Objective: This investigation used a single-subject, multiple baseline across behaviors design to examine the effect of the peer-mediated intervention (PMI), Stay-Play-Talk (SPT), on social initiation in three peer model/play partner dyads enrolled in a district preschool. Method: Peer model social initiations were coded during a 10-minute center-time activity for play organizer statements, share offers, and offers of assistance. The number of occurrences of each behavior during baselin and intervention was graphed for visual analysis of data. Peer model and teacher perceptions of participating in SPT were measured through weekly Likert scales and a post-intervention questionnaire. Results: All three peer model participants exhibited and immediate therapeutic increase in the number of occurrences across all three social initiation behaviors following SPT training. Weighted Tau-U effect sizes demonstrated statistically significant increases in social initiation across all three peer models for play organizer statements, share offers, and offers of assistance compared to baseline. A weighted Tau-U effect size of .92 also demonstrated statistically significant increases in the number of verbalization in response to peer model initiations across all three play partners. Peer models and their teacher reported overall positive experiences participating in SPT intervention. Conclusions: Visual and data analysis suggested a correlation between SPT and increased social initiations and responses across three peer model/play partner dyads. These results further support the use of SPT as a PMI for preschoolers at risk for social language delays.