WEEK 2 - Breaking Down Problems and Gaining Insights

Question 1

What is a logic tree:

Answer 1

A visual and analytical decision support tool.

Why use it:
1. Allows you to visualise your problem - breaking it down into different visual components
2. Allows you to remove irrelevant information
3. Oftens leads to a clear hypothesis

NOTE: Logic trees can be created from multiple frames i.e progress flow and components

Question 2

Creating logic trees

Answer 2

Good practice to use a logic tree at every stage of the The McKinsey 7-Step Problem-Solving Framework, and after the logic tree to make a priotisation matrix.

Ensure mutliple branches for depth
- Branch 1 should be for breadth
- 2nd and 3rd etc should be for depth

RULES FOR LOGIC TREE:
Branches should be MECS (mutually exclusive, collectively exhaustive)
- ME = Branches of the tree don’t overlap
- CE = The tree needs to contain all relevant elements of the core problem

NOTE: Trialing is useful to identity the best fit solution to the problem

Question 3

Why use statistics?

Answer 3

Statistics is a detective’s toolkit:

Gaining Insights:
Statistics are used to understand patterns, trends and underlying relationships

Theory Validation:
Statistics gives us a systematic approvide to validate theories (supporting or rejecting initial beliefs)

Predicting Power:
Allows us to see the bigger picture, making prediction with confidence.

Question 4

Population vs sample:

Answer 4

Population: The entire group that fits your criteria (People containing elements of anything that you want to know)

Sample: A subset of a population -> taking data from the target population which allows you to draw general conclusions.

NOTE: Samples are often used due to the impractical, costly, time-consuming, inconvenient and unmanageable of collecting from the entire population.

Question 5

Types of data: Quantitative and Qualitative

Answer 5

Quantitative: Data that can be measured.
- Discrete: Whole numbers, can’t be broken down.
- Continuous: Numbers that can be broken down.

Qualitative: Non-numerical data that is categorical.
- Nominal: Data used for naming variables such as hair colour.
- Ordinal: Data used to order. variables e.g 1 = happy, 2 = neutral.

NOTE: Using tables and visual representation with qualitative sense of data trends, its essential to further summarise and quantify these insights for precise decision-making.

Question 6

Cross-sectional Data:

Answer 6

Observation or measurements taken on one or more variables at a single point of time

E.G A survey assessing customer satisfaction

Question 7

Time series data:

Answer 7

Observation or measurements of a single variable captured at different points in time.

E.G Monthly sales data from Jan to Dec 2023

Question 8

Panel Data:

Answer 8

Observations on multiple subjects (like products, stores) over multiple points in time.

E.G Tracking monthly sales and customer feedback scores across multiple stores over three years.