Begin with the observational unit, not the Calculate button
Imagine a service team recording ticket-resolution hours for one month. Before pasting values, decide whether each number represents a ticket, a customer, or a daily aggregate; whether reopened tickets are separate observations; and whether missing cases were omitted. Arithmetic cannot repair a list whose rows refer to different units. Retain the source list and its inclusion rule so the summary can be reproduced.
The page accepts comma-, space-, or line-break-separated finite numbers and returns count, sum, mean, median, modes, minimum, maximum, range, variance, and standard deviation. Before interpreting those exact outputs, the estimate sums calculator can provide a rough scale check for a long source list. Use this full report when center and variability must be read together.
Choose population or sample before interpreting variance
Complete finite population
If the list contains every ticket in the defined month and the question describes exactly that month, population variance divides squared deviations by N. The output summarizes that finite group without estimating an unseen population.
Sample used for inference
If the observed tickets are treated as a sample from a broader process, sample variance divides by n−1. The change is a modeling choice, not a cosmetic notation switch, and requires at least two observations.
Ambiguous collection frame
Do not select whichever result looks preferable. Document the target population and sampling process first, then preserve that basis next to the reported variance and standard deviation.
Interpret the delivery-time profile in layers
For the illustrative hours 2, 4, 4, 4, 5, 5, 7, and 9, each statistic answers a different operational question.
- Confirm extent and center
There are eight observations, the minimum is 2, the maximum is 9, and the range is 7. The mean is 5, the median is (4+5)/2=4.5, and 4 is the unique mode.
- Read variability in the selected frame
Treating all eight values as the population gives variance 4 and standard deviation 2. The sample versions are larger because the squared-deviation total is divided by 7 rather than 8.
- Inspect frequencies
The frequency visualization reveals repeated fours that the mean alone hides. To report the repeated fours as an exact share of the eight observations, use the fraction calculator; if spread is the decision variable, use the standard deviation calculator.
- Standardize only with a defensible reference
Once a mean and positive standard deviation describe the intended reference group, the z-score calculator can place one observation in standard-deviation units. That is a separate question from describing this list.
Prevent a complete-looking table from becoming a complete claim
- Mean, median, and mode need not agree. A long right tail can pull the mean upward while leaving the median comparatively stable, so inspect frequencies and collection context before naming a typical case.
- Range uses only two observations and standard deviation weights large deviations through squaring. Neither measure diagnoses an outlier, establishes normality, or explains why observations vary.
- Exact repeated numeric values define modes here. Rounded measurements may create artificial ties, while continuous measurements may have no repeated value even when their distribution has a clear peak.
- This descriptive page does not calculate quartiles, confidence intervals, hypothesis tests, correlations, regressions, sampling weights, missing-data adjustments, or causal effects. Those require additional data and assumptions.