Performance assessment in the context of multiple objectives: a multivariate multilevel analysis.

Area of Science:

• Public Administration
• Health Services Research
• Econometrics

Background:

• Public sector organizations often pursue multiple, potentially conflicting objectives, complicating performance assessment.
• Evaluating objectives in isolation overlooks interdependencies, while single performance indices require subjective weighting.
• Accurate performance measurement is crucial for accountability and resource allocation in public services.

Purpose of the Study:

• To develop and apply a method for assessing the performance of public sector organizations with multiple objectives.
• To explicitly account for correlations between different performance objectives.
• To compare the results of a multivariate multilevel modeling approach with traditional methods.

Main Methods:

• Utilized a multivariate system of equations to analyze 13 performance objectives simultaneously.
• Employed hierarchical data structure, nesting electoral wards within health authorities.
• Applied multivariate multilevel models and compared estimates with ordinary least squares (OLS) and standard multilevel models.

Main Results:

• Evidence of significant correlations between performance objectives was found, indicating complementary and trade-off relationships.
• Multivariate multilevel models produced different performance estimates compared to OLS and standard multilevel models.
• The magnitude of performance differences varied across objectives and individual health authorities.

Conclusions:

• Multivariate multilevel modeling provides a more comprehensive approach to assessing public sector performance by considering objective interdependencies.
• Ignoring correlations between objectives can lead to inaccurate performance evaluations.
• The findings highlight the importance of using appropriate statistical methods for performance analysis in complex public service environments.