The purposes of the current study were 1) to investigate predictive variables among various clinical factors used for assessing swallowing disorders induced by stroke; and 2) to identify the relationship between the factors and lesion sites by reviewing 16 articles on dysphagia after stroke. The results of the analysis revealed that the relationship was not conclusive because of the following four reasons: 1) there was not a clear operational definition of each clinical variable; 2) each variable was evaluated by different rating scales; 3) each research tends to contain heterogeneous patient groups; and 4) the lesion sites were not clearly reported. These observations suggest that further research should be directed at collection of prospective data, use of valid and reliable rating scales, and inclusion of lesion data. By doing so, we should be able to aid stroke patients to rehabilitate their swallowing dysfunctions in a more effective way.