Plagiarism in Online Text-based Structured Interviews

Study Details

• Data Sample: Over 247,500 candidates from various organisations worldwide, participating in online chat-based structured interviews.
• Analysis Process: Plagiarism detection using the Jaccard similarity coefficient to compare new responses against a database of over 6.4 million previous answers.

Key Findings

• Prevalence of Plagiarism: 3.28% of candidates were found to have plagiarized at least one answer
• Gender Differences: Males plagiarized more than females, with rates of 3.57% and 3.04%, respectively.
• Device Type: Candidates using mobile devices plagiarized more frequently than those using desktop computers.
• Interview Questions: Certain questions, particularly those related to teamwork, efficiency improvements, and customer service, were more prone to plagiarism.

Key Takeaway
Plagiarism, akin to faking in traditional assessments, can significantly impact the construct and criterion-related validity of interview results. Detecting and addressing plagiarism is crucial to maintaining fairness and integrity in hiring practices. Implementing enhanced detection tools and regularly updating interview questions can help mitigate this issue.

For more information, download the full paper: