mrvillafria / pewlett-hackard-analysis

Creating entity relationship diagrams, performing data modeling, and completing analysis on an employee database using SQL techniques

Employee Database with SQL

Overview of Project

For this week's project, we will be using Structured Query Language, also known as SQL, to organize and query our data. SQL is one of the most commonly used query languages that will help us perform our analysis to discover relationships between our data in a database. In order to design and build our database, we will also need to understand how to organize our data. To visualize the flow of information, we will create an entity relationship diagram (ERD) that will show the database as a whole but also highlights the tables and their relationships. Although there are many different interfaces to use SQL, we will be using Postgres and pgAdmin in this project. We'll be using Postgres as our relational database system. PgAdmin is the user-friendly interface that interacts specifically with Postgres and where we will be writing and executing our queries.

Purpose

The purpose of this week's analysis was to help Pewlett Hackard organize and analyze their employee data. Bobby, an up-and-coming HR analyst, was tasked with finding the amount of retiring employees at Pewlett Hackard so they can prepare to fill these positions. Bobby's boss determined that anyone born between 1952 and 1955 will begin to retire. The company would also like to implement a mentoring program that will allow new associates to learn from these experienced employees. Instead of retiring, these employees would be working part time to mentor the newly hired.

Pewlett Hackard was mainly using Excel and VBA to view their data and needed our assistance in building a new employee database. The employee information that was available was stored in 6 different CSV files. These spreadsheets consisted of information around departments, employee information, employee salaries, managers, department employee information, and employee titles. We were able to use these spreadsheets and export the data into tables in our new database. We used SQL to write queries that retrieved employees eligible to retire and their titles, employees eligible to retire and their most recent title, number of employees eligible to retire by title, and employees who are eligible to participate in a mentorship program.

By putting Pewlett Hackard's employee data into a new database, their data is now more organized and easier to query for information. Creating queries will provide more insight to decision makers and managers to help future proof Pewlett Hackard.

Results

Please feel free to reference any of these SQL statements from this analysis at the following link:

Employee Database Challenge Queries - Link

Retirement Titles

The information in this table below consists of employees eligible to retire and their titles. As mentioned previously, Bobby's boss has determined that any employee born between 1952 and 1955 will start to retire. This table will be the foundation to our next two queries. In order to retrieve this information based on the data we had, we joined the existing Employees and Titles tables on their primary key, which is the Employee Number.

Retirement Titles table consists of the following data:

• Employee Number
• Employee First Name
• Employee Last Name
• Employee Title
• Start Date of Title
• End Date of Title

Retirement Titles Export - Link

Unique Titles

The information in this table below consists of employees eligible to retire and their most recent title. We noticed in the Retirement Titles table that there were duplicate entries because employees changed their titles over the years. In order to remove these duplicates and only keep their most recent title, we used DISTINCT ON in our query to retrieve the first occurance of the employee number.

Unique Titles table consists of the following data:

• Employee Number
• Employee First Name
• Employee Last Name
• Employee Title

Unique Titles Export - Link

Retiring Titles

The information in this table below consists of number of employees eligible to retire per title. We used COUNT to count the number of employees and grouped the table by title.

Retiring Titles table consists of the following data:

• Count of Employees
• Employee Title

Retiring Titles Export - Link

Mentorship Eligibility

The information in this table below consists of number of employees eligible to participate in a mentorship program. These are current employees that were born between January 1, 1965 and December 31, 1965.

Mentorship Eligibility table consists of the following data:

• Employee Number
• Employee First Name
• Employee Last Name
• Employee Birth Date
• Start Date of Title
• End Date of Title
• Employee Title

Mentorship Eligibility Export - Link

Analysis

Based on the new tables we created, here are four key points we discovered:

1. There are 90,398 total employees born between 1952 and 1955 that are expected to retire soon.
2. By looking at the employees' most recent titles, we find that majority of retiring employees have Senior titles or are Engineers.

3. There are 29,414 Senior Engineers and 28,254 Senior Staff that are eligible to retire. Their Senior title gives an indication that they are experienced in their current role which would be a desirable quality for being a potential mentor.
4. There are 1,549 employees that fit the criteria for being a mentor in a mentorship program.

Summary

"Silver Tsunami" Summary

As the "silver tsunami" begins to make an impact, there are a few key points we can look at when determining how many roles will need to be filled. Based on our analysis, we can see there are 90,398 total employees born between 1952 and 1955 that are expected to retire soon. This does not necessarily mean all of these employees will retire. We would need to do additional research to find statistics around how many people continue to work through retirement. Additionally, we could do more research on average age of retirement. It would also be a great idea to have the company evaluated to see if all of these roles are actually needed.

An additional query to run would be to find employees that still work for the company that are eligible to retire and their most recent titles. The original query did not take into account if the employees still worked there. We find after adding the filter to see if they still worked at the company, there are actually 72,458 employees instead of 90,398.

Unique Titles Currently Employed Export - Link

We also created a new count of employees eligible to retire per title based on if they were still employed.

Mentorship at Pewlett Hackard

When looking into potential mentors for the next generation of Pewlett Hackard employees, it would be ideal to have them be experienced employees. Based on our analysis, there are 1,549 employees that fit the criteria for being a mentor in a mentorship program. There is a potential to fill 72,458 roles if all of the current employees eligible to retire plan to retire in the same year. If we divide the amount of potential employees retiring by the amount of mentors, this could mean every mentor would have around 47 employees that they would have to train.

An additional query to run would be to find the amount of mentors per title. This way we can make sure we have enough mentors to train the new employees who will be filling their specific title.

After running the additional query to see the amount of mentors per title, we will need to consider expanding and editing our criteria for mentors. There is a significant amount of potential employees retiring and not enough mentors per title based on our numbers. Instead of looking at birth year for criteria, we could potentially look at the amount of time an employee has spent in their current role. We would need more information on when and what year this data was pulled so we could run more specific queries. It would be best to do more research into what makes an employee a good mentor such as time in that role and proficiency in their skills.