By the third stage an informatics course usually turns to whether the data underneath your analysis can bear weight, and to who is accountable for keeping it that way. Completeness, accuracy, timeliness and consistency are the properties a dashboard silently assumes, and governance is the organizational machinery that makes those assumptions defensible: stewardship, definitions, access rules and a place where disputes about a number are settled. Your section may print this as NR 706 or NR706; it is the same course. Chamberlain publishes no syllabi outside Canvas. The placement here is our teaching judgment from the course's catalog arc; your section's rubric decides what your week actually asks.
What NR-706 Week 3 asks for
Two reports left the same organization in the same month with different falls numbers, and the meeting spent forty minutes arguing about which one was right. It turned out both were. One counted falls by the unit where the event occurred, the other by the unit that held the patient at midnight, and no written definition existed to adjudicate. That is a data quality problem that is really a governance problem, and it is the shape of the argument this stage wants you to be able to make: a number is not true or false on its own, it is true relative to a definition somebody has to own.
The written work here usually asks you to assess the quality of the data supporting the problem you named earlier, using recognized dimensions rather than a general sense of reliability. Each dimension has a failure mode you can name from your own setting. Completeness fails when a field is optional and busy. Accuracy fails when the easiest value to select is not the true one. Timeliness fails when capture happens at the end of a shift rather than at the moment of care. Consistency fails when two units interpret the same field differently, which is the falls example above.
At doctoral altitude you are also expected to reach the governance layer. Who is the steward of this element. Where is its definition written. What committee approves a change to a report. Who may see the data and under what authority. A paper that assesses quality without asking who is accountable has diagnosed a condition and declined to name the service responsible for treating it, and graders in a systems course read that omission as a gap in altitude rather than a stylistic choice.
Expect the deliverable to be an assessment with a structure, often a table of dimensions with findings and a written interpretation, and sometimes a set of recommendations. If a discussion runs alongside it, remember that specific claims about your organization's data quality are the kind of statement that should be phrased carefully and de-identified, because posts are visible to a section and do not reopen after submission in Canvas.
The NR-706 Week 3 method, step by step
Six moves for assessing data quality without either flattering or slandering your organization.
-
Pick the specific elements your argument depends on
Assess three or four data elements, not the record. Name each one exactly as it appears in the system, note its field type, and state which claim in your problem statement rests on it. Assessing everything produces a paper that assesses nothing.
-
Adopt a published set of quality dimensions and attribute it
Use a named framework of dimensions rather than inventing categories. Your reader can then check your reasoning against a standard, and the paper inherits vocabulary that is stable across the literature. Cite it with a year in the sentence.
-
Convert each dimension into a testable question
Completeness becomes what proportion of eligible encounters have this field populated. Timeliness becomes how many minutes or hours elapse between the event and the entry. A dimension you cannot phrase as a question you could answer will produce a paragraph of adjectives.
-
Run the smallest honest audit you can
Even thirty records reviewed against a written rule beats an impression. Record your sampling approach, your inclusion rule and your window, and report counts with denominators. If no audit is possible, design one on paper and say what it would cost to run.
-
Trace each defect back to the workflow that produced it
A missing field is a symptom. The cause is a design decision, a competing priority or an ambiguity in the definition. This is where the previous stage's workflow map earns its keep, and it is the step that turns an audit into analysis.
-
Name the governance structure, or its absence
Say who stewards the element, where the definition lives, which body approves report changes and how access is authorized. If those answers do not exist in your organization, that is a finding, and it should be written as one rather than left as a blank.
A layout that makes a data quality assessment defensible
Our frame for a data quality and governance assessment, sized for roughly 1,300 to 1,600 words. It is our own outline rather than anything the university issues, and your week's rubric outranks it wherever the two disagree.
| Section | What belongs in it | Word target |
|---|---|---|
| Elements under assessment | Three or four named elements with field types, and the claim each one supports in your larger argument. | 140 to 180 |
| Framework and definitions | The dimension set you adopted, attributed and dated, with each dimension defined in your own words. | 170 to 210 |
| Audit method | Sampling rule, inclusion criteria, window, who reviewed and against what written standard. | 150 to 190 |
| Findings by dimension | Counts with denominators for each element and dimension, presented so a reader can see the worst cell immediately. | 280 to 340 |
| Causes in the workflow | For each significant defect, the step and design decision that produced it, drawn from your process map. | 240 to 290 |
| Governance and recommendations | Stewardship, definition location, approval path, access rules, and what should change in each. | 220 to 270 |
Evidence craft for quality and governance writing
Define before you measure. Every quality claim needs its rule written first: eligible means these encounters, populated means this field contains a value from this list. Papers that report percentages without the rule behind them are unauditable, and a doctoral grader will say so.
Use the regulatory and standards literature at the right altitude. Privacy and security obligations, retention rules and reporting requirements shape governance and belong in the paper, but summarize their function rather than reciting statute. Cite the source, state what it requires of your organization in one sentence, then return to your data.
Distinguish a data defect from a care defect. An unpopulated field may mean the assessment was not done, or that it was done and recorded elsewhere. Writing which one you found, and how you established it, is the single most credibility-earning move available in this stage.
Report the good cells too. An assessment that finds every dimension failing reads as advocacy. Where an element is well captured, say so and say why, because the mechanism that makes one field reliable is usually the design you will recommend applying to the unreliable ones.
Keep the audit de-identified and aggregate. Report counts and proportions. Do not include record identifiers, dates of service that could locate a case, or details that would identify the staff member who made an entry. State that discipline in your method paragraph.
Five mistakes that cost points in this week's territory
- Dimensions used as adjectives. Saying the data is incomplete and untimely without a measured question behind each word produces four paragraphs of nothing.
- No denominator. Proportions without their base cannot be weighed, and in a quality assessment the base is half the finding.
- Skipping governance. Quality without stewardship is a description of weather. The accountable body, or its absence, is the doctoral half of the paper.
- Recommending a new report. Proposing a dashboard on top of a field nobody reliably populates is the classic informatics error this course exists to prevent.
- Assessing the whole record. Breadth here buys nothing. Three elements examined properly outscore twenty listed.
Before you submit
- Each assessed element is named as it appears in the system, with its field type
- The dimension framework is attributed with a year inside the sentence
- Every quality claim has a written rule and a denominator behind it
- Each defect is traced to a specific workflow step or design decision
- Stewardship, definition location and approval path are named or their absence is stated as a finding
- The audit description makes clear that no identifying detail appears in the paper
Writing a data quality assessment for NR-706?
Send the rubric and your audit notes out of Canvas. A premium original draft comes back in 24 to 48 hours with every dimension phrased as a testable question and every defect traced to a workflow step, and revisions run until the grade lands.