Development and preliminary psychometric evaluation of a modified Chinese instrument for self-rated mastery of diabetes exercise knowledge (DEKS-C)
4 affiliations
- 1Nursing Department, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine
- 2School of Nursing and Institute of Nursing Research, School of Medicine, Zhejiang University
- 3Department of Nursing, The Fourth Affiliated Hospital of School of Medicine, and International School of Medicine, International Institutes of Medicine, Zhejiang University
- 4The D. H. Chen School of Universal Health, Zhejiang University
Abstract
Background: Exercise knowledge is seldom evaluated as a standalone construct in China, and no validated instrument exists for self-perceived mastery of exercise knowledge among patients with type 2 diabetes mellitus (T2DM). Adapting the Turkish Diabetes Exercise Knowledge Scale (DEKS) required item consolidation, supplementary items and a revised response format, yielding a modified, DEKS-derived instrument rather than a direct translation.
Methods: This sequential exploratory mixed-methods study was conducted at a tertiary hospital in Hangzhou, China (February–July 2025). In total, 438 patients with T2DM completed the 17-item DEKS-C following forward-backward translation, content validation and three rounds of cognitive interviewing ( n = 23). Internal consistency and the exploratory factor structure were evaluated using Cronbach’s α and principal axis factoring with oblimin rotation; factor retention followed Horn’s parallel analysis, with the three-factor solution re-examined on a polychoric matrix.
Results: Content validation consolidated nine overlapping items into three and added three new items. Cognitive interviewing then consolidated or relocated four items and merged one into the foot-care item, producing 17 items scored 1–4 (total range 17–68). Cronbach’s α was 0.967 (corrected item-total correlations 0.668–0.837), the Kaiser-Meyer-Olkin measure 0.965, and Bartlett’s test χ ² (136) = 6945.15, p < 0.001. Floor and ceiling effects were 13.47 and 8.90%, arising entirely from the 59 all-lowest-response and 39 all-highest-response respondents, the former answering faster (median 89 vs. 232 s, p < 0.001). Horn’s parallel analysis identified three substantially intercorrelated factors (|r| = 0.51–0.79) explaining 55.73–71.23% of total variance (71.23% in the full sample), corresponding to exercise benefits cognition, exercise implementation and exercise safety; an equivalent structure emerged on a polychoric matrix. One pattern coefficient nevertheless exceeded unity and inter-factor correlations were elevated, leaving the solution provisional.
Conclusion: The DEKS-C showed high internal consistency and a three-factor grouping that was broadly stable across sensitivity samples and alternative correlation matrices, supporting content relevance. However, because the factors were substantially intercorrelated and the structure remains unverified, the proposed multidimensional model cannot be regarded as firmly established; independent multicenter and multiregional confirmatory factor analysis is required.
Related
- Development and Preliminary Psychometric Evaluation of a Questionnaire on Comprehensive Protection, Quality, and Humanization in Pediatric Hospital Care
- Subjective word-finding complaints in older adults: Development and preliminary psychometric evaluation of a novel self-reported questionnaire
- Development and preliminary psychometric evaluation of a spectacle trial-wear questionnaire for distance single-vision prescriptions.
- Development and Preliminary Evaluation of the Patient-Artificial Intelligence Rating Instrument
- Preliminary Psychometric Evaluation of an Ultraviolet Radiation-Related Questionnaire in a Case-Control Sample
BibTeX
@article{mei2026development,
title = {{Development and preliminary psychometric evaluation of a modified Chinese instrument for self-rated mastery of diabetes exercise knowledge (DEKS-C)}},
author = {Lei Mei and Hongying Pan and Kaiyan Zhu and Dan Yin and Huilan Yao and Mingxia Zhu and Yufeng Zhou and Lan Lv and Jing Shao},
journal = {Frontiers in Public Health},
year = {2026},
month = oct,
doi = {10.3389/fpubh.2026.1902317},
url = {https://www.frontiersin.org/articles/10.3389/fpubh.2026.1902317/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Development and preliminary psychometric evaluation of a modified Chinese instrument for self-rated mastery of diabetes exercise knowledge (DEKS-C)
AU - Lei Mei
AU - Hongying Pan
AU - Kaiyan Zhu
AU - Dan Yin
AU - Huilan Yao
AU - Mingxia Zhu
AU - Yufeng Zhou
AU - Lan Lv
AU - Jing Shao
PY - 2026
DA - 2026/10/20
JO - Frontiers in Public Health
DO - 10.3389/fpubh.2026.1902317
UR - https://www.frontiersin.org/articles/10.3389/fpubh.2026.1902317/full
AB - Background: Exercise knowledge is seldom evaluated as a standalone construct in China, and no validated instrument exists for self-perceived mastery of exercise knowledge among patients with type 2 diabetes mellitus (T2DM). Adapting the Turkish Diabetes Exercise Knowledge Scale (DEKS) required item consolidation, supplementary items and a revised response format, yielding a modified, DEKS-derived instrument rather than a direct translation. Methods: This sequential exploratory mixed-methods study was conducted at a tertiary hospital in Hangzhou, China (February–July 2025). In total, 438 patients with T2DM completed the 17-item DEKS-C following forward-backward translation, content validation and three rounds of cognitive interviewing ( n = 23). Internal consistency and the exploratory factor structure were evaluated using Cronbach’s α and principal axis factoring with oblimin rotation; factor retention followed Horn’s parallel analysis, with the three-factor solution re-examined on a polychoric matrix. Results: Content validation consolidated nine overlapping items into three and added three new items. Cognitive interviewing then consolidated or relocated four items and merged one into the foot-care item, producing 17 items scored 1–4 (total range 17–68). Cronbach’s α was 0.967 (corrected item-total correlations 0.668–0.837), the Kaiser-Meyer-Olkin measure 0.965, and Bartlett’s test χ ² (136) = 6945.15, p < 0.001. Floor and ceiling effects were 13.47 and 8.90%, arising entirely from the 59 all-lowest-response and 39 all-highest-response respondents, the former answering faster (median 89 vs. 232 s, p < 0.001). Horn’s parallel analysis identified three substantially intercorrelated factors (|r| = 0.51–0.79) explaining 55.73–71.23% of total variance (71.23% in the full sample), corresponding to exercise benefits cognition, exercise implementation and exercise safety; an equivalent structure emerged on a polychoric matrix. One pattern coefficient nevertheless exceeded unity and inter-factor correlations were elevated, leaving the solution provisional. Conclusion: The DEKS-C showed high internal consistency and a three-factor grouping that was broadly stable across sensitivity samples and alternative correlation matrices, supporting content relevance. However, because the factors were substantially intercorrelated and the structure remains unverified, the proposed multidimensional model cannot be regarded as firmly established; independent multicenter and multiregional confirmatory factor analysis is required.
SN - 2296-2565
ER -