Hot Topics
- Home
- /
- Performance Management Overview
- /
- Applying analytics to improve public sector services
Figure 1 – Identifying students who need assistance is an operational activity performed in the primary education sector. The above graphs attempt to show a few parameters that influence student performance in primary education. However, the graphs (as displayed currently) are not adequate to analyze and demonstrate many to many dependencies of these parameters in a manner that can be directly used to decide on educational assistance programs.
Applying analytics to improve public sector services
Introduction
ICT in government is faced with developing comprehensive strategy to seamlessly use information residing in the environment to improve performance of departments while implementing policies and service strategies. However, progress in government performance management has been at foundational level such as investing on conducting citizen surveys or clean accounting audits. Not enough effort has been applied to implement initiatives that can use such foundational data in a manner that provides true insight to administrators.
Challenges in implementing analytics
While authorities use Corporate Performance Assessment Reports & Audit Commission Reports to measure performance and set benchmarks, they lack ability to scientifically analyze underlying data for cause and effect relationships. This is because their current ability restricts them to at best a two-dimensional relationship analysis. However, this is inadequate because such basic analysis cannot reveal dependencies that define complex relationships between data elements in the government environment.
The dependencies of measures in the context of government services are many to many. Manual analysis on multiple dimensions while taking into account effects of all variables on the metric of interest is impossible. Therefore administrators are unable to discern patterns and obtain insight from data.
Measuring performance of Govt. agency
Government departments adopt ad-hoc approach to measure performance of their services. They need access to organized metrics to manage their services better.
Here is a scenario a school administration would face relating to improving the effectiveness of primary education program. Monitoring performance of students and taking appropriate corrective actions is an important activity in a school that needs to be performed on continuous basis. As shown in Figure 1, the school authority analyzes scores obtained by students with respect to various parameters such as education level of parents, nature of surrounding community and exposure of the child to English as a language to arrive at further education assistance programs. Whether these are the right parameters to analyze, and if indeed they are, their degree of influence on student performance is ambiguous. Clearly, these parameters have many to many dependencies which cannot be envisaged and analyzed through two-dimensional plots.
Integrated data vital to service citizen
The push for e-governance has resulted in departmental silo-based ICT infrastructure. As the valuable data is confined within department, there exists disparity in the way multi-agents deliver citizen services.
Solution
Analytics can help administrators identify meaningful relationships and patterns in underlying data to identify issues and opportunities. While analytics cannot replace the knowledge and understanding of humans, it can aid in making effective decisions. Analytics uses a combination of statistics, machine learning, neural computing and artificial intelligence techniques to accomplish this purpose.
Let us look at how the challenge shown in Figure 1 can be addressed. An analytical solution was implemented using Cluster and Discriminant analysis. Education of parents, nature of surrounding community, exposure of the child to English as a language were selected as parameters (variables) for this analysis. Cluster Analysis was applied on these parameters, which grouped the data based on pattern similarity.
Figure 2 – Clustering the students for educational assistance
Data on how classification was done historically for similar performing students along with the results of Cluster Analysis is fed to Discriminant Analysis.
Education of parents and English language fluency were found to have greater significance on educational assistance requirement than other parameters. These two parameters along with other parameters with appropriate weightage were then factored into 2 factor axes namely F1 and F2. Thus the complexity of a multidimensional data set was simplified using these analytical methods and presented in two-dimensions with the help of factor axis. Once the students are classified into different clusters, the programs can be structured differently for each of these groups.
With this approach, school authority will be able to design assistance programs such as English for students speaking other languages, speech therapy, additional mentoring. Similarly crime investigation analysts can make use of database of arrests, crime incidents, past calls to police and then deploy resources more knowledgably to deter crime.
Data visualization is also an important aspect when it comes to data analysis. As implementation of analytics project involves complex statistical functions, it is important to provide the user with simple visual output they can easily understand.
Metrics based performance management
This case study focuses on the solution delivered by EXILANT to one of the premier education institutes. The system will monitor, track and report on project performance to various stakeholders.
Government performance management
Government agencies face a unique set of challenges while managing performance of their service execution. We assist them to exploit technology for managing such challenges.
Summary
For government agencies to be more effective, they need solutions that can analyze past data to anticipate future behavior. We have put together a multi-disciplinary team with expertise in customer service, data analysis, technical implementation, statistics, visual representation and project management. With our expertise in implementing analytics, we are better positioned to assist you with leveraging the technology for performance management needs.
Recent Links :
Measuring performance of Govt. agency
Government departments adopt ad-hoc approach to measure performance of their services. They need access to organized metrics to manage their services better.
Integrated data vital to service citizen
The push for e-governance has resulted in departmental silo-based ICT infrastructure. As the valuable data is confined within department, there exists disparity in the way multi-agents deliver citizen services.
Metrics based performance management
This case study focuses on the solution delivered by EXILANT to one of the premier education institutes. The system will monitor, track and report on project performance to various stakeholders.
