NURS 8740 Week 2 Data Cleaning Memo Example

Course	NURS 8740 Doctoral Seminar IV
Week	Week 2
Paper type	Data cleaning memo
Length	About 900 words, 5 pages
Format	Memo, APA 7 citations
School	Walden University
Program	Doctor of Nursing Practice (DNP)
Updated	October 2026

Free sample paper for NURS 8740 Week 2

Memorandum

To: DNP project chair; ICU nurse manager

From: DNP student, project lead

Date: Month Day, Year

Re: Cleaning of the ICU mobility extract before analysis: 352 rows pulled, 276 qualifying ventilator days analyzed, every removal traced to a written rule

Provenance

This memo records what was removed from the mobilization project's data file, why and on whose authority, before any summary was calculated. The file is the saved flowsheet report ICU Mobility Weekly, run from the hospital's reporting tool for the dates January 5 to March 28, which covers the four baseline weeks and the first eight weeks of the protocol. I ran the report on April 2, exported it to the project spreadsheet on the unit's secured drive and saved a read-only copy the same day, labeled as the frozen file. Every step below was applied to a working copy, so the frozen file can be used to repeat the cleaning from the start.

Each row in the export is one ventilated patient-day: a study number and date, the hours of ventilation already completed at midnight, the day's peak level on the ICU Mobility Scale (Hodgson et al., 2014), the screen result and its recorded time. The export holds 352 rows.

What this page is doingProvenance comes first because a reader cannot judge any cleaning rule without knowing which file it was applied to and whether the original still exists.

Condition of the File

The file arrived with four kinds of defect. Six rows were exact duplicates, all on days when a patient moved between two ICU beds and the flowsheet opened a second row for the new bed. Fifty-one rows were ventilated days that did not meet the 24-hour threshold at 0000 set by the operational definition. Nineteen rows fell on days with one of the four exclusions recorded at 0700. Fourteen rows had no mobility level entered at all. Two rows carried a level of 11 on a patient whose next row showed reintubation, which looked implausible at first and is discussed below.

Data cleaning is usually described as an activity, but it is better treated as a sequence of screening, diagnosis and treatment, each documented, so that another analyst can see what was done and why (Van den Broeck et al., 2005). The rules in the next section follow that order.

What this page is doingListing every defect before any rule shows the reader the full condition of the file, including the unflattering parts, rather than only the defects that were easy to fix.

Rules Applied

Rule 1: Duplicates. Where two rows share a study number and a date, keep the row with the higher mobility level and delete the other. The six pairs had identical levels and screen results in every case, so the choice did not change any value.

Rule 2: Qualifying days. Keep only rows where hours ventilated at 0000 are 24 or more. This applies the operational definition written for the project and removes rows from the first day of ventilation.

Rule 3: Exclusions. For each remaining row, look in the 0700 order record for any of the four excluding conditions named in the definition: comfort care, an open abdomen, a spinal injury not yet stabilized, or bed rest ordered by a physician. Mark each exclusion with its reason in the Excluded column and remove the row from the analysis file. The nineteen exclusions were nine bed rest orders, seven comfort-care orders and three open abdomens.

Rule 4: Blank levels. Keep rows with no level recorded in the denominator and count them as not reaching level 3. Do not fill them from therapy notes or nursing narrative. This follows the definition and avoids reintroducing the reviewer judgment that the prospective measure was designed to remove.

Rule 5: Implausible values. Flag any level of 8 or above on a day when the patient was still ventilated at 1800, and check the flowsheet. The two level 11 entries were real: both patients walked with the respiratory therapist in the morning and were reintubated in the evening for unrelated reasons. The entries stand. No value was changed in the file.

What this page is doingEach rule is written as an instruction another analyst could follow, and the memo reports what happened when it was applied, including when a suspicious value turned out to be correct.

Record Disposition

Table 1 traces the records from the count pulled to the count analyzed.

Table 1

Disposition of Rows in the Mobility Extract, January 5 to March 28

Step	Rows removed	Rows remaining
Pulled from the saved report		352
Rule 1: exact duplicates from bed moves	6	346
Rule 2: under 24 hours ventilated at 0000	51	295
Rule 3: exclusions at 0700 (bed rest 9, comfort care 7, open abdomen 3)	19	276
Analyzed: qualifying ventilator days		276
Of which baseline, January 5 to February 1		90
Of which protocol weeks 1 to 8, February 2 to March 28		186

The arithmetic reconciles at every step: 352 minus 6, 51 and 19 leaves 276, and 90 plus 186 is 276. The fourteen blank-level days remain inside the 276, four in the baseline and ten in the protocol weeks.

What this page is doingThe disposition table lets a reader check the arithmetic in seconds, which is why reviewers look at it before reading any finding.

Judgment Calls and Their Direction

Two decisions involved judgment and could affect the results. Keeping blank levels in the denominator lowers the measure, and more blanks fell in the protocol weeks than in the baseline, so this choice works against the project, not for it. Removing exclusions at 0700 rather than at any time in the day could leave in a patient whose bed rest order was written at noon, which would also lower the measure. I chose both rules before looking at any result and recorded them here so a reader can see that the direction of each choice is conservative.

