This study aimed to develop a learning agility measuring instrument based on De Meuse's theory (2015) with a sample of Government Employees from the National Police of Indonesian Republik and members of Polda Metro Jaya (Lokal Police). The data was collected using an online questionnaire survey distributed via Jotform. The results of the statistical test using a differential power test where the researcher aborted 35 items in the pilot study, which were initially 95 items so that the items that could be continued to the next test stage in a large sample totaled 60 items, then from the average item reliability value has a value per -item ≥ 0.9, this figure shows consistent numbers using the same measuring instrument (test-retest reliability) when testing on small samples and also large samples. Construct Validity value with item homogeneity technique using Product Moment correlation in all items with a p ≤ 0.05. Moreover, it can also be seen from the Pearson correlation value of all items ≥ 0.279 (r-table) of N=50. So that all items in the second phase of the trial can be said to be valid and feasible to use, it can be seen that there is also a positive correlation between dimensions. The results of the factor loading test show that the Standardization Estimate All numbers for each dimension are ≥ 0.71. these figures indicate that each dimension is a factor of the latent construct.
