4.8 Take home points

Hypothesis:

• Statistical inference includes estimation and hypothesis testing.
• Hypothesis testing involves rejecting or not rejecting a hypothesis based on data.

Null Hypothesis Significance Testing:

• Involves null and alternative hypotheses, significance level, p-values, and test power.
• Importance of sample size and effect sizes.

Reporting Test Results:

• Emphasizes clarity and transparency.
• Report test statistics, p-values, effect sizes, and confidence intervals.

Statistical Test Selection:

• Choose tests based on data type, groups compared, and study design.
• Includes decision-making frameworks and examples.

Confidence Intervals:

• Provide a range of plausible values for population parameters.
• Can be used to infer hypotheses, with bootstrapped intervals as an alternative.

Bayesian Hypothesis Testing:

• Bayesian approach updates prior beliefs with data.
• Utilizes prior, likelihood, and posterior distributions for decision-making.

Critical Discussion:

• Examines limitations of null hypothesis significance testing.
• Discusses misinterpretation of p-values, overemphasis on significance, and publication bias.