Peer groups for organisational learning: Clustering with practical constraints.

Daniel W Kennedy,Jessica Cameron,Kerrie Mengersen,Paul P -Y Wu,Renato Cordeiro De Amorim

doi:10.1371/journal.pone.0251723

Abstract

Peer-grouping is used in many sectors for organisational learning, policy implementation, and benchmarking. Clustering provides a statistical, data-driven method for constructing meaningful peer groups, but peer groups must be compatible with business constraints such as size and stability considerations. Additionally, statistical peer groups are constructed from many different variables, and can be difficult to understand, especially for non-statistical audiences. We developed methodology to apply business constraints to clustering solutions and allow the decision-maker to choose the balance between statistical goodness-of-fit and conformity to business constraints. Several tools were utilised to identify complex distinguishing features in peer groups, and a number of visualisations are developed to explain high-dimensional clusters for non-statistical audiences. In a case study where peer group size was required to be small (≤ 100 members), we applied constrained clustering to a noisy high-dimensional data-set over two subsequent years, ensuring that the clusters were sufficiently stable between years. Our approach not only satisfied clustering constraints on the test data, but maintained an almost monotonic negative relationship between goodness-of-fit and stability between subsequent years. We demonstrated in the context of the case study how distinguishing features between clusters can be communicated clearly to different stakeholders with substantial and limited statistical knowledge.

Highlights

Organisational benchmarking has become increasingly widespread across industries, including health, education and government [1]
We found that applying the same algorithm over successive years did not necessarily produce highly similar sets of peer groups
From a principal component analysis (PCA) analysis we found that clusters 1 and 2 had heavy overlap in the main directions of variation in the data, cluster 1 was more concentrated for some principal components

Summary

Introduction

Organisational benchmarking has become increasingly widespread across industries, including health, education and government [1]. An inherent consideration in any benchmarking exercise, which is suggested as a means for organisational learning and quality assurance, is that of comparing and contextualising an organisation’s performance against its peers, known as peer-grouping [2, 3]. Identifying relevant peer organisations is challenging due to the variability and uncertainty in the mix of clients and services provided by each organisation [4,5,6]. Unsupervised clustering methods might appear as the solution to the peergrouping problem, a key challenge to practical uptake lies in the management context where clustering outcomes need to be interpretable by stakeholders of varied backgrounds.

Methods

Discussion

Conclusion