DataX study plan

Map every DataX domain to its exam weight and rank your personal weak spots so the remaining seven weeks target real gaps

Define data governance, stewardship, ownership, and custodianship in your own words and write a one-page glossary you will reuse all eight weeks

Diagram a source-to-target ETL and ELT pipeline end to end, labeling extraction, transformation, load, and orchestration stages

Compare full-load versus incremental-load strategies and list three criteria that decide which one a pipeline should use

Build a metadata inventory for a fictional retail dataset covering technical, business, and operational metadata categories

Draft a data lineage record that traces one metric from source table through transformation to final report

Write a data retention and archival policy for that dataset that names owners, review cadence, and disposal triggers

Explain idempotency and reprocessing in batch pipelines and describe how to make one load job safely rerunnable

Compare normalized third-normal-form modeling against star and snowflake schemas and state when each is preferred

Design a star schema with one fact table and at least three dimension tables, including grain, keys, and slowly changing dimension handling

Classify platform components (storage, compute, orchestration, catalog, serving layer) and map each to a concrete example tool

Distinguish OLTP from OLAP workloads and justify a storage choice for each with two technical reasons

Write five analytical questions answerable from your star schema and identify the exact columns each query needs

Compute mean, median, standard deviation, and interquartile range by hand for a 12-row sample and verify with a tool

Interpret a correlation coefficient correctly and write two sentences explaining why correlation does not establish causation

Define a sampling method for a validation dataset and justify the choice against bias and cost

Enumerate at least eight data quality dimensions and write a testable rule for each on a real dataset

Distinguish completeness, accuracy, consistency, uniqueness, timeliness, and validity with one concrete failure example per dimension

Design a data cleaning sequence for duplicates, missing values, outliers, and format inconsistencies, and justify the order

Specify thresholds and alert routing for automated quality checks so a failing rule triggers a defined response

Apply your quality rulebook to a messy sample dataset and document every cleaning decision with its rationale

Classify data as public, internal, confidential, or restricted and assign handling requirements to each tier

Distinguish anonymization, pseudonymization, masking, and tokenization and state one appropriate use for each

List the core obligations of major privacy regimes (GDPR-style consent and erasure, HIPAA-style safeguards, PCI DSS-style cardholder protection) without overstating specifics

Select the correct chart type for five different analytical questions and justify each choice by data shape and audience

Redesign one weak chart by fixing axis, scale, labeling, and color issues, and note each change

Design a dashboard layout with a stated audience, three key metrics, and a clear visual hierarchy

Apply least-privilege access to that dashboard by defining row-level and column-level restrictions per role

Synthesize all eight domains into a single integrated case study from raw source data to governed, secured, visualized output

Complete two timed full-length practice sets and score each domain separately to expose remaining weak areas

Re-drill your three weakest domains using your own glossary and rulebook rather than new material

Write a one-page exam-day strategy covering time allocation, question triage, and flagging rules