CBA Certified Data Analytics Professional

Statistical Foundations for Analysts20%

• Calculate and interpret measures of central tendency, dispersion and position, selecting the measure appropriate to the data's shape and scale.
• Distinguish between population and sample, parameter and statistic, and correlation and causation in described business scenarios.
• Interpret probability statements, distributions and expected values as they arise in everyday analytical questions.
• Evaluate the reliability of a sample-based conclusion using confidence intervals, margin of error and sample size reasoning.
• Select the appropriate basic significance test for a described comparison and interpret its result, including what a p-value does and does not say.
• Identify common statistical traps in analyses, including survivorship bias, Simpson's paradox, regression to the mean and misuse of averages.

Preparing and Structuring Data20%

• Identify data quality defects in a described or shown dataset, including duplicates, inconsistent categories, invalid values, and mixed data types.
• Select the appropriate cleaning action for a defect, weighing deletion, correction, imputation and flagging, and state its effect on later analysis.
• Distinguish between wide and long data layouts and select the reshaping needed for a stated analysis or chart.
• Interpret relationships between tables, including primary and foreign keys, one-to-many and many-to-many structures, and the effect of granularity on aggregation.
• Evaluate a described data collection or combination step for bias, leakage of meaning, or loss of information, including date, text and unit handling issues.
• Select documentation and reproducibility practices that let another analyst retrace a preparation workflow.

Analysis with Spreadsheets and SQL Concepts20%

• Select the correct spreadsheet function or function combination (lookup, conditional aggregate, logical, text, date) for a described analytical task.
• Interpret what a given formula or SQL query returns, including its filtering, grouping and calculation logic.
• Distinguish between SQL clauses and their spreadsheet equivalents, including filtering rows versus filtering groups and joining versus looking up.
• Calculate summary results that a described pivot table or GROUP BY query would produce from a small dataset.
• Evaluate an analytical approach for correctness and efficiency, including common formula and query errors such as wrong join type, absolute versus relative references, and aggregation before filtering.
• Select appropriate verification steps for a formula, query or AI-generated draft of either, before its output is used in a decision.

Visualisation and Dashboards15%

• Select the chart type that best answers a stated analytical question, given the data types and comparison involved.
• Identify design faults that distort or obscure data, including truncated axes, dual axes misuse, 3D effects, overplotting and rainbow category colouring.
• Evaluate a described dashboard against its audience and decision purpose, including layout hierarchy, filter design and refresh expectations.
• Interpret charts accurately, including log scales, stacked and 100 percent stacked bars, box plots and heat maps.
• Select accessibility and labelling practices that make a visual readable for colour-blind viewers and international audiences.

Communicating Insight10%

• Select the framing of a finding that is accurate, decision-relevant and appropriately caveated for a stated audience.
• Distinguish between observation, interpretation and recommendation in analytical statements.
• Evaluate a written analytical claim for overreach, including generalising beyond the data, hiding uncertainty and burying assumptions.
• Interpret stakeholder questions to identify the underlying decision and the analysis actually required.
• Select an appropriate structure for an analytical narrative, including headline-first summaries and the role of appendices and reproducibility notes.

Introduction to Predictive Methods15%

• Distinguish between descriptive, diagnostic, predictive and prescriptive questions, and between supervised and unsupervised approaches.
• Interpret the output of a simple linear regression, including slope, intercept, R-squared and residual patterns, in business terms.
• Select an appropriate baseline forecasting approach for a described time series, including naive, moving average and trend-with-seasonality reasoning.
• Calculate and interpret basic evaluation measures, including forecast error measures and classification accuracy, precision and recall from a small confusion matrix.
• Identify overfitting, data leakage and unrepresentative training data as causes of models that fail after deployment.
• Evaluate a described predictive claim or AI-generated model output for plausibility, fairness concerns and fitness for the stated decision.