This study examined the linearity of the electromyographic (EMG) amplitude vs. concentric and eccentric squat force relationships for monoarticular and biarticular thigh muscles. Fourteen resistance-trained men (mean age ± SD: 22 ± 2 years; estimated thigh muscle cross-sectional area: 221.9 ± 22.7 cm) performed concentric and eccentric squats using a novel testing device from 10 to 90% of their maximum average force. Surface EMG signals were detected from the right vastus lateralis, rectus femoris, and biceps femoris. Linear regression was used to examine the relationships between EMG amplitude and force, and repeated measures analyses of variance were used to assess differences among the muscles. Moderate-to-high coefficients of determination were found for the vastus lateralis for both concentric and eccentric testings (r = 0.587-0.992). For the biceps femoris, the mean linear slope coefficient was significantly greater for concentric vs. eccentric testing (0.044 vs. 0.013 μV RMS·N; p = 0.002; effect size = 1.44). Although EMG amplitude for the vastus lateralis and rectus femoris increased with changes in eccentric force output, the electrical activity of the biceps femoris remained stable. These results demonstrated that the EMG amplitude vs. force relationships for the vastus lateralis were linear, despite the fact that force production during the squat is related to the activation of muscles that must simultaneously function as agonists and antagonists. Our findings for eccentric force testing are in agreement with investigations showing reduced hip extensor activity during concurrent extension at the hip and knee joints.