The aim of this study was to conduct content validation through expert judgement of an instrument which explores the nutritional knowledge, beliefs, and habits during pregnancy. This is a psychometric study in which 14 experts participated in the evaluation of each of the questionnaire items, which were divided into two blocks according to the characteristics of sufficiency, clarity, coherence, and relevance. Fleiss’ κ statistic was used to measure strength of agreement. A pre-test with 102 participants was conducted to measure the degree of understandability of the instrument. The strength of agreement obtained for each of the dimensions was almost perfect. For each pair of experts, strength of agreement ranged between substantial and almost perfect. Sufficiency was the characteristic of the questionnaire that obtained the highest values in the two blocks, and was also the most statistically significant (p < 0.001). Coherence was the most statistically significant characteristic in the first block (p = 0.030). Clarity was the most statistically significant characteristic in the second block (p = 0.037). The wording of five of the twenty original items was corrected. The new version of the instrument attained a high degree of understandability. The results suggest that the instrument is valid and may therefore be applied.