Tools · Open Biostatistics
Compute, understand, cite. Each calculator solves instantly in your browser, shows the classic equation and a simple explanation, interprets the result in plain language as you type, hands you the R code that reproduces the calculation and cites the original authors of the method. Open source; no data are transmitted.
Compute, understand, cite
Sensitivity, specificity, predictive values, likelihood ratios and post-test probability with Fagan’s nomogram.
Enter the four counts of the 2×2 table (test versus reference standard) and get Sn, Sp, PPV, NPV, prevalence, accuracy, Youden's index, LR+, LR− and DOR with their confidence intervals, a plain-language interpretation and the equivalent R code.
TypeScript Active
Open calculator →Enter the pre-test probability and the likelihood ratios of the test (LR+ and LR−) and get the post-test probability after a positive and after a negative result, the change in percentage points, Fagan's nomogram, the interpretation and the equivalent R code.
TypeScript Active
Open calculator →Enter sensitivity, specificity and the prevalence where the test will be used (and, if you know them, the sizes of the validation study) and get PPV and NPV with their confidence interval, the likelihood ratios, natural frequencies per 1,000 people, the curve of both values against prevalence, the interpretation and the equivalent R code.
TypeScript Active
Open calculator →Relative risk, odds ratio, absolute risk reduction and NNT; chi-squared, Fisher’s exact test and McNemar.
Enter the four counts of the 2×2 table (exposure or treatment versus outcome) and get the risk in each group, the relative risk, the odds ratio, the absolute and relative risk reduction and the number needed to treat with their confidence intervals, a plain-language interpretation and the equivalent R code.
TypeScript Active
Open calculator →Enter the four counts of the 2×2 table and get the expected frequencies, Pearson's chi-squared with and without Yates' correction, the N−1 variant, Fisher's two-sided exact test, the signed phi coefficient and the conditional odds ratio with its interval, plus the test that Cochran's rule recommends for these counts.
TypeScript Active
Open calculator →Enter the four cells of the paired 2×2 table (two tests or two time points on the same subjects) and get McNemar's test in its three versions, the paired difference of proportions with its confidence interval and the paired odds ratio, with a plain-language interpretation and the equivalent R code.
TypeScript Active
Open calculator →Estimate a proportion or a mean, compare two proportions or two means, precision of a diagnostic test and correlation, with power curves.
Enter the proportion you expect to find and how precisely you want to estimate it; get the sample size, the effect of a finite population, the recruitment needed if you expect losses, and the precision you would reach with a sample size you already have.
TypeScript Active
Open calculator →Enter the standard deviation you expect and how precisely you want to estimate the mean; get the sample size from the normal formula and from the t variant, the effect of a finite population, the recruitment needed if you expect losses, and the precision you would reach with a sample size you already have.
TypeScript Active
Open calculator →Enter the two proportions you expect, the significance level and the power, and get how many participants are needed in each group, with and without continuity correction, adjusted for expected losses, together with the power curve, a plain-language interpretation and the equivalent R code.
TypeScript Active
Open calculator →Enter the difference in means you want to be able to detect, the expected standard deviation, the significance level and the power, and get how many participants each group needs, with the power curve, a plain-language interpretation and the equivalent R code.
TypeScript Active
Open calculator →Enter the mean change you want to be able to detect and the variability of the differences (directly, or from the standard deviation of each measurement and the correlation between them), and get how many pairs are needed, with the power curve, a plain-language interpretation and the equivalent R code.
TypeScript Active
Open calculator →Enter the sensitivity and specificity you expect, the prevalence of the disease where the test will be used and the precision with which you want to estimate them, and get how many diseased people, how many non-diseased people and how many consecutive patients are needed, the interpretation and the equivalent R code.
TypeScript Active
Open calculator →Enter the correlation you want to be able to detect, the significance level and the desired power, and get how many pairs of observations you need, the power you would reach with the ones you already have, the interpretation and the equivalent R code.
TypeScript Active
Open calculator →Cohen’s kappa, confidence intervals for a proportion and a mean, mean and SD from the median, and descriptive statistics with a chart.
Enter how many cases show the characteristic (x) out of a total (n) and get the proportion with its confidence interval by six methods, a comparison among them, a plain-language interpretation and the equivalent R code.
TypeScript Active
Open calculator →Enter the mean, the standard deviation and the number of observations to get the confidence interval for the mean (Student's t), the standard error, the interval for the standard deviation and a plain-language interpretation, with the equivalent R code.
TypeScript Active
Open calculator →Choose which statistics the study reported, enter them together with the number of participants and get the mean and standard deviation estimated by the three published methods, with the skewness warning, a plain-language interpretation and the equivalent R code.
TypeScript Active
Open calculator →Paste a column from your spreadsheet and get the full summary of that variable: mean with confidence interval, standard deviation, median and quartiles, skewness and kurtosis, the Shapiro-Wilk normality test and outliers, together with the histogram and box plot you should look at before deciding what to report.
TypeScript Active
Open calculator →Paste the agreement table of two raters, from 2 × 2 up to 10 × 10, and get Cohen's kappa with its confidence interval, the Landis and Koch band, the largest kappa the marginals allow and, when there are two categories, the PABAK and the prevalence and bias indices that explain why high agreement can still give a low kappa.
TypeScript Active
Open calculator →Kaplan-Meier with Greenwood CI, median and log-rank, available now; logistic, Cox and linear regression with real R in the browser, in preparation.
Paste the follow-up time, the event indicator and (if you have one) the group variable, and get the Kaplan-Meier curve with its confidence band, median survival, survival at the times you care about, and the comparison of two curves with the log-rank test.
TypeScript Active
Open calculator →Calculations run in your browser; the data you enter are neither transmitted nor stored.