This study examines 29 journal abstracts that were completed reviews for publication in the year 2012. It was done to investigate the number (percentage) of abstracts that involved with errors, the most erroneous part of the abstract, and the types and frequency of errors. Also the purpose expanded to compare the results with those of the previous study done on error analysis of abstracts in 2007. This comparison was to check if the corrective feedback on the abstract in the past several years played any role in reducing errors. The parts which had most errors were methods (79%) and results (76%), respectively. Throughout all four parts, purpose, methods, results, and conclusion, using the wrong words (expressions) was the most troublesome type of errors. Preposition errors, omission and addition of words along with some miscellaneous errors decreased compared to the results of 2007 study. Giving corrective feedback on the abstract during the past years might have helped the researchers in nursing profession reduce errors since some of them continuously submitted research papers to the same journal and some others may have referenced the previously published abstract prior to their submission. It is advised to provide L2 learners with corrective feedback so that they can notice, understand errors and build cognitive error correction processes. (Contains 4 tables.)