View source for WBR0594

 {{WBRQuestion
|QuestionAuthor= {{YD}} (Reviewed by  {{YD}})
|ExamType=USMLE Step 1
|MainCategory=Biostatistics/Epidemiology
|SubCategory=Gastrointestinal
|MainCategory=Biostatistics/Epidemiology
|SubCategory=Gastrointestinal
|MainCategory=Biostatistics/Epidemiology
|SubCategory=Gastrointestinal
|MainCategory=Biostatistics/Epidemiology
|MainCategory=Biostatistics/Epidemiology
|MainCategory=Biostatistics/Epidemiology
|SubCategory=Gastrointestinal
|MainCategory=Biostatistics/Epidemiology
|SubCategory=Gastrointestinal
|MainCategory=Biostatistics/Epidemiology
|SubCategory=Gastrointestinal
|MainCategory=Biostatistics/Epidemiology
|SubCategory=Gastrointestinal
|MainCategory=Biostatistics/Epidemiology
|MainCategory=Biostatistics/Epidemiology
|SubCategory=Gastrointestinal
|Prompt=A researcher is studying the role of omeprazole in gastric acid secretion. He conducts an experiment involving 10 volunteers, half of which are administered daily doses of omeprazole, and the other half administered similar daily doses of placebo. Six months later, accurate measurements of gastric acid are performed using esophageal pH testing. Although omeprazole is truly associated with suppression of gastric acid in the study, the researcher fails to demonstrate such significant association at a 95% confidence interval during his statistical analysis. Which of the following is true regarding the researcher's trial?
|Explanation=The researcher has a type II (beta) error, or false-negative error, whereby he failed to demonstrate statistical significance despite its presence in the study. The researcher most probably could not demonstrate a significant association between omeprazole and gastric acid suppression in his study because the sample size in his trial is too small (n=10). As such, the study does not have substantial power (Power = 1-beta). Based on his statistical findings, the researcher could not demonstrate a significant association between omeprazole and acid suppression. Accordingly, he cannot reject the null hypothesis and has to accept it. Had the patient had a larger sample size, the study would have been more statistically powerful, and the researcher would have been able to demonstrate significant association and to reject the null hypothesis. On the other hand, a type I (alpha) error, or false-positive error, is defined as the perception of statistical significance when in fact there isn't any.
|AnswerA=The researcher has committed a type I error
|AnswerAExp=Alpha error is also called type I error. However, the research has a type II (beta) error.
|AnswerB=The researcher has committed a false-negative error
|AnswerBExp=Type II error is also called false-negative error.
|AnswerC=The researcher's findings cannot be changed by further study
|AnswerCExp=Increasing the sample size can reduce type II error and increase the statistical power of the study.
|AnswerD=The researcher is able to reject the null hypothesis based on his statistical findings
|AnswerDExp=Based on his findings, the researcher cannot reject the null hypothesis because he failed to demonstrate statistical significance at a 95% confidence interval.
|AnswerE=The power of the study is independent of the probability of making a type II error
|AnswerEExp=The power of the study is defined as: P = 1 - "type II error" . Power is dependent on type II error.
|EducationalObjectives=Type II error is also known as false-negative error. It is present when a researcher fails to demonstrate significance when in fact there is one.
|References=First Aid 2014 page 57
|RightAnswer=B
|WBRKeyword=Type 1 error, Type I error, Type 2 error, Type II error, Error, Statistical error, False positive, False negative, Hypothesis, Fail, Reject, Accept, Null, Alpha, Beta, Power, 1-Beta, Sample size
|Approved=Yes
}}