Although vestibular deficits can have severe repercussions on the early motor development in children, vestibular assessment in young children has not yet been routinely integrated in clinical practice and clear diagnostic criteria to detect early vestibular deficits are lacking. In young children, specific adjustments of the test protocol are needed, and normative data are age-dependent as the vestibular pathways mature through childhood. Therefore, this study aims to demonstrate the feasibility of an extensive age-dependent vestibular test battery, to provide pediatric normative data with the concurrent age trends, and to offer a clinical framework for pediatric vestibular testing. This normative study included 133 healthy children below the age of 4 years (mean: 22 mo, standard deviation: 12.3 mo, range: 5-47 mo) without history of hearing loss or vestibular symptoms. Children were divided into four age categories: 38 children younger than 1 year old, 37 one-year olds, 33 two-year olds, and 25 three-year olds. Children younger than 3 years of age were examined with the video Head Impulse Test (vHIT) of the horizontal semicircular canals, cervical vestibular evoked myogenic potentials (cVEMP) with bone conduction stimuli, and the rotatory test at 0.16, 0.04, and 0.01 Hz. In 3-year old children, the vHIT of the vertical semicircular canals and ocular vestibular evoked myogenic potentials (oVEMP) using a minishaker were added to the protocol. The horizontal vHIT appeared to be the most feasible test across age categories, except for children younger than 1-year old in which the success rate was the highest for the cVEMP. Success rates of the rotatory test varied the most across age categories. Age trends were found for the vHIT as the mean vestibulo-ocular reflex (VOR) gain increased significantly with age (r = 0.446, p < 0.001). Concerning the cVEMP, a significant increase with age was found for latency P1 (r = 0.420, p < 0.001), rectified interpeak amplitude P1-N1 (r = 0.574, p < 0.001), and averaged electromyographic (EMG) activity (r = 0.430, p < 0.001), whereas age trends for the latency N1 were less pronounced (r = 0.264, p = 0.004). Overall, the response parameters of the rotatory test did not show significant age effects ( p > 0.01), except for the phase at 0.01 Hz (r = 0.578, p < 0.001). Based on the reported success rates and age-dependent normative vestibular data, straightforward cutoff criteria were proposed (vHIT VOR gain < 0.7, cVEMP rectified interpeak amplitude < 1.3, oVEMP interpeak amplitude < 10 µV) with accompanying clinical recommendations to diagnose early vestibular impairment. In this large cohort of typically developing children below the age of 4 years, the vHIT and cVEMP were the most feasible vestibular tests. Moreover, the age-dependent normative vestibular data could specify age trends in this group of young children. Finally, based on the current results and clinical experience of more than ten years at the Ghent University Hospital (Belgium), a clinical framework to diagnose early vestibular deficits in young patients is proposed.