Psychometric Testing of SPIDER: Data Capture Tool for Systematic Literature Reviews

• Autores: Sherrilene Classen, Sandra Winter, Kezia D. Awadzi, Cynthia W Garvan, Ellen D.S. Lopez, Swathy Sundaram
• Localización: American Journal of Occupational Therapy, ISSN 0272-9490, Vol. 62, Nº. 3, 2008, págs. 335-348
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • OBJECTIVE. Systematic literature reviews contribute to evidence-based occupational therapy, yet no data capture tool currently exists to validly and reliably appraise the characteristics and quality of primary studies.

    METHOD. We determined the psychometrics of Systematic Process for Investigating and Describing Evidence-Based Research (SPIDER) and piloted it with 201 studies included in a systematic literature review.

    RESULTS. Content validity showed item relevance with 73% agreement between two experts. For the quality construct, seven of nine quality indicators were positively (p < .05) correlated with the overall quality score. The quality scores were positively correlated (p < .05) with two objective measures, inferring criterion validity. Intrarater reliability was moderate to perfect (? = 0.4�1.0). Cross-tab analyses showed less variation in experienced reviewers� interrater reliability.

    CONCLUSION. SPIDER provides plausible opportunities for occupational therapy researchers and graduate students to appraise the characteristics and quality of primary studies but requires testing across other settings