Speech recognition measures are a fundamental component of the audiometric test battery, providing valuable information regarding an individual’s communication difficulties, extending beyond that conveyed by the audiogram. The University of Canterbury Auditory-Visual Matrix Sentence Test was developed in New Zealand English (O’Beirne, Trounson, McClelland, Jamaluddin, & Maclagan, 2015; Trounson, 2012) with the goal of affording an accurate portrayal of these difficulties encountered in real world scenarios. Owing to the cognitive demands of conventional matrix sentence tests, the current study endeavoured to modify the University of Canterbury Auditory-Visual Matrix Sentence Test to develop an audiometric speech recognition measure suitable for use with the paediatric population in New Zealand. Following this, the current study aimed to evaluate the newly developed paediatric measure, alongside its parent test, in order to establish the equivalence of the sentence lists and the conditions in the auditory-alone and auditory-visual modalities for each test individually. Evaluation of the sentence lists with 43 participants with normal hearing suggested that while the sentence lists were equivalently difficult in the auditory-visual modality, the same was not true of the auditory-alone modality. Further evaluation regarding the equivalence of the conditions within each modality indicated that although the accuracy of estimating a listener’s speech recognition threshold was found to be equivalent, the speech recognition threshold values were not. Equivalence is of pivotal importance, allowing speech recognition results to be compared across appointments and clinics; consequently these findings warrant consideration in future research.