Researchers often disregard the potentially negative effects of unbalanced sampling on power estimates when using multilevel models. The purpose of this study was to investigate the effects that unbalanced sampling had on the estimated level-one power in multilevel random coefficient models. Twelve combinations of conditions were used to compare the effects that a range of proportionally different sized sampling ratios had on the level-1 power for the random coefficient model. Aggregate data presents the summary for sample sizes 200, 500, and 800, three sampling ratios of 0.25 : 0.75, 0.20 : 0.80, and 0.15 : 0.85, four intraclass correlations of 0.2, 0.1, 0.05, and 0.01, three effect sizes of 1, 0.8, and 0.5, and the 108 mean power estimates. Results indicate that as sampling ratios change from 0.25 : 0.75 to incrementally larger unbalanced sampling ratios of 0.15 : 0.85, the estimated power is lowered in almost every case. This effect was more pronounced for smaller sample sizes of 200. Fourteen cases displayed differences larger than 5% in aggregate power estimates with 11 of these cases having larger differences exhibited between the 0.20 : 0.80 and the 0.15 : 0.85 sampling ratios.