This flow chart provides guidance for selecting the appropriate statistical test for hypothesis tests and confidence intervals based on characteristics of the data being analyzed. It outlines questions to consider regarding whether the data is quantitative or categorical, from a single set or multiple sets, paired or independent, and directs the user to select from tests such as z-tests, t-tests, chi-squared tests, or correlations as applicable to analyze proportions, means, distributions, or associations in the data.
This flow chart provides guidance for selecting the appropriate statistical test for hypothesis tests and confidence intervals based on characteristics of the data being analyzed. It outlines questions to consider regarding whether the data is quantitative or categorical, from a single set or multiple sets, paired or independent, and directs the user to select from tests such as z-tests, t-tests, chi-squared tests, or correlations as applicable to analyze proportions, means, distributions, or associations in the data.
This flow chart provides guidance for selecting the appropriate statistical test for hypothesis tests and confidence intervals based on characteristics of the data being analyzed. It outlines questions to consider regarding whether the data is quantitative or categorical, from a single set or multiple sets, paired or independent, and directs the user to select from tests such as z-tests, t-tests, chi-squared tests, or correlations as applicable to analyze proportions, means, distributions, or associations in the data.
Decision flow chart for hypothesis tests and confidence intervals
Is the data quantitative or categorical?
Are you analyzing the proportion of a single set of data, comparing proportions of two sets of data, or is it some other situation? T test or T-interval for a single mean Is the data paired, or are the two sets of data from independent samples? Z test or Z-interval for a single proportion Z test or Z-interval for a two proportions T test or T-interval for matched pairs T test or confidence interval for independent samples Are you testing several proportions or counts against some known distribution? Is the data in a 2x2 table? Are you testing that the distribution is different for each column? Or testing that the row and column variables are independent? chi-squared test for goodness of fit chi-squared test for association/independence/ homogeneity Quantitative (means) Categorical (proportions or counts) single set two sets paired independent single set two sets other yes no yes Are you analyzing the mean of a single set of data, comparing means of two sets of data, or determining if there is a correlation between two variables? T test for correlation, or T-interval for the slope correlation