The new method introduced in the article helps scientists create more precise confidence intervals for their research results. By using a normal scores transformation, the intervals can be narrower than traditional ones, especially when dealing with skewed or long-tailed data. When the dataset has uncorrelated or moderately correlated variables, the confidence intervals maintain their accuracy. However, highly correlated variables may slightly affect the coverage probability. The normal scores method is recommended for datasets with 50 or more observations, as it provides reliable results without being computationally intensive.