A finding can be statistically significant but have limited practical significance because the effect size is small. Which of the following is true?

A finding can be statistically significant but have limited practical significance because the effect size is small. Which of the following is true?

• A. False
• B. True
• C. Only when the sample size is large
• D. Only when the p-value is extremely small

Answer: (B)

Statistical significance and practical significance can diverge: a result can be unlikely under the null hypothesis (statistically significant) even if the actual difference or relationship is tiny (practically insignificant). The p-value tells us whether the observed effect could plausibly occur by chance, and sample size plays a big role—large samples can produce very small p-values even when the effect size is small. Practical significance looks at how big the effect is and whether that size matters in real-world terms, which is where measures of effect size and the real-world impact come in. So a finding can meet the threshold for statistical significance yet have limited practical importance if the effect is small, even though the data were unlikely under the null. Tools like Cohen’s d, Pearson’s r, or confidence intervals help gauge this practical importance.

Statistical significance and practical significance can diverge: a result can be unlikely under the null hypothesis (statistically significant) even if the actual difference or relationship is tiny (practically insignificant). The p-value tells us whether the observed effect could plausibly occur by chance, and sample size plays a big role—large samples can produce very small p-values even when the effect size is small. Practical significance looks at how big the effect is and whether that size matters in real-world terms, which is where measures of effect size and the real-world impact come in. So a finding can meet the threshold for statistical significance yet have limited practical importance if the effect is small, even though the data were unlikely under the null. Tools like Cohen’s d, Pearson’s r, or confidence intervals help gauge this practical importance.