4-2 Project One 1
4-2 Project One
DAT-300
Southern New Hampshire University
4-2 Project One 2
Preliminary data
The data from National Motors offers the information on motor sales from five different sites.
The dataset for National Motors contains motor sales data spanning across 28 months, while the
Kansas City Motors dataset contains data for only 27 months and an addendum for the 28th
month of sales. The additional missing month could be attributed to a data analyst not inputting
the data, or perhaps the tool that Kansas City Motors was using, not performing correctly, or
suffering a system outage.
These two companies utilize different databases for collecting and store their data in different
formats. To merge these systems both National Motors and Kansas City Motors, will need to
switch to one format. Data from both companies must be analyzed and prepared in the same data
format so all information can be stored and later analyzed with the same level of effort. The
Kansas City database was formatted on excel; due to the warehouse’s use of SQL servers. This
causes the data to be stored in a scientific method, which in turn, has caused incompatibilities
between Nation Motors and Kansas City Motors
Data sets analysis
Using the data that has been provided from the two databases has been useful information.
Adding Kansas City motors to the National Motors database would be the easiest transfer, other
processes must be performed to ensure that the data is used correctly. Data cleansing is an
approach used by companies to manage the data that has, been collected review the orders, and
help identify the errors and inaccuracies.
There is always the chance that the data currently in use was entered into the systems
incorrectly or is now inaccurate, depending on the level of human interaction. There is certain
data that needs to be confirmed to ensure the correct data is being entered into the system. In the
4-2 Project One 3
event the Kansas City Motors databases was just copied into the National Motors databases set, a
set of analysts should verify the data accuracy prior to merging. There is additional information
from the National Motors dataset from the 28th month shows that 286 motors were sold at the
Kansas City factory. There might have been an error in the number sold from the previous
months because it seems the sales have either dropped or were entered in error. Causing the
fluctuation of the sales number within the past couple of months.
Completeness and Accuracy
After reviewing the preliminary dataset for quality and completeness. It shows that each value
is recorded in a specific format and that each value matches a set format. The number for the
28th month looked to have been missing from the Kansas City Motors dataset at one point, but a
correction was created to repair that and finish the dataset. It looks like the facility sold 3,000
motors between the 21st and 22nd months of sales. It would be important to verify if sales data
are the same for both months or if there was an error that led to a duplicate entry.
Furthermore, having data stored in a uniform format for analysis is crucial. All tools, program,
and software consist of coding, and a set of “Digital instructions”, which the creators of the
system program construct, in order to instruct the toolset on how to decode/store/process
incoming data. When incompatible systems are used for this scenario, syntax errors can cause
massive problems in the execution of the programs function. “Humans are adaptable. They can
structure a sentence in many ways, and it will still make sense. In contrast, computers require
precise instructions. If you break the rules a bit, a command may make sense to a human, but a
computer won't be able to interpret it” (Muelaner, 2020). Syntax errors simply create an instance
where the computer/tool, does not know how to interpret the incoming data or command, which
is exactly what has happened in this scenario. The Kansas City Motors data was extracted into a
4-2 Project One 4
spreadsheet and the data was stored in a AS400 database, in scientific format, whereas the
warehouse uses Microsoft SQL, which presents the incompatibility at play.
Data Analyzed
@ When analyzing both data sets, we must ensure the correct amount of information is retained
from both datasets. Some of the data may not be relevant or represent the issues that are
occurring. Storing all the data from each dataset seems to be the smart move due to there being
limited data provided to work with. Using the data from the word document to correct the errors
that were extracted and put into the Kansas City spreadsheet. Corrections should be based upon
the information from the word document to correct the cells that have errors. Not correcting
these cells leaves missing data or errors in the data, which contributes to inaccuracies.
Limitations of Data
Some companies have rules limiting data sharing across their departments and restrict the flow
of data. When collecting or preserving data there are no limits or restrictions on specific data.
The data supplied is the number of products sold monthly at each location. There would be rules
applied if the data held any customer information or payment history within these datasets. Due
to the dataset not having any personal or payment method associated with it, there are no rules
applied when sharing this data.
References
4-2 Project One 5
Muelaner, J. E. (2020, September 15). Lifewire: Tech for Humans. Retrieved from Syntax
Errors: What They Are and Why They're a Problem: https://www.lifewire.com/what-is-a-
syntax-error-4687347
