PEOPLE: International Journal of Social Sciences ISSN 2454-5899 © 2017The author and GRDS Publishing. All rights reserved. Available Online at: http://grdspublishing.org/ 423 Aleksić & Arula Volume 3 Issue 2, pp. 423 - 436 Date of Publication: 08 th September, 2017 DOI-https://dx.doi.org/10.20319/pijss.2017.32.423436 COLLABORATIVE TESTING – IMPLICATIONS ON KNOWLEDGE RETENTION Katarina Aleksić Branislav Nusić Primary School, Belgrade, Serbia katarina.aleksic@gmail.com Verica Arula Branislav Nusić Primary School, Belgrade, Serbia verica.arula@gmail.com Abstract Long-term monitoring of students’ achievements (5 th to 8 th grade, Branislav Nušić Primary School, Belgrade, Serbia) on Serbian Language and Computer Science tests, led authors to the conclusion that achievements, in most cases, are lower than expected. Test scores are inconsistent with the level of proficiency students are showing at school, during classes, when knowledge is not formally evaluated or is assessed in a different way. It is also noted that the knowledge that students demonstrate on tests usually does not keep. The lack of functional knowledge is confirmed by students’ weak achievements on the Primary School Final Exam. The authors have noticed that process of testing in both subjects causes discomfort in students, a decline of confidence, uncertainty and fear that they will not be successful enough. The authors have searched for a new type of testing that would have a positive impact on students’ achievements, prolong knowledge retention and provide the freedom of communication that encourage peer learning and encourage self-testing and auto-correction. The authors assumed that deficiencies of traditional testing could be overcome by introducing collaborative testing. Through action research, authors have concluded that collaborative testing contributes to the knowledge retention. Students achieve better results on tests. Testing