Adjusted Fisher–Pearson Sample Skewness
Measure the direction and size of sample asymmetry from three or more observations using the adjusted Fisher–Pearson skewness coefficient.
Summarize a data set, measure its spread, fit a regression or calculate event probabilities. See which sample or population convention each tool uses.
Try a shorter search or choose another category.
Measure the direction and size of sample asymmetry from three or more observations using the adjusted Fisher–Pearson skewness coefficient.
Find the arithmetic average of a list of numbers. Each value carries the same weight, including repeated measurements entered more than once.
Calculate the chance of one or more successes over repeated independent attempts using their shared success probability and the number of attempts.
Update the probability of a hypothesis after observing evidence. Use its prior probability and the likelihood of that evidence under each possible case.
Find the chance of exactly k successes in n independent trials when every trial has the same success probability, such as heads in a series of coin flips.
Calculate Pearson’s goodness-of-fit statistic from counts in matching categories and the expected frequencies predicted under your model.
Find the percentage of variation in y explained by a simple linear regression with an intercept, using matched observations of x and y.
Compare relative spread by expressing the sample standard deviation as a percentage of the mean. Use measurements with a meaningful zero and positive mean.
Compare two independent samples by dividing their mean difference by the pooled within-group standard deviation, using each group’s raw values.
Count how many groups of k distinct objects you can choose from n objects without replacement when rearranging the group leaves it unchanged.
Find the chance that a uniformly distributed value falls within chosen limits by comparing their overlap with the full distribution range.
Calculate covariance for two matched lists of observations. Choose a sample or population denominator to measure how the two variables vary together.
Measure the span of a data set by subtracting its smallest observation from its largest. The result has the same units as the original values.
Find an event’s probability as a percentage from the number of favorable outcomes and the total number of equally likely possibilities.
Find the expected value of a discrete outcome by pairing each possible result with its probability. The probability list must sum to one.
Find the chance an event arrives by a specified time under an exponential waiting-time model. Use a constant event rate and matching time units.
Find the factor that scales a with-replacement standard error when a simple random sample is drawn without replacement from a known population.
Find the geometric mean of positive values, useful for averaging growth factors over equal periods. Enter a 5% gain as a factor of 1.05.
Find the probability that the first success comes on a specified trial. Enter the success chance per independent trial and count the first trial as one.
Find the harmonic mean of positive rates. For average speed over equal travel distances, enter the speed of each leg in a consistent unit.
Find the probability of drawing exactly k target items from a finite group without replacement, using the population size, target count and draw size.
Calculate the chance that two independent events both occur. Enter each event’s probability as a decimal to get the joint probability as a percentage.
Find the spread of the middle half of your data. The calculator interpolates the 25th and 75th percentiles and subtracts the lower from the upper.
Fit a least-squares straight line to paired x and y observations. The fitted equation gives the slope and intercept in the form y = mx + b.
Find the average absolute distance between your observations and their arithmetic mean. Each distance contributes in its original measurement unit.
Find the mean absolute error of a forecast by comparing each prediction with its matching actual value. The result keeps the units of the observations.
Find the middle value in a numerical data set. For an even number of observations, the median is the average of the two central values after sorting.
Calculate unscaled median absolute deviation by finding the median distance from the sample median, a measure of spread resistant to extreme values.
See which value occurs most often in a list. The mode calculator returns every tied value when several share the highest frequency.
Estimate the percentage of a normal distribution at or below a threshold. Enter the threshold, population mean and standard deviation.
Find the height of a normal density curve at a chosen x using the mean and standard deviation. Density is expressed per unit of the input variable.
Compare event odds across two groups by entering the event and nonevent counts for each group, using the second group as the reference.
Calculate a paired t statistic from ordered before-and-after measurements and a hypothesized mean difference, keeping each pair together.
Calculate Pearson’s r from paired x and y values. The coefficient measures the direction and strength of their linear association on a scale from −1 to 1.
Count the ordered ways to fill k positions from n distinct items without replacement, such as arranging a lineup or assigning different roles.
Estimate the chance of exactly k events in an interval with a known average count. The Poisson model assumes independent events at a constant average rate.
Find the chance that event A, event B or both occur. Enter each probability and their known overlap so outcomes shared by both are counted once.
Calculate root mean squared error from paired forecasts and actual values. Squaring the residuals gives larger prediction errors greater influence.
Find the value at a chosen percentile of your sample using inclusive linear interpolation, from the smallest observation at 0 to the largest at 100.
Enter the probabilities of discrete outcomes to measure their average information content using base-two logarithms, with the result in bits.
Find Spearman’s rho from paired observations by ranking both lists, averaging ties and measuring how their orderings track each other.
Calculate standard deviation from a list of measurements. Choose sample or population to measure spread around the mean in the original units.
Estimate the standard error of a sample mean from independent observations. The calculation uses the sample standard deviation and sample size.
Estimate the responses needed to measure a proportion in a large population. Set the confidence critical value, expected proportion and margin of error.
Calculate a mean after trimming a chosen percentage of observations from each tail. Values are sorted before the two ends are removed.
Find Tukey’s lower and upper outlier fences from your data using 1.5 times the interquartile range. Compare observations with these screening boundaries.
Find sample or population variance from your observations. Sample variance divides the sum of squared deviations by one less than the sample size.
Calculate a weighted average from matching lists of values and weights. Use frequencies, percentages or relative weights with a positive total.
Estimate a confidence interval for a proportion from the number of successes and trials. The Wilson score method also handles zero or all successes.
Convert an observation to a z-score using its reference mean and standard deviation. The sign shows which side of the mean the observation falls on.