What does Kestwick's delivery extract contain, how was it built, and where is it thin? This HI560 Unit 2 description answers all three before a single rate is computed. Searches like "hi 560 unit 2 assignment example", "hi560 unit 2 sample" and "hi560 unit 2 example" land here.
What a finished HI560 Unit 2 data set description looks like
Five pages and a data dictionary appendix. Page one states provenance: a mother-level extract pulled from the electronic record's delivery summary on a stated date, joined to the coded abstract by encounter number, covering January 2024 through December 2025. North contributes 3,688 rows and South 2,724. A field table lists all 38 variables with type, source system, allowed values and percent missing. The missing-data section is the center of the paper. Parity is blank on 131 rows, 97 of them at South. Pre-pregnancy body mass index is missing on 4.1 percent of North rows and 11.9 percent of South's. Labor onset reads none, spontaneous or induced, but 5.6 percent of rows carry a default the paper flags as suspect. A flow diagram then narrows the file to the 2,248 NTSV births later units compare.
How a HI560 Unit 2 example is structured
The description answers three questions in order: where the data came from, what each row means, and what the file cannot say. Provenance comes first because every later figure inherits it, so the extraction date, query logic and join key are stated before any count. The unit of analysis is defined next, one row per delivering mother, with twin births carried as a single row and a note on why babies are not the unit here. Variables are grouped by purpose, identifiers removed, demographics, obstetric history, labor and delivery, and coded fields, rather than listed alphabetically. Missingness is reported by field and by hospital, because a gap concentrated at one site can masquerade as a difference between sites. An NTSV flow diagram and a short list of questions for the data owner close the paper.
Provenance in four lines
Source system, extraction date, query filters and the encounter number used to join the coded abstract. A reader could request the same pull and expect the same 6,412 rows to come back.
One row, one mother
Twins appear once, with the second baby's fields in their own columns. Stating the unit early prevents the commonest error in delivery data: counting babies where a rate needs mothers.
Gaps by field and by site
Parity blank on 131 rows, body mass index missing at nearly three times North's rate at South, and a labor onset default that appeared after an interface change. Each gap carries its count and its hospital.
From 6,412 to 2,248
A flow diagram removes multiparous, preterm, multiple and non-vertex births one step at a time, with the 131 unknown-parity rows held in a separate box rather than dropped silently.
Questions for the data owner
Why South's onset field defaults to spontaneous, whether weights come from the first prenatal visit, and whether a transfer in labor leaves the sending hospital recorded anywhere in the row.
Where marks go in HI560 Unit 2
Descriptions that list variables and stop are where HI560 graders usually find the most to criticize. Most versions of this assignment want the file's origin, its unit of analysis and its gaps, and the gaps carry the weight: a field missing on 12 percent of one hospital's rows is a finding, not a footnote. Reporting missingness only for the whole file hides exactly that pattern. Precision about definitions matters too; parity, gravidity and gestational age each have exact meanings, and a paper that blurs them signals trouble for every later unit. Silent exclusions cost credit, so the unknown-parity rows need their own box. The strongest descriptions end with specific questions for whoever owns the data, since those questions show the writer has read the file rather than its column headers.
Get a HI560 Unit 2 example written to your instructions
Attach the file your section supplied, or its data dictionary if the file is large, plus the rubric and the assignment text for Unit 2. The free first sample is ready in 24-48h, built from that file's own fields and gaps rather than Kestwick's, with missingness reported by group wherever the data allows it.
HI560 Unit 2 questions, answered
How much detail belongs in a data set description?
Enough that a reader could decide whether the file suits a question without opening it. That usually means source, date range, extraction date, unit of analysis, row count, a field table with types and missing percentages, and a note on anything derived. Summary statistics generally wait for the descriptive unit, so this paper stays with structure and completeness.
Should missing values be filled in before describing the data?
No. Describe the file as received, with every gap counted, and leave handling decisions for later units where they can be justified against a specific analysis. Imputing values at this stage hides the very pattern a reviewer should notice, especially when one site or one group accounts for most of the blanks.
What if my data set has no data dictionary?
Build a short one inside the description, labeled as a reading of the file rather than an official source. For each field, infer the type and allowed values from the data, note anything ambiguous, and list the questions a data owner would need to answer. That list of questions often earns credit on its own, because it shows the file was actually examined.