Nurses' self-efficacy and outcome expectancy in evidence-based practice: Translation, construct validity and internal consistency of the Dutch scales

Background

Evidence-based practice (EBP) is crucial for appropriate, effective, and affordable care. Despite EBP education, barriers like low self-efficacy and outcome expectancy limit nurses’ engagement in EBP. Reliable scales are essential to evaluate interventions aimed at improving self-efficacy and outcome expectancy in EBP. The English Self-efficacy and Outcome Expectancy in EBP scales are psychometrically sound.

Objectives

To describe the translation, construct validity and internal consistency of the Dutch Self-efficacy and Outcome Expectancy in EBP Scales.

Method

The scales were translated forward and backward, piloted for comprehensibility and completeness and then administered among Dutch nurses and nursing students.

Results

Pilot testing confirmed comprehensibility, completeness, and relevance of the items. Confirmatory factor analysis (CFA) (n = 769) tested a second-order model for the Self-efficacy scale (Comparative Fit Index (CFI)=0.96, Tucker-Lewis Index (TLI)=0.95, Root Mean Square Error of Approximation (RMSEA)=0.06, Standardized Root Mean Residual (SRMR)=0.04) and a single-factor model for the Outcome Expectancy Scale (CFI=0.99, TLI=0.99, RMSEA=0.06, SRMR=0.01). Chi-squared tests remained significant. Hypothesis testing confirmed construct validity of the Self-efficacy (r = 0.77) and Outcome Expectancy Scale (r = 0.74). Both scales exhibited high internal consistency with McDonald's Omega and Cronbach's Alpha values above 0.95.

Discussion

Both scales exhibit theoretical soundness and positive fit indices. Significant chi-square tests and high correlations between weighted and unweighted scores support using unweighted scores over utilizing the estimated model to calculate weighted scores.

Conclusions

Construct validity and internal consistency of the Dutch Self-efficacy and Outcome Expectancy in EBP Scales are good. Future research should prioritize responsiveness and test-retest reliability.