From peer feedback literacy to second language writing anxiety: a Bayesian SEM and network analysis of the indirect role of feedback literacy behaviors
Reading and Writing, 2026 (SSCI, Scopus)
- Yayın Türü: Makale / Tam Makale
- Basım Tarihi: 2026
- Doi Numarası: 10.1007/s11145-026-10913-2
- Dergi Adı: Reading and Writing
- Derginin Tarandığı İndeksler: Social Sciences Citation Index (SSCI), Scopus, IBZ Online, Agricultural & Environmental Science Database, EBSCO Education Source, Education Abstracts, Educational research abstracts (ERA), ERIC (Education Resources Information Center), Index Islamicus, Linguistic Bibliography, MLA - Modern Language Association Database, Psycinfo, MLA International Bibliography, Social Science Premium Collection (ProQuest), Education Collection (ProQuest), Education Source Ultimate (EBSCO), Linguistics Collection (ProQuest)
- Anahtar Kelimeler: Bayesian structural equation modeling, Feedback literacy behaviors, Peer feedback literacy, Psychometric network analysis, Second language writing anxiety
- Eskişehir Osmangazi Üniversitesi Adresli: Evet
Özet
The contribution of feedback to learning depends not only on the quality of the comments provided but also on students’ capacity to understand, use, and manage them effectively. However, little is known about how peer feedback literacy (PFLS) relates to feedback literacy behaviors (FLBS) and second language writing anxiety (SLWAI). This cross-sectional study examined these relationships at the construct and item levels among 418 pre-service teachers of English as a foreign language (EFL) from two Turkish state universities. Structural relationships were assessed using Bayesian structural equation modeling (BSEM), while item-level associations between FLBS and SLWAI were examined through regularized psychometric network analysis. Compared with first-year students, second-, third-, and fourth-year students reported higher PFLS (β =.241,.295, and.287, respectively). PFLS was strongly and positively associated with FLBS (β =.442), whereas FLBS was negatively associated with SLWAI (β = −.385). The evidence regarding the direct path from PFLS to SLWAI was inconclusive because its credible interval included zero (β = −.160); however, the model-implied cross-sectional indirect association through FLBS was negative and its credible interval excluded zero (β = −.170). In the exploratory network analysis, sample-specific conditional associations were observed between some FLBS and SLWAI items; however, low bridge stability did not support the reliable identification or ranking of specific bridge items or subdimensions. Overall, the findings indicate cross-sectional associations among PFLS, self-reported FLBS, and SLWAI but do not establish temporal ordering or causal mediation.