Extracellular signal-related kinases 1 and 2 (ERK1/2) are signaling proteins involved in cell survival and proliferation and are overactive/phosphorylated (forming pERK1/2) in approximately 1/3 of human cancers. Measuring pERK1/2:ERK1/2 is therefore common in cancer-related research, but conventional methods are some combination of slow, expensive, and labor-intensive, and large samples are often required. Microfluidic devices offer the possibility of determining pERK1/2:ERK1/2 rapidly, with lower costs, and with substantially reduced labor and reagent requirements compared to conventional methods. This would facilitate more expedient medical decisions and increase experimental throughput in research that requires determining pERK1/2:ERK1/2, such as that related to tumor cell signaling and anti-cancer drug development. This dissertation describes the development of a microfluidic immunoassay for determining the ratio of pERK1/2:ERK1/2 in human cell lysate. Early experiments used antibody-coated beads loaded into chips with poly(dimethylsiloxane) (PDMS) microwell arrays. The fluorophore, assay buffer, incubation protocol, and capture antibody used in all subsequent experiments were chosen here, but these experiments suffered from high (>10%) coefficients of variance. To improve the sensitivity and clinical relevance of the pERK1/2:ERK1/2 assay, the assay was systematically examined until it was determined that the process of loading beads into the PDMS arrays was mechanically damaging their capture antibodies. The next chapter introduced a new chip design for performing pERK1/2:ERK1/2 assays without harsh mechanical loading. These chips were used to optimize assay conditions and determine pERK1/2:ERK1/2 in human Jurkat cell lysate at physiologically relevant concentrations. Further improvement of the assay called for automation to minimize labor/user interaction during experiments, and this required valves to control fluid flow on-chip. The next chapter therefore focused on developing computer-controlled, Peltier-based freeze-thaw valves for microfluidic chips. In the final chapter, these valves were coupled with a chip design for assaying up to 8 samples simultaneously. These chips could be used to perform multiple replicate measurements of a sample or to produce a calibration curve on the same device used to measure samples. Finally, the 8-sample chips were used with the freeze-thaw valves to determine pERK1/2:ERK1/2 in cell lysate with automated fluid control, marking a significant advance toward the realization of automated kinase assays.