فرزانه حمیدی

Replication research in Applied Linguistics (AL) is often undervalued due to an emphasis on originality, leading universities to deem it unsuitable for graduate theses. Without replication, findings remain untested (Plonsky, ۲۰۲۴). Hence, it is crucial to reassess the role of replication and establish clearer evaluation criteria to enhance scientific rigor, starting with universities. This sequential explanatory mixed methods study, conducted in two phases, compared the originality of replication studies with theses and examined academic supervisors' perceptions of replication viability as thesis topics. Adopting the method proposed by Hou et al., (۲۰۲۲) and using Natural Language Processing (NLP) techniques, the researcher quantified originality by analyzing semantic similarities in research topics, methodologies, and results, providing a more objective measure than traditional peer review assessments. The data for this phase consisted of three corpora: a main corpus of AL, which served as the basis for knowledge units to compute originality score replication studies and theses, a corpus of replication studies, and a corpus of theses. These corpora were gathered using ProQuest TDM Studio, where the coding process was also conducted, with the Linguistics & Language Behavior Abstracts (LLBA) database employed. The base corpus provided the knowledge units of AL and did not include the other two corpora. The other two corpora, replication studies and theses, were divided into three parts: Research Topics, Research Methods, and Research Results. The originality of each part was then compared to its counterpart in the other corpus. Additionally, for greater accuracy, the theses corpus was further divided into master's theses and doctoral dissertations. Replication studies showed greater originality in results than master's theses and doctoral dissertations, while doctoral work excelled in topics and methodologies. This finding challenges the belief that replication lacks originality, advocating for its inclusion in graduate programs. For the second phase, ۲۳ supervisors were recruited globally through LinguistList, an online platform, providing linguists' affiliation details and contact information, and they were interviewed asynchronously via email. The supervisors’ responses were collected and explored utilizing Nvivo version ۱۳ and were interpreted adopting theme analysis. They saw replication research as beneficial for advancing knowledge, verifying findings, and refining methods. They noted its role in bridging research gaps and ensuring applicability across contexts, aligning with Institutional Theory's (DiMaggio & Powell, ۱۹۸۳) focus on organizational influences. Supervisors cited originality concerns and institutional pressures as major challenges to replication research. Universities prioritizing novel studies and limited time, resources, and funding create barriers. Accessing original data and comparable methodologies also poses difficulties for students. Supervisors stressed the need for mentorship in replication research, requiring deep literature understanding and methodological critique. They suggested universities offer more coursework and training on replication methodologies to equip students. Supervisors agreed on replication's academic value and advocated for its wider acceptance in graduate programs with clear evaluation guidelines. They also suggested dedicated publication spaces for replication research. Lastly, this study calls for redefining originality in AL research beyond novel theories. It highlights the need for institutional support, including funding and mentorship, and advocates for increased visibility and recognition of replication research.