Navigating COVID19 Item Spend Page

Navigating to the second page of the dashboard, this page shows us the payments by procurement items.

Click on column names to sort by the column in ascending or descending order

Once you've navigated to the second page of the COVID19 dashboard named "Items Spend", you can click on the total payments column. This will sort the below table by total payments in descending order. Click on it again, it will sort the table by total payments in ascending order.

For all the tables on the dashboard, clicking on a column header will sort the column.

Click on the Plus to expand

We can see the top paid items are "Others" and Item "0". We might wonder what is this item other and item 0. Note on the left of “Others” there is a small plus button, click on the button will unfold the row. The departments paying for this item will be shown on the table.

Click on a row to filter

We can further click the National department of Health under Item "Others", this will be a filter too, meaning all information on the dashboard will only now be about Department of Health’s Others order.

On the right table called vaccine transactions, we can see an about 540 million Rand payments. This means among the 690 Million payments on item of Others, 540 million are vaccine orders.

Inconsistent Department Names

Under Others, there is "National Department of Health", "Health", and "Gauteng Department of Health".

Is "National Department of Health" the same as "Health"? This reveals some of the data quality issues. From data on the COVID-19 dashboard, we realize that there are usually different entries of names for the same department, which shows inconsistency in the data.

For example, in the COVID-19 transactions dataset, we found four names related to the Education Department, respectively “Basic Education”, “Departmen of Education”, “Department of Education” and “Education.”.

While some of these might seem like obvious typos, others we simply can't assume.

Later we found "Education." is actually Gauteng Education department. We did this by comparing the PDF report published by the department with the orders on the dashboard. So even though "Health" and "National Department of Health" look similar, we can’t simply assume they are the same department. Rather we need to remember the same department is written as different names in the dataset during the analysis.