Building a Data Portfolio

Select a question and an audience

Select a question and an audience

A portfolio artifact should answer a defined question for a reader who can inspect how the answer was obtained. Choose a manageable question with permitted data, explicit definitions and a feasible validation method. Reuse an existing course capstone rather than creating several near-identical projects. A focused extension, such as subgroup coverage or a changed-input test, can demonstrate independent judgment. State why the question matters without inventing a real client, employer or commercial outcome. An attractive presentation cannot replace a correct method.

Follow a concrete example

Choose one reader and one question, then list the artifacts that let that reader check your answer. A region extension to an existing capstone is often more informative than another copied end-to-end tutorial. Define which decision or validation is yours, and what was supplied. Scope the work to the data and time you can actually inspect.

Further reading: DataCareerHub learning quality. The case and exercises are authored for DataCareerHub.

Your task

Choose an existing project and write its audience, question, source permissions, scope and independent extension.

Use learner_worksheet.txt and the provided source/code files. Save your own artifact and compare the separate solution after attempting the task.

Organise reproducibility and evidence

Organise reproducibility and evidence

A reader needs source/provenance, dictionary, method, executable or editable work, checks and findings. Use relative paths where practical and document environment/commands. Keep raw and derived artifacts distinct; remove secrets and confidential records before sharing. A public repository is optional, not required. If you publish, inspect the actual files and rights first rather than assuming every downloaded dataset may be redistributed. The course’s fictional fixtures can be used for these exercises, with clear attribution to supplied learning material.

Follow a concrete example

Trace a headline through the folder: source files→contract→query/script→validation→result→finding. Each link should work without a live account or a remembered notebook state. If publishing is later chosen, review actual file contents for secrets, permissions and personal data first. A public repository is optional; reproducible local evidence is already useful.

Inspect the reference implementation

Language: Python 3.12. The complete downloadable file includes the named helpers and the fresh-output command-line runner. Trace the source, transformation and result against this lesson’s contract before running it locally.

def career_evidence(source,out):
 base=py_project(source,out)
 return {'project_claim':'Implemented an analysis of fictional Haven Workshop orders with reconciled line and order grain.','evidence':base,'avoid_claims':['client delivery','employment outcome','accreditation','causal sales improvement'],'independent_extension':'Compare region-level order counts and value, retaining cancelled-order exclusions.'}

Further reading: DataCareerHub learning quality. The case and exercises are authored for DataCareerHub.

Your task

Create a portfolio folder and a reproduction README. Link each claim to its source calculation and test artifact.

Use learner_worksheet.txt and the provided source/code files. Save your own artifact and compare the separate solution after attempting the task.

Present contribution and limitations ethically

Present contribution and limitations ethically

Describe what you built, what was supplied, which decisions you made and what remains uncertain. A fictional exercise can demonstrate skill without being represented as paid client work. Do not invent business impact, certification, employer endorsement or placement outcomes. Use specific observable results and appropriate scope: “reconciled a fictional 24-order dataset” is defensible; “increased real revenue” is unsupported. Reflect on a failure or changed assumption and how it affected your conclusion. This helps a reviewer assess judgment rather than a polished claim alone.

Follow a concrete example

Rewrite “improved revenue” as a factual artifact statement unless you actually evaluated a real intervention. For example, say you reconciled a fictional24-order source and added an independently tested subgroup comparison. Credit reference code and describe limitations. This specificity gives a reviewer something to assess without inventing clients, jobs or certification.

Further reading: DataCareerHub learning quality. The case and exercises are authored for DataCareerHub.

Your task

Write a 120-word description and an attribution/limitations note. Mark every statement requiring evidence.

Use learner_worksheet.txt and the provided source/code files. Save your own artifact and compare the separate solution after attempting the task.