Data Analytics Foundations of Data Analytics Syllabus

Name two common techniques used in predictive analytics.

Regression analysis, machine learning classification, time-series forecasting, and decision trees.

What does predictive analytics produce, and does it guarantee outcomes?

It produces probabilities or forecasts of future events; it estimates likelihood and does not guarantee that outcomes will occur.

What is prescriptive analytics?

The most advanced analytics type, which recommends specific actions to take by combining predictions with optimization to achieve a desired outcome.

Name two techniques associated with prescriptive analytics.

Optimization, simulation, and recommendation engines (often powered by machine learning or AI).

How does prescriptive analytics build on predictive analytics?

Predictive analytics forecasts what will happen; prescriptive analytics goes further by recommending what actions should be taken to influence or optimize that outcome.

What are the six phases of the data analysis process (Google framework)?

Ask, Prepare, Process, Analyze, Share, and Act.

In the data analysis process, what happens in the 'Ask' phase?

You define the problem, understand stakeholder expectations, and frame clear, answerable questions that will guide the analysis.

What is a SMART question in the context of defining a problem?

A question that is Specific, Measurable, Action-oriented, Relevant, and Time-bound, ensuring the analysis targets a clear, answerable goal.

Why is properly defining the problem (the 'Ask' phase) critical to an analysis?

A poorly defined problem leads to analyzing the wrong data or answering the wrong question, wasting effort and producing insights that do not address the real business need.

In the data analysis process, what happens in the 'Prepare' phase?

You identify, collect, and store the data needed, deciding what data to use, where it comes from, and how it will be organized.

What are 'metadata' and why do they matter in the Prepare phase?

Metadata is data about data (e.g., source, format, creation date); it helps analysts understand, organize, and trust the data they collect.

In the data analysis process, what happens in the 'Process' phase?

You clean the data, checking for and fixing errors, removing duplicates, handling missing values, and ensuring data quality and integrity.

Name three common data-cleaning tasks.

Removing duplicates, handling missing values, correcting inconsistent formatting, fixing typos/errors, and removing outliers or irrelevant data.

What is 'data integrity' and why is it emphasized during cleaning?

Data integrity is the accuracy, completeness, and consistency of data throughout its lifecycle; maintaining it ensures analysis results are trustworthy.

In the data analysis process, what happens in the 'Analyze' phase?

You organize, transform, and examine the cleaned data to find patterns, relationships, and trends that answer the defined question.

Name two activities performed during the 'Analyze' phase.

Sorting and filtering data, performing calculations and aggregations, identifying trends/correlations, and creating pivot tables.

In the data analysis process, what happens in the 'Share' phase?

You communicate insights to stakeholders through visualizations, dashboards, and reports so the findings are clear and actionable.

Why is data visualization important in the 'Share' phase?

Visualizations make complex data easier to understand quickly, highlight key insights, and help stakeholders grasp patterns they might miss in raw numbers.

In the data analysis process, what happens in the 'Act' phase?

Stakeholders use the insights to make decisions and take action, applying the analysis to solve the original business problem.

What are the three main categories of data by structure?

Structured, semi-structured, and unstructured data.

What is structured data? Give an example.

Data organized in a predefined format such as rows and columns (a fixed schema), easily searchable; e.g., a relational database table or spreadsheet.

What is unstructured data? Give an example.

Data with no predefined format or schema; e.g., text documents, emails, images, audio, and video.

What is semi-structured data? Give an example.

Data that does not fit a rigid table but contains tags or markers to separate elements (partial structure); e.g., JSON, XML, or emails with headers.

What is the difference between qualitative and quantitative data?

Qualitative data is descriptive and non-numeric (categories, qualities, opinions), while quantitative data is numeric and measurable (counts and amounts).

Give one example each of qualitative and quantitative data.

Qualitative: eye color, customer feedback text, product category. Quantitative: height in cm, number of sales, temperature.

Within quantitative data, what is the difference between discrete and continuous data?

Discrete data can only take specific, countable values (e.g., number of cars), while continuous data can take any value within a range (e.g., weight or time).

What is the difference between primary and secondary data collection?

Primary data is collected firsthand by the analyst for a specific purpose (e.g., surveys, experiments); secondary data is gathered by someone else and reused (e.g., existing reports, public datasets).

Name three common data collection methods.

Surveys/questionnaires, interviews, observation, experiments, and extracting data from existing databases or sensors.

Define first-party, second-party, and third-party data.

First-party data is collected directly by your organization from its own customers; second-party data is another organization's first-party data obtained directly from them; third-party data is aggregated from many sources and sold by an outside provider that has no direct relationship with the individuals.

Why is first-party data generally considered the most reliable?

Because you collect it directly from your own audience, so you know its source, accuracy, and how it was gathered, and it raises fewer privacy concerns than purchased third-party data.