CMCE10002 Chap.1 What Business Analytics Is: Three Kinds of Question
What Business Analytics Is: Three Kinds of Question
Three dimensions, three different evidence bars
Foundations of Business Analytics opens by splitting the field into three dimensions: descriptive analytics, which establishes what happened in the past; predictive analytics, which anticipates what happens next; and causal analytics, which uncovers the cause-and-effect relationships that explain why something happened.
The subject treats that split as a working decision rather than a taxonomy, because each of the three demands a different kind of evidence.
Why the split does real work
A descriptive claim is about rows that exist and can only be wrong through miscalculation. A predictive claim is about rows that do not exist yet, so it can only be judged on data it was not built from.
A causal claim is about a world that did not occur, and it can be wrong even when every calculation is correct.
Recognising the moment a sentence changes level is the judgement this chapter builds.
From a raw record to a decision
The subject describes a toolkit running from gathering, transforming and storing raw data through to crafting narratives that influence business decisions, and it places the ethics of working with sensitive information inside that chain rather than beside it.
Where each question type breaks down
Descriptive work fails quietly when the record is incomplete: an average over the rows that survived a system migration describes the survivors and nothing else.
Predictive work fails when the past stops resembling the future, which is why a rule that scored well last year is not evidence about this one. Causal work fails when the group that received the change was never comparable to the group that did not. Naming the failure mode that belongs to your question type is the fastest way to find the weak point in someone else's analysis, and the fastest way to protect your own.
What this chapter covers
- 01
Descriptive, predictive and causal questions, and how to tell them apart
- 02
Reading the verb in a request to identify the question type
- 03
The evidence each question type demands before it can be answered
- 04
The four layers from raw record to a narrative a manager can act on
- 05
Bias, individual rights and transparency as constraints on the method
Turning a manager's sentence into an analytical plan
- 1Name the question type from the decision being proposed.
- 1State the evidence that question type requires.
- 1Say what the existing records do support.
- 1Name the design that would answer the question as asked.
Key terms
- Descriptive analytics
- Summarising what already happened in a record, using only rows that exist.
- Predictive analytics
- Anticipating a value that has not been observed yet, judged on data the rule never saw.
- Causal analytics
- Establishing whether one thing produced another, which requires a comparison with what would otherwise have happened.
- Counterfactual
- The outcome that would have occurred without the intervention, which is never present in the data.
- Grain
- What a single row of a table represents, such as one customer, one order or one firm-year.
What Business Analytics Is: Three Kinds of Question FAQ
What are the three kinds of analytics question?
The subject names descriptive analytics for what happened in the past, predictive analytics for what happens next, and causal analytics for why something happened. Each demands different evidence, so naming the type first decides what a good answer must contain.
How do you tell a causal question from a descriptive one?
Look at what the analysis will be used for. When the result is meant to argue for changing a procedure, policy or practice, the question is causal. Verbs such as improve, increase and lead to carry the same signal even inside a sentence that sounds purely observational.
Does an introductory analytics subject assume programming experience?
No prior programming experience is expected. The subject states that basic numeracy, curiosity about how data changes the way businesses operate, and a willingness to work through structured problems are what position a student to succeed.
Can the same dataset answer more than one kind of question?
Often, but not to the same standard. A table of transactions supports strong descriptive statements, weaker predictive ones once you hold rows back to score against, and causal statements only when something in how the data arose created a fair comparison. The dataset does not change; what changes is how much of it the claim is entitled to lean on.
Exam move
Practise sorting real requests into the three types before touching any code. For each one, write the evidence that type requires and then say whether the dataset in front of you can supply it. Most marks lost in short answers come from answering a descriptive question when a causal one was asked.
Working through What Business Analytics Is: Three Kinds of Question in CMCE10002? Sia is AskSia’s AI Statistics tutor — ask any CMCE10002 What Business Analytics Is: Three Kinds of Question question and get a clear, step-by-step explanation grounded in how CMCE10002 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
