Blog

Part 3: Hoji as an Alternative Tool for Data Verification in DQA

Posted by: Gitahi Ng'ang'a on

In the first part of this series, we discussed the importance of data verification as a critical component of DQA. In the second installment, we talked about Microsoft Excel as a tool for conducting data verification. We discussed why it is currently the most popular choice and examined some of the challenges associated with it. In this third and final installment, we shall discuss our easy-to-use and robust mobile data collection and analysis technology as an alternative to Microsoft Excel. We shall demonstrate how it delivers all the advantages of spreadsheets without the drawbacks identified in part two. If you haven’t already read the preceding installments of the series, please do so before continuing.

Mobile data collection and analysis

Now, if you do not already know about Hoji, it is a mobile data collection and analysis platform. As a company, Hoji Ltd was founded in Nairobi in 2014 as an easy-to-use, robust and comprehensive mobile data collection and analysis solution. If you are familiar with ODK or KoboCollect, you have a good idea what I am talking about. Or do you?

We set out with the vision is to close the gap between data gathering and decision making by collapsing collection and analysis into one real-time process. We have come a long way since then, and take pride in being able to deploy even the most complex surveys within a few days. Simpler surveys can be deployed in a matter of hours. But, we also discovered that there was a need to expand the scope of our solution beyond surveys to support DQA and routine M&E.

The first time we encountered the DQA use case was in 2016 when working with the National AIDS and STI Control Program (NASCOP) in Kenya. Through discussion with subject matter experts there, we quickly found out that even our own solution fell short of meeting their needs. While Hoji already provided fairly rich data analysis features especially for household surveys, we couldn’t at the time customize the analysis we provided or “integrate” it with secondary data sources like the DHIS and DATIM. Yet this is what the problem at hand demanded. So we got to work.

Having already introduced DQA and data verification and covered Microsoft Excel as the most common tool used to solve the problem in the first two parts of this series, I will proceed directly to discussing how choosing Hoji as an alternative addresses the issues associated with spreadsheets.

If you may recall, national programs favor using spreadsheets for DQAs because they support both data entry and data analysis. As such, DQA data can be entered and analyzed at the facility, allowing for immediate results dissemination and review.

The first challenge with this approach is that you end up with as many workbooks as the facilities you sample, presenting a major merging problem before sub-national and national results can be aggregated and analyzed. The second challenge is that even data that already exists in digital format from the DHIS and DATIM must also be manually entered, making the process not just inefficient but also error-prone. These are the problems we set out to address. And we did.

Automatic data collation

The first part of our solution is that, as a mobile data collection system, Hoji works both online and offline. When online, data is submitted to a central server in real-time. Otherwise, it sits on the device until connectivity is restored. Ultimately, all data collected through the system is automatically collated in one database, thereby eliminating the merging problem created by using disjointed Excel workbooks.

On the server, Hoji provides sophisticated data analysis and visualization features, including a pivot table utility that allows the dataset to be sliced and diced in almost limitless ways. The data can also be easily filtered and aggregated, greatly simplifying the process of preparing facility, sub-national and national reports. Highly flexible calculations allow data managers to define calculated variables, such as summing up values for the same indicator across different age and sex disaggregations. Best of all, these summaries and analysis can be prepared in advance for later visualization in interactive read-only charts. Besides eliminating the need for merging data, these features also make it impossible for field officers to tamper with the formulas used to generate the DQA reports.

Preloading DHIS and DATIM data

The second part of our solution involves the ability to preload the values that are already digitally recorded elsewhere, in this case, data from both the DHIS and DATIM. As with any other mobile data collection tool, Hoji provides the means to preload onto the mobile app the list of all health facilities to be visited. Additional information such as counties and sub-counties may also be included. More interestingly, Hoji also provides the means to associate every single program indicator for each month under review with its preloaded value from the DHIS or DATIM.

What does this mean?

This means that field officers only need to enter the indicator values for the data sources at the facility, in this case, the Register and the MOH 731 summary form. This is important because it achieves two things. Firstly, it reduces the amount of data entry work for field officers by at least half. Secondly, and more importantly, it keeps the data verification process itself from introducing errors, since the preloaded values cannot be changed and are necessarily accurate unless the import and export process from DHIS is somehow compromised.

Efficiency, quality and cost saving

Through these improvements over Excel-based data verification, national programs like NASCOP have been able to execute large-scale DQAs efficiently and have preliminary results ready in real-time and the final report within a matter of weeks. Results at individual health facilities are also disseminated in real-time, without the error-prone process of tinkering with complex Excel formulas.

On-the-spot DQA results dissemination meeting at a health facility using Hoji.

On-the-spot DQA results dissemination meeting at a health facility using Hoji.

Using Hoji instead of Microsoft Excel for data verification and DQA as a whole affords our clients enhanced efficiency, data quality and cost savings. If you have any questions or comments, please leave them below. We shall get back on each one of them. If you are convinced and would like a demo or to deploy Hoji for your next DQA, don’t hesitate to contact us.

Leave a Reply