Kennedy Omondi Otieno
St. Augustine University of Tanzania (SAUT), Arusha.
Email: omondiken2016@gmail.com

Charles Gervas Rufyiriza
St. Augustine University of Tanzania (SAUT), Arusha.
Email: rufcharles@gmail.com

Abstract: This study assessed the plagiarism policies in public and private higher learning institutions (HLIs) and their effectiveness towards mitigating Artificial Intelligence (AI) assisted breach of originality. The study tested if there is statistically significant difference between the effectiveness of the plagiarism detection software in public and private HLIs on mitigating AI assisted breach of originality and significant difference between the number of times students’ work is subjected to testing for plagiarism in public and private HLIs. The study was grounded on Deterrence Theory. The study employed convergent mixed methods and correlational research design and tested hypotheses using T-test and Chi-square. Target population included 200 course instructors, 1200 students and 7 IT experts from 7 public and private HLIs, with a final sample of 147 respondents. Criterion purposive sampling was used to select seven system administrators while 20 course instructors and 120 students were selected using stratified and simple random sampling. Questionnaire and structured interview were used to collect data. Cronbach coefficient Alpha was employed to test for reliability of instruments; QS=0.759 & QCI=0.738. Credibility and dependability of qualitative data was established through detailed reporting of the research process. Descriptive statistics and thematic analysis were employed. All the two hypotheses yielded the results of p> 0.05 hence the null hypotheses were accepted. In conclusion, both public and private HLIs have established plagiarism policies aimed at addressing AI-assisted breach of originality although they’re not effective and recommended that institutions should more advance tools and guidelines in detecting AI plagiarism in academic work.

Received January 23, 2020; Reviewed March 26, 2020; Accepted April 17, 2020

Abstract: This online study investigated levels of plagiarism from different regions, for example, Africa, Asia, U.S.A and Europe. Specifically, the study investigated theses similarity index according to the following indicators: Sex, University, Region, Level of thesis -Doctoral or Masters and Year the thesis was written. The paper also tested if there was any significant difference between levels of plagiarism and sex of student, regional institutions, level of thesis and year of thesis writing. The results indicate that females plagiarized more than the males, doctoral theses had highest degree of plagiarism than the masters theses, there was significant difference between the sex of student and similarity index, there was no significant difference in the level of plagiarism when grouped according to sex of student, there was no significant difference in the level of plagiarism when grouped according to region, there was no significant difference in the level of plagiarism when grouped according to the level of thesis, there wasa significant difference between the plagiarism similarity index when grouped according to the year of the thesis.The following are the recommendations shared from the study: institutions of higher learning should train students and academic staff on the importance of staying plagiarism free, academic staff should remain keen on plagiarized work and once detected, legal action should be taken,including punishing the student, each institution of higher learning should have plagiarism policy in place, student supervisors should work with students showing them how to avoid plagiarism in their academic writing.