This paper describes the effect of missing sounding reports on temperature and pressure mean values for mandatory levels using the aerological information from the Camagüey Meteorological Centre. Also it is described the effect of missing data on mean temperature and pressure values at the multiple tropopause levels. The case study belongs to one station for a time lag of eight years. Up to the present these types of studies have been conducted using simulated datasets. The present one uses a real inhomogeneous radiosonde dataset. The main reason for missing reports were transmission problems and possible encoding-decoding difficulties. It has been found that profiles of the mean temperature and altitude show little differences between the complete and incomplete datasets. Moreover, no statistical significant differences were found for the mean values of the variables for the complete and incomplete datasets. The most probable reason for those results is that the cause of the missing reports has a random behaviour. Finally we have found that the only two effects noticed on the statistics were slightly higher values of the mean temperatures in the complete dataset and the decrease in the percent of multiple tropopause reports for the incomplete dataset.