Proportions
What type of study design is most suitable for comparing two treatments to test for a difference in outcomes?
In hypothesis testing for two proportions, what does a larger standard error indicate about our point estimate?
A presidential candidate wanted to know if support for her was significantly different between women and men. Her team randomly sampled 1000 voters of each gender and found that 756 of the 1000 women supported her and 560 of the 1000 men supported her. Are all three conditions of a two-proportion z-test satisfied?
Why must we check that individual observations within each group are independent when conducting a hypothesis test for two proportions?
In a 2019 experiment studying differences in purchasing habits between college students and working adults, a 95% confidence interval for the difference in proportions of students vs. adults who prefer shopping online is (-0.03, +0.12); what conclusion can be drawn?
Given that the sampling distribution of the difference between sample proportions approximates normality under certain conditions, which methodological errors might invalidate the testing procedure if these conditions aren't met?
In a study comparing two treatments with binary outcomes, how does increasing the sample size impact the width of a confidence interval for the difference in population proportions?
How are we doing?
Give us your feedback and let us know how we can improve
Which method selects individual units in such a way that each unit in the population has an equal chance of being included in the sample?
How would increasing number observations per group potentially impact power associated with statistical test aimed at distinguishing real effects versus null hypothesis scenario?
A researcher conducts hypothesis tests comparing two population proportions using distinct methods resulting in different z-scores; if all else remains constant across methods except allocation ratios (sample sizes), what inference can we draw about these disparities in z-scores?