who have not seen the ad. The sample mean of the ad group is $50, and the sample mean of the non-ad group is $30. The population standard deviation is $10, and the sample size is 100. What type of test should the researcher use?

Answer: t-test. Explanation: The researcher wants to compare the means of two groups, which is a classic application of the t-test.

Last-Minute Revision

• Z-test formula: Z = (X̄ - μ) / (σ / √n)
• t-test formula: t = (X̄1 - X̄2) / (s / √n)
• F-test formula: F = (s1^2) / (s2^2)
• Chi-Square test formula: χ^2 = Σ [(observed - expected)^2 / expected]
• Null hypothesis: H0: μ = 0
• Alternative hypothesis: H1: μ ≠ 0
• Type I error: α = 0.05
• Type II error: β = 0.20
• Power: 1 - β = 0.80
• Effect size: d = (X̄1 - X̄2) / σ
• Confidence interval: CI = X̄ ± (Z * (σ / √n))
• Sample size: n ≥ 30 for Z-test and t-test
• Population standard deviation: σ must be known for Z-test
• Pooled standard deviation: s must be calculated for t-test
• Degrees of freedom: df = n - 1 for t-test and F-test