Hypotheses in Statistical Testing

MendStat 159.1.004: Hypothesis Testing

Introduction

This section outlines the process of hypothesis testing in statistics, specifically focusing on the testing of a population proportion regarding defectives.

Terms

  • Null Hypothesis (H0): A statement asserting that there is no effect or no difference, and it serves as the default or initial assumption for the statistical test.
  • Alternative Hypothesis (Ha): A statement that contradicts the null hypothesis, suggesting there is an effect or a difference.
  • Population Proportion (p): The proportion of a certain characteristic (in this case, defectives) in the entire population being studied.

Situation Described

  • A statistical test is being conducted to evaluate if the proportion of defectives has decreased.
  • The current threshold for the proportion of defectives is set at 0.8%.

Hypotheses

  • Null Hypothesis (H0):
    • The null hypothesis asserts that the proportion of defectives has not decreased and remains equal to 0.8%.
    • Mathematically, this is expressed as:
      H0:p=0.008H_0: p = 0.008
  • Alternative Hypothesis (Ha):
    • The alternative hypothesis posits that the proportion of defectives has indeed decreased below the threshold of 0.8%.
    • Mathematically, this can be expressed as:
      H_a: p < 0.008

Interpretation

  • The null hypothesis, H0, will be tested against the alternative hypothesis, Ha, using appropriate statistical methods to determine which hypothesis is supported by the sample data provided.
  • A rejection of the null hypothesis would suggest that there is sufficient evidence to conclude that the proportion of defectives has decreased below 0.8%.