Read for shifts, trends, runs and astronomical points under Perla, Provost and Murray's rules, the NU803 Unit 7 chart shows a sustained drop in low-glucose days without licensing a causal claim. Searches like "nu 803 unit 7 assignment example", "nu803 unit 7 sample" and "nu803 unit 7 example" land here.
What a finished NU803 Unit 7 run chart interpretation looks like
One annotated chart dominates four pages: twenty-eight weekly points with the baseline median of [6.95] drawn across the sixteen baseline weeks and extended forward, dashed. Go-live is marked, as are the start of the handover line and a [two]-day card printing outage. Beneath the chart, a table applies each of the four rules in turn, with the count that decides it. Shift: [ten] consecutive points below the median from week [3] after go-live, where six would suffice. Trend: the longest consecutive rise or fall covers [four] points, short of five. Runs: [nine] across [28] useful observations, under the table's lower limit of [ten]. Astronomical point: week [2] at [9.49] is discussed and not declared one. What the chart does not show gets the last section.
How a NU803 Unit 7 example is structured
Rules come before data, so the reader can see nothing was chosen after looking. Each rule is given accurately. A shift needs at least six points in a row all above or all below the median, and a point sitting exactly on it is skipped, neither extending nor breaking the run. A trend needs five or more points all rising or all falling, with repeated values counted once. Runs are counted as crossings of the median plus one and compared with the published table for the number of points not on the median. An astronomical point is a judgment by people who know the process, with no numerical threshold. Read together, a shift and too few runs agree, and the audit, early doses down from [38] to [14] percent, is consistent with them. Cause is not claimed.
Rules stated before results
The four rules from Perla, Provost and Murray appear first, in their published form, so the reading that follows cannot be accused of choosing a test to fit.
The shift
Ten points below the extended baseline median, starting in week [3], meet the six-point rule with room to spare; the two-week delay is noted, not explained.
No trend, and why that matters
The longest run of consecutive rises or falls stops at [four] points, so the drop reads as a change in level rather than a steady slide.
Too few runs
Nine runs across twenty-eight useful observations falls below the table's lower limit, a second signal reached by a different rule than the shift.
The tallest point
Week [2] at [9.49] is weighed as a possible astronomical point and not declared one, pending a patient-level look at who contributed its low days.
Where marks go in NU803 Unit 7
Accuracy about the rules is the first thing an NU803 marker checks in a Unit 7 interpretation. A shift described as five points, a trend counted to include repeated values, or runs judged by eye instead of by the table suggest the rules were half-remembered, and that tends to cost more than any single misreading. Comparing the average before with the average after, as though the chart were a bar graph, misses what the unit is teaching. Annotations matter: a chart without the go-live line, or without the events that might explain a point, leaves the reader guessing. Credit follows interpretations that read signals together and admit what the design cannot rule out. Causal language, the bundle reduced hypoglycemia, is corrected in many sections; a sentence reporting a sustained change after go-live reads as disciplined.
Get a NU803 Unit 7 example written to your instructions
Upload the weekly figures behind your chart, or the chart itself with its median marked, with the date your change began and the Unit 7 prompt and rubric, and a free first custom sample returns within 24-48h applying each run chart rule to your series, the deciding count shown, plus a short list of claims the chart cannot support.
NU803 Unit 7 questions, answered
Should the median come from the baseline or the whole series?
For judging whether a change moved the process, the baseline median extended forward is the usual choice, since a median computed over all points is pulled toward the new level and weakens the comparison. The sample freezes its baseline median at [6.95] and notes that a new median could be calculated once the new level has held long enough.
Is a run chart enough, or does the analysis need a control chart?
For many improvement projects a run chart is enough, and it needs fewer assumptions. Control charts add limits based on the data's variation and suit longer series or stable processes. The sample uses a run chart because twenty-eight weekly points and a changing process fit its strengths; a later unit adds a statistical comparison for readers who expect one.
What counts as an astronomical data point?
A value obviously and blatantly different from the rest, judged by people who know the process rather than by a formula. It is the most subjective of the four signals. The sample examines its tallest week and declines to call it astronomical, because a few patients with repeated lows could produce it, and says what data would settle the question.