visualising proportions

Question 1

what is promotion data

Answer 1

Proportion data is grouped by categories, subcategories.

For example, total expenses (the whole) is aggregated by expenses of departments (categories or the parts)

Question 2

what are the 3 most important things for proportion data

Answer 2

maximum
minimum
overall distribution

Question 3

what is ranking analysis

Answer 3

sorting the expenses is called ranking analysis

Question 4

what is part-to-whole analysis s

Answer 4

Express expenses as percentage

Question 5

what are the 7 data proportions pattern

Answer 5

uniform (all values are roughly the same)

uniformly different ( differences from one value to the next decrease the same amount)

non uniformly different (differences from one value to the next vary significantly)

increasingly different (differences from one value to the next increase)

decreasingly different (differences from one value to the next decrease)

alternating differences (differences from one value to the next begin small, then increase, and then become small again)

exceptional (one or more values are extraordinarily different from the rest)

Question 6

what are some charts which are used for Proportions data display

Answer 6

–> Pie chart
–> Donut chart
–> Stacked area chart
–> Stacked bar chart
–> Treemap
–> Pareto chart

Question 7

what is a pie chart

Answer 7

The circle represents the whole, and the size of wedge represents the part or
percentage of that whole

Question 8

which chart is similar to pie chart

Answer 8

donut chart

with a hole cut out in the middle. The same arguments of angles and human perception still apply

Question 9

when to use the stacked area chart

Answer 9

Use the stacked area chart if you want to show changes over time for several variables

vertically it always adds up to 100%

Question 10

what is facetting

Answer 10

Technique used in data visualization to create multiple individual plots or charts that share the same axes, data scales, and structure but display different subsets of the data.

These smaller plots are arranged in a grid-like format to facilitate easy comparison between the subsets.

Question 11

when to use stacked bar chart

Answer 11

If you have only a few distinct points in time, you can use the stacked bar chart in the
same way you use the stacked area (just set the bars vertical).

Question 12

when to use a treemap

Answer 12

–> Treemap uses the areas of rectangles to show relative proportions. It works especially well if your data has a hierarchical structure with parent nodes, children, etc.

–> Display quantities for each category via area size

–> Colour/gradient can be a 2nd measure

Question 13

what is a Pareto chart

Answer 13

–> Besides ranking values, important to examine the cumulative contribution of parts to the whole

–> Can be constructed using bar chart and line chart togethe

Question 14

what is the Pareto principle

Answer 14

80% of the efforts come from 20% of the causes

Question 15

what are the techniques and best practices for visualising proportions

Answer 15

–> Grouping categorical items in an ad hoc manner
–> Using Pareto charts with percentile scales
–> Re-expressing values to solve quantitative scaling problems
–> Using line graphs to view ranking changes through time

Question 16

what is grouping

Answer 16

–> Segment data into meaningful grouping for analysis

–> Ability to group categorical items in an ad-hoc manner is critical to uncover interesting insights

Question 17

where are percentile scales used ?

Answer 17

Used in interval scale (size of orders from largest to smallest, grouped into percentile
intervals)

Question 18

what is percentage with example

Answer 18

absolute; an exact quantitative rating

(Your Marks/ Total Marks ) * 100

Question 19

what is percentile with example

Answer 19

comparative quantitative measurement. The 99th percentile is the top 1%, 98th percentile is the top 2%

(no. of pple behind you/total num of ppl) *100

Question 20

what problem does reexpression solve

Answer 20

When a set of ranked values extends a vast scale, it could be difficult to see and compare with the lowest values at the end graph

reexpression solves this scaling problem so that values are evenly distributed across the quantitative scale in a single graph

Question 21

what is wrong with using reexpression

Answer 21

It alters distances between values and distorts the actual magnitudes of
the difference so use it with care

Question 22

2 reexpression methods

Answer 22

1. Square root re-expression
2. Logarithmic re-expression

Question 23

We should never use line chart to display __________ but to
visualise the changes over time is fine

Answer 23

single ranking relationship

note: only show the changes in the ranking and not the actual values