Calculates bootstrap confidence intervals by resampling the data B times with replacement and computing the statistic for each resample. Bootstrap CIs do not assume a parametric distribution — percentile method CI is [2.5th, 97.5th percentile of bootstrap distribution] — BCa method corrects for bias and acceleration.
Inputs
Mean Sample
Arithmetic average. Sensitive to outliers — if your data has extreme values, the median may be more representative.
Std Sample
Average spread around the mean. In a normal distribution: 68% within ±1 SD, 95% within ±2 SD, 99.7% within ±3 SD.
N Sample
Number of data points collected. Larger samples give narrower confidence intervals. Rule of thumb: 30+ for most statistical tests.
Conf Level Pct
How certain you want to be that the interval contains the true value. 95% is standard. Going to 99% widens the interval.
Results
confidence interval lower bound
Range within which the true population value lies at the stated probability. A 95% CI means: if you repeated the study 100 times, 95 of those intervals would contain the true value.
confidence interval upper bound
Range within which the true population value lies at the stated probability. A 95% CI means: if you repeated the study 100 times, 95 of those intervals would contain the true value.
margin of error (±)
The difference between the computed result and the exact or true value — a measure of approximation accuracy.
standard error SE = s/√n
The difference between the computed result and the exact or true value — a measure of approximation accuracy.
z-score for
Number of standard deviations a value is from the mean. |z| below 2: within normal range. |z| above 2: in the outer 5% of the distribution. |z| above 3: in the outer 0.3%.