Which of the following is a non-parametric statistical technique?

The Mann-Whitney U test is indeed a non-parametric statistical technique, which distinguishes it from parametric methods such as ANOVA, linear regression, and t-tests. Non-parametric tests do not assume a normal distribution of the data, making them useful when data do not meet the assumptions required by parametric tests.

The Mann-Whitney U test specifically assesses whether there are differences in the distributions of two independent groups without the need for the data to conform to a normal distribution. This makes it suitable for ordinal data or for interval data that may not be normally distributed. It ranks all the data points from both groups together and compares the ranks, rather than relying on means and standard deviations as parametric tests do.

In contrast, the other mentioned techniques—ANOVA, linear regression, and the t-test—all operate under the assumption that the data follows a particular distribution (usually normal) and involves parameters such as means and variances. Therefore, the Mann-Whitney U test is a valuable alternative when these assumptions cannot be satisfied, providing researchers with a robust tool for analyzing differences between groups.