Knowledge graph completion (KGC) is an important technique for implicitly identifying missing entities or relations in knowledge graphs (KGs) that are employed in various real-world applications such as question answering, information retrieval, and making recommendations. Knowledge graph embedding (KGE) is a typical KGC approach that embeds entity and relation vectors into a low-dimensional vector space for this purpose. In this paper, we present a novel KGE approach called QLogicE. It integrates translation and quantum embedding to capture features of elements within a fact, in the form of word embedding and the logical relationship among facts over KG, in the form of quantum logic. Extensive experimental results on challenging benchmark datasets confirm that the proposed approach QLogicE achieves impressive and (sometimes) surprising performance on 8 embedding dimensions, whereas state-of-the-art KGE approaches typically achieve their best performance at approximately 200 embedding dimensions. In addition, the proposed model achieves 94.84% Hits@1 on the challenging dataset FB15k237, which is almost twice as good as the best performance reported in this metric.