00:00:01 Introduction

Today, we are going to go through the following topics. When we introduce a system, this is a big change within an organization. So we're going to see what are the forces of change and resistance to change in this aspect. Then, in order to implement any case management system, we want to consider the steps that we need in order to build, launch, and maintain an information system that supports case management. That is why we will describe the steps, and one core component of those steps is the data model, and that is why we will place specifically a big part of the presentation today on that.

Then best practices on data are interconnected with the model. At the end, we will go through best practices for sustaining system use. So we will go more in depth on capacity building, which is core, as well as other resources that an organization needs to maintain in order to ensure that the system actually can be implemented, launched, and made successful.

00:01:21 Introducing a case management system

So let's see what happens when we use a case management system. When we introduce a case management system, we introduce a change within an organization. We will leverage the idea of Force Field Analysis to describe how an organizational change occurs. For the record, Force Field Analysis was created by Kurt Lewin in the 1940s. The idea behind it is that situations are maintained by equilibrium between forces, where some are for change and others are resistance. For change to happen, the driving forces must be strengthened or the resisting forces weakened.

Imagine that the proposed change is the introduction of a new information management system for case management. We have valid forces for change, such as having a common system, saving time in data collection and analysis, and implementing standards. However, we have resistance when we introduce a change. This may reflect past negative experiences, lack of resources, lack of timing and staffing, plus lack of trust. We may find ourselves in a position where we do not trust the senior management. There is also loss of control over the new situation, fear of failure, and the unknown.

We need to strengthen the forces for change and we need to weaken the forces of resistance. To begin with, we can build an internal Advisory Board, and we can create an internal user community. We need capacity, which is crucial. We need to invest in continuous capacity building and consider different kinds of capacity that is needed for administrators, data entry users, or field staff. We need to have an internal resource center where we have available guides and manuals. We need to celebrate wins and spread success stories using visible internal channels. To weaken the forces resisting change, we need to communicate; we need a communication plan that outlines key messages and timings. We need to have support in place, creating multiple channels for receiving and providing feedback.

Capacity building is core when we introduce a change within an organization. Under the design phase, we have our Standard Operating Procedures (SOPs). Those refer to different aspects of case management, such as the protocol case management workers follow, how they communicate with beneficiaries, and what type of data they collect. The MEAL system design and information management system design match them as we work close to one another. When we are preparing for launch and adoption, we need to capacitate both end users—like caseworkers and supervisors—but also those that administer the system, such as IT, the Information Management Officer, or program managers. The increased capacity leads to real-time data use and creates incentives for system use, which is core to sustainability.

00:06:39 Design and roll out a case management system

As we mentioned in the previous webinar, the MEAL standards exist within a system. We need a design based on context analysis and specific aspects of the country where we are implementing. The service is guided by documented processes and protocols. We need appropriate staff and budget. We can identify two main drivers for implementation: organizational strategy and the need to use the data collected to support the beneficiaries timely.

Let's explore the specific steps. First, we need to have a clear vision. We use a term from monitoring and evaluation: Theory of Change—how and why a desired change is expected to happen in a particular context. As part of this, we need to specify the level of implementation. Do we implement at a project level, program level, regional, or organizational level? Do you have any critical hypotheses, such as lack of specific policies or lack of staff? Finally, what is our main objective? For example, we may need to improve evidence-based decision making within the projects.

The specific steps involve assessing the needs at the user level, designing a system tailored to those needs, launching within a specific timeframe, and finally, the adoption phase. During assessment, we identify the implementation level and critical hypotheses. The output is the Theory of Change. We need to identify data requirements, data collection methods, and project requirements like offline collection. The outcomes of this process are the system requirements, the data flows, and finally, the data model.

We translate the output of the previous step into the design step. The data model is translated into the collection forms. The data flow is translated into roles and permissions. These two components lead to the database design. From the database design, we move to the report design where learning requirements are translated into reports. Following this, we evaluate the capacity our administrators need. During launch, we identify capacity building approaches for end users. The output is training sessions and manuals. During the adoption phase, we monitor system use and actively seek to collect feedback to improve both the system and accompanying policies.

00:12:25 Data model as a key to structure data

The case management reality is complex, emergent, non-linear, and adaptive. Within that system, we need the information management system to help us organize and define the data we collect. Here, the key is the data model. A data model is a visual representation of a conceptual framework that organizes and defines the data elements, showing how they interact with each other. By mapping out data structures and relationships, we understand how data is stored, organized, and retrieved.

