Analysis Report On Findings
As stated in the Part A, In order for the college administrators to monitor the salary differences. The administrator will need to set up a timely monitoring of the graph, how the graph populate, this will give them automatic insight into determine the salary differences as any changes in salary would populate the graph and they will notice it automatically by timely visualization. The Salary increase could be determined basically by visualizing staff rank on the graph, the moment there is a change in rank, it will result in increment on the graph length, It is more easier to visualizes using the histogram than using the box plot, though box plot could be used in reaching other results in more real time, in event of determining the staff numbers. It is easy to know this through the scatters plot. Professors receive higher salary at the peak of their careers why it there is evidence of drop in salaries earn towards retire. All this were made possible through data set visualization using the histogram, the salaries could be determined by the years spent in service and rank. Association professors get the median salary which is based on the rank; it follows that of the professors. Administrators can easily utilize this analysis in planning budgets, monitoring expenses and also to have more insight into their operational cost and staffs per each rank and the salary per each rank without having to personally audit the account, this could be done by utilizing the data analytics program which could also be programmed and run at end of each time to determine changes in rank and which automatically results in salary increase and this also inform the urgent need for the administrators to adjust the overall budget to meet the salary increase of staff promoted to the next rank. This is the best methods, to set up a timely monitoring of the graph, data generated from salary records will be analysed and it is going to automatically help the administrator to determine the increase in salary by change of rank and also determine the salary increase by promotion in rank. This are the key attributes to determine the changes in salary. The above graph plotted using histograms, boxplot and gg-plot were utilized in sorting the data and plotting them in a physical shape in order for the analyst to visualize and derive information requested from the data. The Raw dataset withhold several information and there are several algorithms to arrive at the desired results, the algorithms are utilized in gathering desired information and results from data, to determine the correlation in data sets. In this scenario, the results from analysis are discovered after undergoing several algorithms methods that correlate with the desired result.
The probability of problem to come from this arrangement is if staff are promoted and the administrator does not deemed the staff to be worthy of salary increment as part of their policy, It will be difficult to adjust automatically from generated data set if auto analytics tools are being utilized to gather intelligence on the data set generated for staff ranks and salary. Some staff are promoted in organizations but the organization does not necessarily want to increase the salary due to the fact that they are not on merit or deemed them to have earned enough already. Certain measure needs to be considered as condition that certain staff contributed to merit salary increase in the event of promotion, it could either be based on years in service or days spent in the institution. All this has to be considered and provided for in to maintain budgetary integrity and compensated staffs base on their level of commitment and years spent. Not necessarily by rank. Staff with lower rank could be compensated with higher salary base on the volume of their commitment. The database needs to be structured with this condition to effect the real importance of compensation and increment base on commitment and contribution. The data structure is not robust enough to give the administrator more leverage and flexibility to alter decision based on some other staff attributes beyond ranks and years in service. These need to be considered as a matter of important and urgency to have a unique system of staff monitoring, payment, compensation, salary increment and population by rank.
The purpose of the overall statistical analysis is to have tangible insight into the raw date (school record) and to discover the pattern and changes in them by using the algorithms provided by the R-Studio. R-Studio is a programming Analytical software but there are several other software that are less complex to achieve the same goal but for purpose of root knowledge, R is good for having the root knowledge of the concepts and analysis terms. The data are programed into graph, the graph could be either used to analyse it or to aggregate data for purpose of sense making. The data sets are used in populating the graph and this reveals hidden information in the graph. It could be used to accurately determine the mean, median and mode and this is utilized in quantifying the purpose of the analysis and identifies the accurate result for decision makers. In the University record scenario, It was utilized in analysing stored records about the staffs and this data could be used automatically through the analysis and use of graph and boxplot to determine when a lecturer is promoted and results in increase in salary and also to be utilized in detecting sets of staff that earns more and the one that are average and more. It is very important to analyse data gathered on staff, in order to have accurate records and also have deep understanding when lecturer salary need to be increased as a result of promotion in rank just by using the graph plot through analysis of their records. Data are infinite, it is more difficult for decision makers to make effective decision without having the data sorted and analysed and setting auto alert in their data base in the event that there are changes in individuals records which need attention and this can only be achieved through data Analysis.