We analyzed observational and model data to study the sources of formaldehyde over oil and gas production regions and to investigate how these observations may be used to constrain oil and gas volatile organic compound (VOC) emissions. The analysis of aircraft and satellite data consistently found that formaldehyde over oil and gas production regions during spring and summer is mostly formed by the photooxidation of precursor VOCs. Formaldehyde columns over the Permian Basin, one of the largest oil- and gas-producing regions in the United States, are correlated with the production locations. Formaldehyde simulations by the atmospheric chemistry and transport model WRF-Chem, which included oil and gas NO_x and VOC emissions from the fuel-based oil and gas inventory, were in very good agreement with TROPOMI satellite measurements. Sensitivity studies illustrated that VOCs released from oil and gas activities are important precursors to formaldehyde, but other sources of VOCs contribute as well and that the formation of secondary formaldehyde is highly sensitive to NO_x. We also investigated the ability of the chemical mechanism used in WRF-Chem to represent formaldehyde formation from oil and gas hydrocarbons by comparing against the Master Chemical Mechanism. Further, our work provides estimates of primary formaldehyde emissions from oil and gas production activities, with per basin averages ranging from 0.07 to 2.2 kg h^-1 in 2018. A separate estimate for natural gas flaring found that flaring emissions could contribute 5 to 12% to the total primary formaldehyde emissions for the Permian Basin in 2018.