By Edwin Musonye
Despite many organisations, and even government agencies, talking considerately on virtues of data-backed planning and action; very few utilise the existing data for their operational needs. In the past, it was justified since statistical modeling was done manually and was prone to time waste, many errors and extreme costs.
However, currently there is an upsurge of computational power and therefore, no genuine excuse can be tolerated for the disregard. The only hurdle is in the possession of skills for competently handling data and datasets; relevant statistical software for analysis; equation and path modeling; and interpretation. This is through quality training, extensive practice and higher professionalism.
At Document Point, using Kenya National Bureau of Statistics’ (KNBS) 2019 data, the above model was effectively estimated. The layout was manually accustomed to center the core economic factors under
the study. These factors are Output, Intermediate Consumption, Value added, Payroll and Operating Surplus. The
model thus represents the impact of the various economic sectors on the selected factors (denoted by the red nodes). This means the distances between the nodes bear no relational importance.
The aim of statistics and data visualisation is to assist decision makers in seeing beyond the obvious relationships and patterns. Whereas, pure statistics give results in numerical values only; data visualisation takes over to diagrammatically display what is really happening. The above visualising has avoided numerical quantification of relationships. Instead it used directional arrows (of varying weights) to represent levels of
impact; and different colours to separate nodes representing factors from those representing sector variables.
Interesting insights are drawn from the data loadings. Yet, an assumption is made that the data as obtained from the source is complete and accurate.
Standing like a loner outliner among the five economic factors, payroll is highly marginalised. Workers compensation is unfelt across all sectors. This could be because of any of the following three reasons,
Remuneration is greatly pegged on productivity. There may exist a reality that Kenyan workers are not very productive. They are not creating sufficient wealth as to justify noteworthy recompense. Nevertheless, workers’ productivity also relies on tools of work. If they are inefficiently resourced, their output -regardless of high training- will be compromised.
Workers are productive but are being unfairly ill-used by employers due to the high unemployment levels. It may be too