Abstract The Evidence theory or Dempster-Shafer Theory (DST) has been frequently used in practical applications to deal with uncertainty or lack of information. It is based on the concept of basic probability assignment (BPA). In DST, it is important to quantify the uncertainty (or information) that a BPA represents. An uncertainty measure, known as Deng entropy, was introduced as an interesting alternative to other measures proposed before. In previous work, it was shown that the Deng entropy does not verify most of the required properties for this type of measure and presents some undesirable behaviors. Two modifications of the Deng entropy have been recently proposed, which improve the original one. In this research, we demonstrate that these modifications do also not satisfy the majority of the necessary mathematical properties, and they present most of the behavioral drawbacks of the original one. Therefore, as the original Deng entropy, the modified ones should be cautiously employed in practical applications.