Min. 1st Qu. Median Mean 3rd Qu. Max.
25.0 742.2 1526.0 3071.9 3554.2 60869.0
summary(customers$Milk)
Min. 1st Qu. Median Mean 3rd Qu. Max.
55 1533 3627 5796 7190 73498
bins <-cut(customers$Frozen, breaks =4, labels =c("Low", "Medium-Low", "Medium-High", "High"))customers$Frozen <-factor(bins)customers$Region <-factor(customers$Region)# Scatterplot using "region" as color aestheticggplot(customers, aes(x = Frozen, y = Milk, color = Region)) +geom_point() +labs(x ="Frozen Variable", y ="Milk Variable", title ="Scatter Plot of Milk vs. Frozen Variables by Region",subtitle ="Relationship between Milk and Frozen Variables with Regional Distribution") +theme(legend.position ="right") +labs(caption ="Data Source: customers dataset")
8.1: Data-Ink Ratio and Chartjunk
Refinement 1: Removing Chartjunk
Intent: Remove unnecessary clutter to focus attention on the data. Rationale: Eliminating non-essential elements enhances the clarity of the plot and makes it easier for viewers to interpret the data.
ggplot(customers, aes(x = Frozen, y = Milk, color = Region)) +geom_point() +labs(x ="Frozen Variable", y ="Milk Variable", title ="Scatter Plot of Milk vs. Frozen Variables by Region") +theme_minimal() +theme(legend.position ="right") +labs(caption ="Data Source: customers dataset")
Refinement 2: Increasing Data-Ink Ratio
Intent: Maximize the ratio of data to ink to improve the efficiency of conveying information. Rationale: By reducing the amount of non-data ink and emphasizing the data points, viewers can focus more on the relationships within the data.
ggplot(customers, aes(x = Frozen, y = Milk, color = Region)) +geom_point(alpha =0.6) +# Reduce point opacitylabs(x ="Frozen Variable", y ="Milk Variable", title ="Scatter Plot of Milk vs. Frozen Variables by Region",subtitle ="Relationship between Milk and Frozen Variables with Regional Distribution") +theme_classic() +theme(legend.position ="right") +labs(caption ="Data Source: customers dataset")
These refinements aim to enhance the clarity and interpretability of the scatter plot. By removing unnecessary elements and increasing the emphasis on data points, viewers can better discern the relationships between milk and frozen variables across different regions.
8.2: Data Density and Overplotting
Refinement 1: Adjusting Point Size
Intent: Control overplotting by adjusting the size of data points. Rationale: By scaling point size based on data density, we can mitigate overplotting issues and improve the visualization of dense regions.
ggplot(customers, aes(x = Frozen, y = Milk, color = Region)) +geom_point(alpha =0.6, size =1.5) +# Increase point sizelabs(x ="Frozen Variable", y ="Milk Variable", title ="Scatter Plot of Milk vs. Frozen Variables by Region") +theme_minimal() +theme(legend.position ="right") +labs(caption ="Data Source: customers dataset")
Refinement 2: Adding Jitter
Intent: Introduce random jitter to data points to reduce overplotting. Rationale: By adding a small amount of random variation to the data points, we can prevent them from completely overlapping, making it easier to identify patterns.
ggplot(customers, aes(x = Frozen, y = Milk, color = Region)) +geom_jitter(alpha =0.6, width =0.1, height =0.1) +# Add jitterlabs(x ="Frozen Variable", y ="Milk Variable", title ="Scatter Plot of Milk vs. Frozen Variables by Region") +theme_minimal() +theme(legend.position ="right") +labs(caption ="Data Source: customers dataset")
These refinements address issues related to data density and overplotting. By adjusting point size and adding jitter, the plots provide a clearer representation of the distribution of milk and frozen variables across different regions, helping viewers identify patterns and trends more effectively.
8.3 Refinement:
Intent: To enhance the visual appeal and clarity of the scatter plot by adjusting the point size and adding transparency to overlapping points. Rationale: By adjusting the point size and adding transparency, we can alleviate the issue of overplotting and provide a clearer visualization of individual data points.
# Scatterplot with adjusted point size and transparencyggplot(customers, aes(x = Frozen, y = Milk, color = Region)) +geom_point(size =2, alpha =0.6) +# Adjust point size and transparencylabs(x ="Frozen Variable", y ="Milk Variable", title ="Scatter Plot of Milk vs. Frozen Variables by Region",subtitle ="Relationship between Milk and Frozen Variables with Regional Distribution") +theme(legend.position ="right") +labs(caption ="Data Source: customers dataset")
This refined plot uses smaller points with added transparency, making it easier to distinguish between overlapping data points. It improves the clarity of the visualization, allowing viewers to identify patterns more effectively.
8.4 Refinement:
Intent: To enhance the plot’s readability, add a grid to aid in assessing the distribution of data points. Rationale: Adding a grid helps viewers assess the distribution of data points more accurately, improving the plot’s readability and facilitating data interpretation.
# Scatterplot with grid linesggplot(customers, aes(x = Frozen, y = Milk, color = Region)) +geom_point(size =3) +geom_smooth(method ="lm", se =TRUE) +labs(x ="Frozen Variable", y ="Milk Variable", title ="Scatter Plot of Milk vs. Frozen Variables by Region",subtitle ="Relationship between Milk and Frozen Variables with Regional Distribution") +theme(legend.position ="right", panel.grid.major =element_line(color ="gray", linetype ="dashed")) +# Add grid lineslabs(caption ="Data Source: customers dataset")
This refined plot incorporates grid lines, enhancing the readability of the visualization by providing reference points for assessing the distribution of data points. The grid lines aid viewers in understanding the relationship between Milk and Frozen variables within each region more effectively.
