The ionosphere is a part of the Earth's atmosphere with complex temporal and spatial distribution and variation. Radio occultation can obtain the vertical distribution of ionosphere, effectively making up for the deficiencies of ground-based Global Navigation Satellite System (GNSS) and other ionospheric sounding technologies; however, the inversion results are subject to the hypothesis model and observation accuracy, making it important to conduct quality assessment. The Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) project started in 2006 and has been renewed since 2019. So far, a large amount of observation data has been accumulated. In this paper, we used the electron density profile data obtained by COSMIC-1 and COSMIC-2 to evaluate the quality. Four evaluation parameters, which only consider the profile's self-characteristics, were used, and the spatiotemporal distribution characteristics of the data quality were analyzed. The experimental results show that the number of occultation electron density profiles of COSMIC-2 is significantly greater, which is about seven times greater than COSMIC-1. In terms of spatial distribution, the unqualified ratio of COSMIC-1 profiles is less than 25% in most mid-latitude areas and about 15% near the equator, while the unqualified ratio rises to 25-50% in high-latitude areas and reaches 65% or even higher in part of the polar region. The unqualified ratio of COSMIC-2 profiles is about 25% near the equator and about 15% in middle and low-latitude areas. In terms of seasonal distribution, the quality of profiles is the worst in winter, followed by summer, and best in spring and autumn. In terms of day-night distribution, the unqualified ratio is higher at night than daytime and changes significantly at the turn of day and night. The nmF2 and hmF2 of the qualified profiles are significantly higher near the magnetic equator than in other regions, exhibiting a double-peaked phenomenon. The case of a large geomagnetic storm shows that the qualified ratio of inversion results can still maintain about 70% during extremely severe magnetic storms.