Creating a Multi-Variates Visuals Using the 5 Basic Principles of Design
1) This week we go over the 5 basic principles of design. Then we will create a scatter plot using multivariate visualization in mind, with the help of ggplot2 and dplyr in RStudio. Dplyr just helps to organize the data set we are working with. Speaking of, the name of which is called consumer goods. I downloaded it for a previous assignment, and I am now reusing for a completely different purpose. With all of that laid out for us lets get to work.
2) The five basic principles of design are: Alignment, Repetition, Contrast, Proximity, and Balance. Alignment helps to create order, repetition strengthens design by tying things together, contrast allows emphasis on certain parts, Proximity creates organization and balance provides stability. Utilizing all these basic design principles allows you to communicate your work more effectively to others.
3) Now using both ggplot2 and dplyr, we will create a multi-variate visual in the form of a bubble scatter plot. Comparing the price of items, to their distribution to their market share. You can find ConsumerGood.csv here.
> #Loading the 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
Warning message:
package ‘dplyr’ was built under R version 4.3.3
>
> #Read the dataset
> data <- read.csv("CENSORD/ConsumerGood.csv")
>
>
> #Creating the multivariate scatter plot
> ggplot(data, aes(x = distribution, y = price)) +
+ geom_point(aes(color = share, size = share), alpha = 0.7) +
+ scale_color_gradient(low = "blue", high = "red") +
+ labs(title = "Price vs. Distribution of Consumer Goods",
+ x = "Distribution",
+ y = "Price",
+ color = "Market Share",
+ size = "Market Share") +
+ theme_minimal() +
+ theme(legend.position = "right")
| |
|
4) The bubble scatter plot shown above shows the distribution and price for consumer goods, with each point's color and size representing market share. Higher values of market share appear larger and transition from blue to red, allowing for quick identification of how distribution and price vary across different market share levels, revealing potential patterns in consumer pricing dynamics. I was not able to combine the color and the bubble size together on the legend, due to limitations in ggplot2 and dplyr, or my own lack of knowledge.
5) In this particular scenario, multivariate visualization is indeed effective here. It reveals the relationships between price, distribution, and market share. The color and size distinctions provide quick recognitions of patterns, such as how market share correlates with price. However, adding too many variables could risk clutter, potentially hindering interpretation and overall clarity.
Comments
Post a Comment