The purpose of this study was to examine the effects of the synthetic phonics training based on explicit instruction on word recognition, oral reading fluency, and spelling of students at risk for dyslexia. The subjects were three second-grade students selected from an elementary school. Their intellectual level was normal and they are assumed to have dyslexia by using the dyslexia screening checklist. This study used a multiple probe design across subjects. As dependent measures, word recognition was to read 10 words with no meaning including forms 1 to 6 in Korean. The formative assessment of the BASA was used for oral reading fluency. Spelling was tested with 10 words with meaning from forms 1 to 6 in Korean. The results are as follows: First, word recognition of subjects was effective and intervention at phoneme level was more effective than that at syllable level. The subjects’ errors were influenced by form of letters and level of difficulty rather than syllables and number of letters. Also, subjects knew the name of each letter, but they did not know the sound of each letter. Second, oral reading fluency of subjects was improved, but the improvement was lowest among the dependent variables. The oral reading errors made by subjects were complex and unique, not common. Third, spelling was the most effective among the dependent variables and the effects were continuously maintained. Some discussions, implications, and suggestions for future research were offered.