This paper presents a novel data validation algorithm for wireless sensor networks. Qualitative methods such as heuristic rule, temporal correlation, spatial correlation, Chauvenet's criterion, and modified z-score were applied for validating sensor data samples with injected faults. Results show that no single method detects all types of data faults without false positives. A new algorithm using a novel combination of heuristic rule, temporal correlation, and modified z-score improved detection and reduced false positives. © 2013 Jaichandran Ravichandran and Anthony Irudhayaraj Arulappan. © 2014 Elsevier B.V., All rights reserved.