Introduction: The diagnosis of childhood growth hormone deficiency (GHD) requires a failure to respond to 2 GH stimulation tests (GHSTs) performed with different stimuli. The most commonly used tests are glucagon stimulation test (GST) and clonidine stimulation test (CST). This study assesses and compares GST and CST’s diagnostic efficacy for the initial evaluation of short children. Methods: Retrospective, single-center, observational study of 512 short children who underwent GHST with GST first or CST first and a confirmatory test with the opposite stimulus in cases of initial GH peak <7.5 ng/mL during 2015–2018. The primary outcome measure was the efficacy of the GST first or CST first in diagnosing GHD. Results: Population characteristics include median age of 9.3 years (interquartile range 6.2, 12.1), 78.3% prepubertal, and 61% boys. Subnormal GH response in the initial test was recorded in 204 (39.8%) children: 148 (45.5%) in GST first and 56 (30%) in CST first, p < 0.001. Confirmatory tests verified GHD in 75/512 (14.6%) patients. Divergent results between the initial and confirmatory tests were more prevalent in GST first than CST first (103/148 [69.6%] vs. 26/56 [46.4%], p < 0.001) indicating a significantly lower error rate for the CST first compared to the GST first. In multivariate analysis, the only significant predictive variable for divergent results between the tests was the type of stimulation test (OR = 0.349 [95% CI 0.217, 0.562], p < 0.001). Conclusions: Screening of GH status with CST first is more efficient than that with GST first in diagnosing GHD in short children with suspected GHD. It is suggested that performing CST first may reduce the need for a second provocative test and avoid patients’ inconvenience of undergoing 2 serial tests.