Intro to Marketing Research: Hypothesis Testing OneSample Tests Mean Proportion

What It Is

One-sample tests are statistical methods used to determine whether a sample's mean or proportion is significantly different from a known population mean or proportion. A classic example of one-sample testing is the study by John B. Watson, a pioneer in behaviorism, who tested the effect of fear on children. In 1920, Watson conducted an experiment where he conditioned a child, Albert, to fear a white rat by associating it with a loud noise. The study aimed to demonstrate the power of classical conditioning, a concept later developed by Ivan Pavlov. This study matters for marketing decision-making because it highlights the importance of understanding consumer behavior and using data-driven methods to inform marketing strategies.

Key Terms & Concepts

• One-sample test: A statistical test used to compare a sample's mean or proportion to a known population mean or proportion.
  • Example: A company wants to determine if its new product's average rating is higher than the industry average.
• Null hypothesis (H0): The default assumption that there is no significant difference between the sample and population.
  • Example: In a study on the effectiveness of a new advertising campaign, the null hypothesis would be that the campaign has no significant impact on sales.
• Alternative hypothesis (H1): The hypothesis that there is a significant difference between the sample and population.
  • Example: In a study on the effectiveness of a new product feature, the alternative hypothesis would be that the feature has a significant impact on customer satisfaction.
• Type I error: The error of rejecting the null hypothesis when it is true.
  • Example: A company mistakenly concludes that a new product is more effective than the competition when, in fact, it is not.
• Type II error: The error of failing to reject the null hypothesis when it is false.
  • Example: A company fails to detect a significant improvement in customer satisfaction due to a new product feature.
• Z-score: A measure of how many standard deviations an observation is from the mean.
  • Example: A company uses a Z-score to determine if a customer's rating is significantly higher or lower than the average rating.
• Standard error (SE): A measure of the variability of a sample mean or proportion.
  • Example: A company uses the standard error to calculate the margin of error in a survey.
• Confidence interval (CI): A range of values within which the true population mean or proportion is likely to lie.
  • Example: A company uses a confidence interval to estimate the average rating of a new product.
• P-value: The probability of observing a result as extreme or more extreme than the one observed, assuming the null hypothesis is true.
  • Example: A company uses a p-value to determine if the results of a survey are statistically significant.
• Critical region: The region of the distribution where the null hypothesis is rejected.
  • Example: A company sets a critical region to determine if a new product's average rating is significantly higher than the industry average.
• Alpha level (α): The maximum probability of committing a Type I error.
  • Example: A company sets an alpha level of 0.05 to determine if a new product's average rating is significantly higher than the industry average.
• Sample size (n): The number of observations in a sample.
  • Example: A company determines the required sample size to estimate the average rating of a new product.
• Cohen's d: A measure of the effect size of a study.
  • Example: A company uses Cohen's d to determine the magnitude of the effect of a new product feature on customer satisfaction.
• Effect size: A measure of the magnitude of the effect of a study.
  • Example: A company uses an effect size to determine the practical significance of a new product feature.

Common Misunderstandings

• Misunderstanding: One-sample tests are only used for comparing means.
• Correction: One-sample tests can be used to compare both means and proportions.
• Misunderstanding: The p-value is the probability of the null hypothesis being true.
• Correction: The p-value is the probability of observing a result as extreme or more extreme than the one observed, assuming the null hypothesis is true.
• Misunderstanding: A Type I error is always more serious than a Type II error.
• Correction: The seriousness of Type I and Type II errors depends on the context and the consequences of the error.

Quick Application / Identification

A company wants to determine if its new product's average rating is higher than the industry average. The industry average rating is 4.2, and the company's sample of 100 customers has an average rating of 4.5. The standard deviation of the industry average rating is 0.5. What type of one-sample test should the company use?

Answer: t-test. Explanation: The company should use a t-test to compare the sample mean (4.5) to the known population mean (4.2).

A company wants to determine if its new product feature has a significant impact on customer satisfaction. The company's sample of 50 customers has an average satisfaction rating of 4.8, and the standard deviation of the satisfaction ratings is 0.8. What is the effect size of the study?

Answer: Cohen's d = 0.8. Explanation: The company should use Cohen's d to determine the magnitude of the effect of the new product feature on customer satisfaction.

A company wants to estimate the average rating of its new product. The company's sample of 200 customers has an average rating of 4.2, and the standard deviation of the ratings is 0.5. What is the margin of error of the estimate?

Answer: ±0.1. Explanation: The company should use the standard error to calculate the margin of error in the estimate.

Last-Minute Revision

• ⚠️ A Type I error is committed when the null hypothesis is rejected when it is true.
• A one-sample test is used to compare a sample's mean or proportion to a known population mean or proportion.
• The p-value is the probability of observing a result as extreme or more extreme than the one observed, assuming the null hypothesis is true.
• A confidence interval is a range of values within which the true population mean or proportion is likely to lie.
• The alpha level (α) is the maximum probability of committing a Type I error.
• A sample size of 30 is generally considered sufficient for a one-sample t-test.
• Cohen's d is a measure of the effect size of a study.
• The effect size is a measure of the magnitude of the effect of a study.
• A standard error of 0.1 is generally considered small.
• A p-value of 0.05 is generally considered statistically significant.
• A confidence interval of 95% is generally considered sufficient for most marketing applications.
• A sample size of 100 is generally considered sufficient for a one-sample proportion test.
• The null hypothesis is the default assumption that there is no significant difference between the sample and population.
• The alternative hypothesis is the hypothesis that there is a significant difference between the sample and population.
• A Type II error is committed when the null hypothesis is not rejected when it is false.