A reader repeating this cleaning should get exactly 276 rows from the frozen file. If not, the difference is in the application of a rule and should be resolved before any finding is compared with mine. Clear documentation of how a data set reached its final form is part of what improvement reporting guidance expects in the methods (Ogrinc et al., 2016).

What this page is doingNaming which way each judgment pushes the measure is what separates a cleaning memo from a description of tidying a spreadsheet.

References

Hodgson, C., Needham, D., Haines, K., Bailey, M., Ward, A., Harrold, M., Young, P., Zanni, J., Buhr, H., Higgins, A., Presneill, J., & Berney, S. (2014). Feasibility and inter-rater reliability of the ICU Mobility Scale. Heart & Lung, 43(1), 19-24. https://doi.org/10.1016/j.hrtlng.2013.11.003

Ogrinc, G., Davies, L., Goodman, D., Batalden, P., Davidoff, F., & Stevens, D. (2016). SQUIRE 2.0 (Standards for QUality Improvement Reporting Excellence): Revised publication guidelines from a detailed consensus process. BMJ Quality & Safety, 25(12), 986-992. https://doi.org/10.1136/bmjqs-2015-004411

Van den Broeck, J., Cunningham, S. A., Eeckels, R., & Herbst, K. (2005). Data cleaning: Detecting, diagnosing, and editing data abnormalities. PLoS Medicine, 2(10), Article e267. https://doi.org/10.1371/journal.pmed.0020267

Reading the NURS 8740 Week 2 instructions

This week's memo typically asks you to show how your raw data became your analysis file. Start with provenance: the system, the date range, who pulled the data and where an untouched copy is kept. Describe the condition of the file honestly, listing every type of defect before any fix. Write each cleaning step as a rule another analyst could apply and get the same result, and report what happened when you applied it, including values that looked wrong but were correct. Trace the record counts from the number pulled to the number analyzed in a table whose arithmetic reconciles. Finally, identify any step that involved judgment and say which direction it could move your results.

How this NURS 8740 Week 2 example is built

The provenance section names the saved report, the run date, the secured drive and the frozen copy, then defines a row. The condition section lists six duplicates, 51 days under the threshold, 19 exclusions, 14 blank levels and two suspicious level 11 entries, citing Van den Broeck and colleagues on screening, diagnosis and treatment. Five rules follow, each with its result: duplicates had identical values, exclusions were nine bed rest orders, seven comfort-care orders and three open abdomens, blanks stay in the denominator and the level 11 entries were confirmed. Table 1 shows the disposition and the arithmetic. The final section explains why both judgment calls lower the measure and cites SQUIRE 2.0 on documenting methods.

Reading the NURS 8740 Week 2 rubric

Faculty usually evaluate a cleaning memo on reproducibility: could another analyst start from the frozen file and end with the same rows? Statements that the data were cleaned and checked earn little without the rules behind them. The disposition table draws close attention, because counts that do not add up undermine every later result. Credit rises when the writer reports defects that reflect badly on the source system and when suspicious values are investigated rather than deleted. Transparency about judgment calls, including their likely direction of bias, is valued. Clear memo format, precise wording and correct APA citations complete the expectations.

Common NURS 8740 Week 2 mistakes, and how to avoid them

Cleaning memos often say that outliers and errors were removed without saying which ones or by what rule. Others fill missing values from notes or memory, which quietly changes the measure. Some delete implausible values without checking them, and later find the values were real. Keep an untouched copy of the original file before you start, and write each rule before you apply it. After cleaning, add up the removals and check that they match the difference between the starting and final counts. Our writers can draft a cleaning memo for your project's data, with provenance, rules, a disposition table and an honest account of the judgment calls.

Get a NURS 8740 Week 2 example written to your instructions

Send the memo prompt, any template your section supplies, and a description of what the file did wrong, and an example comes back matched to that shape. First custom one, no charge, inside 24 to 48 hours. An extract pulled under site permission belongs to the project that pulled it, and none gets reconstructed here.

More NURS 8740 papers and DNP samples

NURS 8740 Week 2 questions, answered

Where can I find a free NURS 8740 Week 2 sample paper?

Above, with its disposition table, the whole memo on cleaning the mobility extract is published for anyone to go through, free.

How many rows did the NURS 8740 Week 2 memo start and end with?

It started with 352 rows from the saved report and ended with 276 qualifying ventilator days, 90 from the baseline period and 186 from the first eight protocol weeks.

Why were blank mobility levels kept in the file?

The operational definition counts them as days not reaching level 3, and filling them from other notes would bring back the reviewer judgment the measure was designed to avoid.

What happened to the two level 11 entries?

They were checked against the flowsheet and found to be real; both patients walked in the morning and were reintubated in the evening, so the entries stayed unchanged.

Why does the memo say its judgment calls are conservative?

Both choices, keeping blanks as failures and applying exclusions only at 0700, would lower the measure, and both were set before any result was seen.