Data Literacy Competency Assessment

Data Literacy Competency Assessment

Purpose

Data literacy doesn't mean everyone needs to know how to write SQL queries. It means the ability to understand, evaluate, and apply data in decision-making. When an AI or data initiative fails for organizational reasons (not technical ones), the cause is often a competency gap in one of four areas — not the tool. This skill produces a structured assessment of where the gap is, and at what organizational level, so development effort is aimed correctly instead of "training everyone on everything."

Anchored in research

• A DALI-type data literacy framework (a synthesis of multiple sources, from citizen- and professional-level data literacy definitions): four core elements — Understanding data, Acting on data, Engaging and influencing through data, and Ethics & privacy as a cross-cutting theme.
• Ackoff, Russell L. — the DIKW hierarchy (Data, Information, Knowledge, Wisdom, 1989) underlying the maturity ladder: data itself isn't knowledge, knowledge isn't understanding, understanding isn't the wisdom to act correctly.
• DAMA International — DAMA-DMBOK (Data Management Body of Knowledge) as the professional standard defining data governance competency (a professional-body standard in the same vein as BABOK/PMI/SFIA elsewhere in this repo).

Method

1. Assess the four areas separately — don't produce a single overall score, because an organization is typically uneven across areas:
   • A. Understanding data: is it understood what data is, where it comes from, who owns it, and that "data is a representation of reality, not reality itself" (see ../data-bias-and-quality-critical-reading/SKILL.md)?
   • B. Acting on data: can data quality be assessed, can misleading reporting be spotted, and does data actually drive decisions and behavior change — or is it "nice-to-know" metrics being collected that lead nowhere?
   • C. Engaging through data: can data be synthesized, visualized, and told as a story that gets a decision-maker to act (see ../data-storytelling-and-business-translation/SKILL.md)?
   • D. Ethics & privacy: are the ethical and legal boundaries of collecting and using data understood (especially in AI models) — this is a cross-cutting theme across the other three, not a separate stage.
2. Use a four-step maturity ladder for each area:
   • Level 1 — Unaware: data is used without questioning its origin or limitations.
   • Level 2 — Aware: limitations are recognized, but not systematically factored into decisions.
   • Level 3 — Applying: limitations are systematically factored in, data repeatedly leads to correct decisions.
   • Level 4 — Embedded: data literacy is part of the organization's default way of operating, not a separate skill that has to be consciously invoked. Score each area (A-D) at level 1-4 separately — an organization can be at level 3 in Understanding but level 1 in Engaging.
3. Differentiate the assessment by role — the same data literacy requirement doesn't apply to everyone:
   • Leadership/decision-makers need C (Engaging — able to demand and interpret data as a story) and D (Ethics — accountable for decisions) above all.
   • Analysts/data professionals need A (Understanding) and B (Acting) in depth above all.
   • Line managers/end users need a sufficient level of A (able to question) and D (able to spot ethical risks in their own work) — not necessarily deep B/C competency. If the whole organization is trained on the same program regardless of role, that's the most common way data literacy investment gets wasted.
4. Identify the biggest bottleneck; don't try to fix everything at once. The data literacy chain is only as strong as its weakest link: if leadership is at level 1 in Engaging (C), the best analysis (A/B at level 4) never leads to a decision, because it can't be interpreted or trusted. Prioritize development effort based on where the weakest link is by area/role, not where it's easiest to train.
5. Produce the assessment as a table: role × area (A-D) × level (1-4) × the biggest observed risk in that cell. This table is the skill's primary output, not a long narrative description.

What this skill does NOT do

• Doesn't design the training program or its content itself — produces a diagnosis on which the training or other development effort is designed separately.
• Doesn't assess individual people's competency by name — assesses roles and organizational levels, not individuals.
• Doesn't replace a technical assessment of data architecture or infrastructure — assesses people's ability to use and interpret data, not the technical condition of systems.
• Doesn't confirm figures or maturity levels from memory — bases the assessment on observations you provide (interviews, surveys, observation) or marks an assumption clearly ([assumption — verify]).

Refinement notes

Areas to keep deepening with real practice:

• a concrete interview/survey template for assessing the level of each area (A-D) (into ../../references/)
• your own observations about which role/area combination is most often the weakest link across different industries
• examples of how imbalanced data literacy (e.g. strong analytics, weak leadership engagement) has blocked a project from moving forward

Once this section is filled in and validated in practice, update the maturity field in skills_index.json to draft, validated, or canonical (see ../../../meta/maturity_levels.md). Don't add new fields to the frontmatter — name and description are the only ones allowed (see ../../../meta/frontmatter_schema.md).

Continue from here

• Next in this pack (if area C is weak): ../data-storytelling-and-business-translation/SKILL.md
• Next in this pack (if area A is weak): ../data-bias-and-quality-critical-reading/SKILL.md
• Related skill in another pack: ../../../change-and-communication/skills/workshop-and-facilitation-design/SKILL.md — if a training/facilitation session is being designed based on the diagnosis.
• A ready-made skill chain for this situation: see ../../../playbooks/
• This pack's shared guardrails: ../../CLAUDE.md

References

• ../../references/ — the pack's shared background material
• ../../CLAUDE.md — the pack's shared guardrails