Ten first grade students who had responded poorly to a Tier 2 reading intervention in a response to intervention (RTI) model received an intervention of video self-modeling to improve decoding skills and sight word recognition. Students were video recorded blending and segmenting decodable words and reading sight words. Videos were edited and viewed a minimum of four times per week. Data were collected twice per week using curriculum-based measures. A single subject multiple baseline across participants design was used. Results indicated an increase in decoding skills and sight word recognition for all participants. A 2-week posttest maintenance assessment showed retention or increases for 70 percent of participants. Results from the study offer promise for a specific intervention that may reach particular students who respond poorly to Tier 2 reading instruction.