When examinees perceive a test as low stakes, it is logical to assume that some of them will not put out their maximum effort. This condition makes the validity of the test results more complicated. Although many studies have investigated motivational fluctuation across tests during a testing session, only a small number of studies have investigated motivational fluctuation across items within a single test. This study aims to examine the pattern of test-taking effort across items in cognitive ability tests when items are presented in a random order manner. Response Time Effort (RTE) was used as a measure of test-taking effort. This measure calculates the proportion of rapid responses in the test based on the response times for each item. Data from 213 university students completing the inductive reasoning test was examined using latent class analysis. The results suggested that examinees in low-stakes testing have different patterns of effort across items. Examinees who consistently provided a high level of effort across items had higher test performance, test-taking engagement, and RTE. Item position and item difficulty are also correlated negatively with test-taking effort. Implications of these results for researchers and practitioners are discussed.