<p>The reproducibility of research findings is important across many disciplines and is a fundamental concept in scientific studies. This paper investigates the reproducibility of statistical hypothesis tests using Nonparametric Predictive Inference (NPI). NPI is a frequentist statistical framework based on minimal modelling assumptions, considering future observations to be exchangeable with the observed data. Its predictive nature makes it particularly suitable for assessing the reproducibility of a test. This paper applies NPI to study the statistical reproducibility of several umbrella alternative tests, including the Mack-Wolfe (MW), Esra and Fikri (EF), and Jonckheere-Terpstra (JT) tests. These tsests evaluate the null hypothesis that location parameters are equal against the alternative hypothesis that they follow a specific order. Several examples are provided to illustrate the application of the proposed methods. The findings suggest that the reproducibility of these tests can be quite poor, especially when the test statistic is close to the critical value.</p>