Page 109 - FULL REPORT 30012024

ii. Malaysia Stroke Mortality Dataset

The data cleaning procedure for the Malaysia stroke mortality

dataset, acquired from the eStatistik database, requires a full
restructure owing to the original condition of disarray and

inappropriate formatting. The dataset revealed a lack of order, with

variables and records not adhering to a defined pattern suitable to
analysis. To remedy this, the dataset underwent a rigorous

rearrangement inside Microsoft Excel. Figure 4.28 displays the
basic dataset before the cleaning procedure.

Figure 4.28 Pre-Cleaned Malaysia Stroke Mortality Dataset Structure.

Columns were designated to represent each pertinent statistic,

including year, number of stroke cases, and fatality rates. The rows
were arranged in a manner that matched each unique record,

creating a consistent grid of data points. The data inputs were
entered manually into the newly organised Excel sheets, which was

an essential step to verify that each piece of data was accurately

placed within the hierarchy of the dataset.

The cleaning method primarily emphasised the reformatting and
rearranging of the dataset. The process included standardising the

headers to provide clarity and uniformity, which is crucial for any

subsequent analytical undertaking. The cell formatting was
92