The Patient-Centered Medical Home (PCMH) is a model of care that has been promoted as a way to transform a broken primary care system in the US. However, in order to convince more practices to make the transformation and to properly reimburse practices who are PCMHs, valid and reliable data are needed. Data that capture patient experiences in a PCMH is valuable, but which instrument should be used remains unclear. Our study aims to compare the validity and reliability of two national PCMH instruments. Telephone surveys were conducted with children who receive care from 20 pediatric practices across Florida (n = 990). All of the children are eligible for Medicaid or the Children's Health Insurance Program. Analyses were conducted to compare the Consumer Assessment of Health Plan Survey-Patient-Centered Medical Home (CAHPS-PCMH) and the National Survey of Children with Special Health Care Needs (NS-CSHCN) medical home domain. Respondents were mainly White non-Hispanic, female, under 35 years old, and from a two-parent household. The NS-CSHCN outperformed the CAHPS-PCMH in regard to scale reliability (Cronbach's alpha coefficients all ≥0.81 vs. 0.56-0.85, respectively). In regard to item-domain convergence and discriminant validity the CAHPS-PCMH fared better than the NS-CSHCN (range of convergence 0.66-0.93 vs. 0.32-1.00). The CAHPS-PCMH did not correspond to the scale structure in construct validity testing. Neither instrument performed well in the known-groups validity tests. No clear best instrument was determined. Further revision and calibration may be needed to accurately assess patient experiences in the PCMH.