Let's imagine that a relational data model is the same as a house blueprint. When we design a house, we meet with owners to gather requirements like the number of rooms and functional requirements. When it comes to data modeling, we meet with stakeholders for system requirements. Data requirements define the main data sets and the association across them. User requirements define the data flow—who uses which data collection form to perform which tasks.

In a house, you have various rooms like a kitchen or bedroom. Similarly, in the data model, we find different entities, such as a beneficiary or referral. The entity is the main unit of operation around which data are being collected. Each room has specific characteristics; in the data model, we use the term attribute. These can be the beneficiary ID, name, age, and gender. For specific characteristics, we select materials to ensure quality, which is equivalent to specific settings that establish restrictions, like validation rules. Hallways connect the rooms; similarly, the data model defines relationships. For example, a beneficiary receives multiple services.

Normalization is key to the relational data model. It supports the division of large tables into smaller tables to reduce redundancy and ensure uniqueness. It also defines the relationships across tables to ensure data validity. For example, if we have a table with beneficiary names and referrals, fields should contain atomic values—first and last names should be separate. We also need to avoid transitive dependency and data redundancy. To solve these issues, we reconstruct tables so that each record depends on a unique identifier. We might create a beneficiary intake table, an assignment table, and a referral table, all connected via relationships (one-to-one, one-to-many, etc.).

00:26:50 Best practices on planning for data use

Planning for data use starts in the assessment phase, specifically when we have the data model ready. Under the data model, we identify four components: data entities, data sets, attributes, and relationships. The data model enables us to identify the data needed for the reports. This acts as a reality check on whether we missed any data entity or data set.

We define learning and reports by breaking them into four components. The first is analysis: defining the type of measures needed. The second is the audience: the stakeholders who receive information. The third is communication: how we communicate findings (visuals, tables). The fourth is dissemination: how we distribute findings.

Regarding measures, we have measures of frequency (count, proportion, ratio, rate), measures of central tendency (mean, median), measures of position (minimum, maximum), and measures of variability (range, variance). For audiences, field staff need simple reports to monitor day-to-day work. Project staff need increased interactivity, such as dashboards with filter options. External audiences like donors may need public availability or specific reports.

For communication, we use visualizations. Line charts are recommended for comparing items over time. Bar charts show distribution. Pie charts describe composition. Scatterplots show relationships between variables. For dissemination, the most recommended option is to create a live report updated in real-time. We can set up roles within our database for internal sharing, or use public availability for external sharing with donors or partners who may not have database access.

00:40:56 Best practices for sustaining system use

The key component for sustaining system use is capacity building. We can approach this using the "Golden Circle" framework: Why, How, and What. The "Why" is the purpose driving our actions—for example, we want to reach beneficiaries at scale while maintaining quality. The "How" refers to specific strategies and competencies, such as capacity in information management system design. The "What" represents the processes and steps, like specific training on data models or privacy procedures.

When designing training, we need to organize a custom plan. Before the training, we create a training outline and a detailed facilitation guide. During the session, we tailor content to learning necessities. A good facilitator asks open questions, ensures full participation, uses body language effectively, and employs a variety of facilitation techniques like case studies or simulations.

We have two major audiences for training: administrators and end users. Administrators need advanced skills in system design and management. End users need training on data collection and adoption. Following the training, we must conduct an evaluation to measure effectiveness, satisfaction, relevance, and learning outcomes.

Finally, we need to maintain a library of additional resources. This should include organizational-level processes, available tools, relevant guidance, focal points, and templates. Key takeaways include: capacity building ensures consistent implementation; the Why/How/What framework helps design educational material; and the relational data model is key to simplifying complex reality.

00:49:18 Q&A and Conclusion

Host: Thank you, Eliza. This has been a lot of information, but really useful. Let's give the audience some time to digest all this information. In the meantime, I've shared in the chat various resources. You can find the recordings of past webinars on our website. You can browse topics such as case management, information management, and many more.

Host: I don't see any questions yet. Perhaps people are getting ready for the end of the year. Okay, then maybe we can wait another two or three minutes.

Host: There was a question regarding the recording. Yes, the recording and the presentation will be available on our website and we'll also share it via the email you used to register.

Host: All right, since we don't see any more questions, we're wishing you a great new year. Thank you for joining this session and the past sessions. Thank you, Eliza, for the presentation. Keep in touch. Thank you everyone.