• Problem to solve
• Data Collected
• Methods – R Studio Software / Microsoft Excel
• Findings
• Conclusion

Problem to solve

This Tech Sales Report details the performance of the sales representatives of a Tech Company for
2021. It contains the methods used to evaluate the salaries of 12,130 sales representatives
from the software product group of a high-tech company. For each employee, the data include
socio-demographic and education information, salary, and a personality indicator. Also included in
the data is the net promoter score, which is an indicator of customer satisfaction with each sales
rep.

There are two parts of the project where we will focus only on the software branch of the firm.

Being the first part the data visualization and summary measures in which we will inspect and

review the data in order to prepare a report for the company’s upper management based on the

four personality types in the software product group. The four personality types are Analyst,

Diplomats, Sentinel, and Explorer.

The second part of the project will consist of a Linear and Nonlinear Regression Analysis that will

predict the salary of the respective sales representatives by creating dummy variables in order to

capture different predictor variables of the model and the effect of the personality into the model.

Data Collected

We are provided eleven (11) different data values about employees which included: Sales Rep ID,

sales division, age of employee, gender, years at the company, education level, personality type,

certificates, feedback, net promoter score and the employees salary base.

Personality Group 1: Diplomats and Explorers

Personality Group 2: Sentinel and Analyst

Methods – R Studio / MSO Excel

In order to do the pertinent analysis, the tools used are:

Microsoft Excel to do the initial analysis of the model. Since the software data is stored in an Excel sheet to create pivot tables and scatter plots to analyze the relationships of different values to each other.

Aside from this, the tool will be used in order to create the Regression Analysis by selecting the Y variable “Salary” and the X variables which will be the predictor variables used in the analysis thatwill determine the level of fit of the different models, signaling which one is the best model. The tool will also be used in order to create the graphs, pie charts, and plots needed to create the data visualization images that will do the infographics.

R studio will also be used as a second tool in the process of creating the Regression Analysis and the selection of the best model.

Findings

The first finding was in excel where we decided to compare personality types with all the other

data points to find any major differences amongst personalities.

Using R we calculated a linear model regression using the personality types as explanatory

variables, first with Sentinel as the reference category and later Diplomat as the reference

category. The p-value showed there is no significant difference within the pairs of the

personality types above grouped.

First we ran a regression using all possible

explanatory variables but using scatterplots

we found that the relationship of Salary and

Age isn’t linear so we decided to include Age

squared as a variable. This final regression

equation got us a higher Adjusted R-Squared

of 0.586

Conclusion – Results

After analyzing the data, we can conclude that the personality type does play an important role on the average salary obtained from the workers. We find a relationship between the personality type, the average of certificates, the average NPS

and the average salary.

The successful personalities (Diplomat and Explorer) happened to have the highest average of

NPS, the less average of years in the company (As well as age) and the highest average of

certificates, all of this variables together resulted in the highest average of salary for this

personalities, being the “ Explorer ” personality the one with the highest average salary of $76,460.

On the other hand, If we analyze the less successful personalities we find that they have the lower

average of NPS, they’ve spent more years in the company, happen to be older (This fact is relative,

since there’s not a big difference in age between the personalities), have worse feedback from

managers and in average have less certificates than the successful or better paid personalities. We

can see their influence in the average salary that they receive is almost $14,000 of difference from

the successful personalities.We must also mention that when it comes to the disparity in gender,

the women seem to be the ones losing. The men have worse NPS scores, worse feedback and

less certificates than women and they still receive more money in compensation (More than a

$1,000 difference) and they have a higher average of salary, with a difference of more than $6,000

between genders.

In conclusion, the company must pay more attention to the female gender of the company.

A change on this matter would increase the female productivity and employee retention.

-Alfredo S.