R Markdown

This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.

When you click the Knit button a document will be generated that includes both content as well as the output of any embedded R code chunks within the document. You can embed an R code chunk like this:

Online_Retail <- read.csv('C:/Users/laasy/Documents/Fall 2023/Intro to Statistics in R/Datasets for Final Project/OnlineRetail.csv')

Question 1 :

A list of at least 3 columns (or values) in your data which are unclear until you read the documentation. E.g., this could be a column name, or just some value inside a cell of your data Why do you think they chose to encode the data the way they did? What could have happened if you didn’t read the documentation?

Answer:

In an online retail dataset, there are some columns that are unclear until I read the documentation. Here are the following coloumns:

  1. StockCode: The “StockCode” column typically contains alphanumeric codes that represent the specific products sold by the retailer. Without reading the documentation, it would be unclear what each code represents. These codes could be internal identifiers for products, and understanding their meaning is crucial for proper analysis. These codes are often consistent across the retailer’s database, making it easier to manage and analyze large volumes of products.If not read the documention, I could misinterpret the meaning of StockCodes and make incorrect assumptions about the products being sold.

  2. Description: The “Description” column provides a textual description of the products sold. This column may contain product names, brief descriptions, or even special notes. Without the documentation, might not fully understand the nature of the products, which could lead to misinterpretation or incomplete analysis. Providing a textual description allows for better human interpretation and is useful for reference. Without knowing the context provided by the Description column, might miss valuable insights related to specific products or categories.

  3. InvoiceNo: The documentation may not explicitly mention that transactions starting with the letter ‘c’ are cancellations. The prefix ‘c’ is not a common convention for cancellations in all values of the dataset, so understanding its significance required reference to the documentation.Sometimes we might incorrectly treat cancellation transactions as regular transactions, leading to errors in the analyses, such as overestimating sales or revenue. The presence of cancellation transactions might lead to data quality issues if they are not appropriately identified and handled. Failing to recognize the ‘c’ prefix could result in a lack of understanding about the retailer’s transaction handling processes, potentially leading to incorrect business decisions.

Question 2:

At least one element or your data that is unclear even after reading the documentation You may need to do some digging, but is there anything about the data that your documentation does not explain?

Answer:

Certainly, the “Quantity” and “InvoiceDate” columns in an online retail dataset are unclear even after reading the documentation.

  1. Quantity:

  • Unit of Measurement: The documentation may not specify the unit of measurement used for the “Quantity” column. For example, it may not explicitly state whether “Quantity” represents individual items, packs, kilograms, or other units. This lack of clarity can affect how you interpret and analyze the data, especially when dealing with products of different sizes or types.

  • Negative Values: In some datasets, the “Quantity” column might contain negative values, indicating returns or cancellations. The documentation might not explain the conventions used for negative quantities, making it unclear whether these values should be included in analyses like calculating total sales.

  1. InvoiceDate:

  • Time Zone: The “InvoiceDate” column usually records the date and time of each transaction, but it may not specify the timezone used. Without timezone information, it’s unclear whether the timestamps are in UTC, the retailer’s local time, or some other timezone. This can lead to issues when conducting time-based analyses or comparing data across regions.

  • Data Resolution: The documentation might not clarify the data resolution of the “InvoiceDate” column. For example, it may not specify whether the timestamps include seconds, milliseconds, or just date and hour. This can affect the precision of time-based analyses and might lead to inaccuracies if assumptions are made.

Question 3:

Build a visualization which uses a column of data that is affected by the issue you brought up in bullet #2, above. In this visualization, find a way to highlight the issue, and explain what is unclear and why it might be unclear.

Answer:

Visualization:

data <- data.frame(Online_Retail)

# Load necessary libraries
library(ggplot2)
library(dplyr)
## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
# Calculate total quantity per country
data_summary <- data %>%
  group_by(Country) %>%
  summarise(TotalQuantity = mean(Quantity))

# Create a bar plot
ggplot(data_summary, aes(x = Country, y = TotalQuantity, fill = Country)) +
  geom_bar(stat = "identity") +
  labs(
    x = "Country",
    y = "Average Quantity",
    title = "Average Quantity by Country"
  ) +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Explanation:

In this bar plot, I have aggregated the data to calculate the total quantity ordered for the 36 countries with the average quantities. The x-axis represents the countries, and the y-axis represents the average quantity. Each country is represented by a different color.

Unclear Elements and Reasons:

The issue here is that it’s unclear why some countries have negative quantities. Without additional context or documentation, it’s challenging to determine the reasons behind these negative quantities and whether they should be included in analyses or treated differently. The negative values are due to cancellation of orders from the customers.

Significant Risks:

The significant risk in this scenario is that without understanding the reasons for negative quantities in certain countries, there might be incorrect assumptions about sales trends, inventory management, or customer behavior. Risks include:

  • Misinterpretation: Treating negative quantities as regular sales can lead to inaccurate country-level sales figures and affect financial reporting.

  • Inaccurate Insights: Not addressing this issue might lead to incorrect insights about the sales performance of different countries and product popularity.

Actions to reduce negative consequences:

To reduce negative consequences, considering the following actions would be better:

  • Data Preprocessing: Depending on the documentation, there might need to preprocess the data by excluding or adjusting negative quantities by country to align with business rules.

  • Flagging Negative Quantities: Creating a separate column or flag to identify negative quantities explicitly by country and handle them differently in the analysis.

  • Sensitivity Analysis: Perform sensitivity analysis to assess how different treatments of negative quantities by country impact the results.

  • Documentation: If the documentation is missing or unclear, documenting the own assumptions and data processing steps to ensure transparency and reproducibility.

By addressing these risks through proper documentation and data preprocessing, there are chances of minimizing the negative consequences and ensure more accurate and reliable analysis of average quantities ordered by country.