Measuring midwives' perceptions of their practice climate across racial‐ethnic identities: An invariance analysis of the Midwifery Practice Climate Scale

• Published in: Research in nursing & health Vol. 46; no. 6; pp. 627 - 634
• Main Authors: Thumm, E. Brie, Giano, Zach, Niles, P. Mimi, Smith, Denise, Howard, Bridget
• Format: Journal Article
• Language: English
• Published: New York Wiley Subscription Services, Inc 01-12-2023
• Subjects: Burnout; Confirmatory factor analysis; Diversification; Employee retention; Ethnic identity; Health disparities; Health status; Hispanic people; Maternal characteristics; Measurement; Measures; Midwifery; Midwives; Perceptions; Workforce
• ISSN: 0160-6891; 1098-240X
• DOI: 10.1002/nur.22349

Abstract
Diversification of the midwifery workforce is key to addressing disparities in maternal health in the United States. Midwives who feel supported in their practice environments report less burnout and turnover; therefore, creating positive practice environments for midwives of color is an essential component of growing and retaining midwives of color in the workforce. The Midwifery Practice Climate Scale (MPCS) is a 10‐item instrument developed through multiphase empirical analysis to measure midwives' practice environments, yet the MPCS had not been independently tested with midwives of color. We conducted invariance analyses to test whether latent means can be compared between midwives of color and non‐Hispanic White samples. A step‐up approach applied a series of increasingly stringent constraints to model estimations with multiple group confirmatory factor analyses with two pooled samples. A configural model was estimated as the basis of multiple group comparisons where all parameters were allowed to freely vary. Metric invariance was estimated by constraining item factor loadings to be equal. Scalar invariance was estimated by constraining intercepts of indicators to be equal. Each model was compared to the baseline model. The findings supported scalar invariance of MPCS across midwives of color and non‐Hispanic White midwives, indicating that the MPCS is measuring the same intended construct across groups, and that differences in scores between these two groups reflect true group differences and are not related to measurement error. Additionally, in this sample, there was no statistically significant difference in perceptions of the practice environments across midwives of color and non‐Hispanic White midwives (
p
 > 0.05).