Velopharyngeal function is essential for intelligible speech production, but can often be impaired. Current clinical care could be improved with the use of reliable and objective methods of assessment appropriate for home use. This paper explores the use of a combined nasal acceleration and acoustic sensor to assess velopharyngeal function. Speech production data in nasalized and non-nasalized contexts is recorded from N=6 healthy participants and three normalization strategies are assessed. Normalizing data to maximally nasal productions results in a reduction of between-speaker variability. Using a filtered speech signal can reduce the effects of intra-speaker variability caused be differences in loudness. The normalization strategies pursued show high discriminability of nasalization in vowels with an inexpensive sensor appropriate for home use.