No procedure appears in this DDBA 8307B audit. It describes one dataset: each variable's meaning and measurement level, the coding of missing values, and the questions the file cannot address. Searches like "ddba 8307b week 1 assignment example", "ddba8307b week 1 sample" and "ddba 8307b week 1 example" land here.
What a finished DDBA 8307B Week 1 data audit looks like
Three to four pages built around a variable table, one row per column in the file. The rows give the variable name, what it records in plain words, the measurement setting in SPSS's Variable View, the level the construct actually supports, the coding scheme, and counts of valid and missing cases. Stevens's scheme of nominal, ordinal, interval and ratio levels frames the table, and the audit notes where the software's setting and the variable's nature disagree: manager tenure is stored as four numbered bands marked Scale, though the bands are ordered categories. A missing-values section follows, flagging a code of 99 on the satisfaction items that was never declared missing and would be averaged as a real score. A final section states, in sentences a regional director could follow, what comparisons the file supports.
How a DDBA 8307B Week 1 example is structured
The audit moves from the file as a whole to its parts and back. An opening paragraph says where the data came from, what one row represents, and how many rows there are, because a reader who misunderstands the unit of analysis will misread everything else. The variable table comes next and carries the body. After it, a measurement section discusses the disputed cases, chiefly whether a satisfaction index summed from five rating items can be treated as interval. Velleman and Wilkinson's critique of rigid level-based rules is set against Stevens there, and the audit takes a stated position. Missing values get their own section, organized by variable and by pattern. A closing paragraph comes back to the whole file in plain language: what it lets the chain compare, and what it says nothing about.
What one row is
A store-month, not a store and not a customer. Stating the unit of analysis first prevents the most expensive misreading in the course, where a result about stores is later written up as a finding about the people who shop in them.
Settings beside substance
Each variable's SPSS measurement setting sits next to the level its construct can support. Tenure bands marked Scale, a region code stored as a number: the table shows where the software would permit a procedure the variable cannot justify.
The summed index, argued
Five one-to-five items added into a satisfaction score raise the classic dispute. The audit gives Stevens's position and Velleman and Wilkinson's reply, then says how the index will be treated and what that choice risks.
Missing, by code and by pattern
Blank cells, undeclared codes of 99, and stores that closed partway through the year are each counted separately. The pattern matters as much as the count, since values missing for one kind of store bias every average that follows.
What the file cannot say
The closing paragraph is addressed to a manager: the file compares stores and regions over one year, and it holds nothing about individual customers, about stores closed before the extract, or about why any score moved.
Where marks go in DDBA 8307B Week 1
Week 1 marks reward seeing the file before using it. An audit listing variable names with their SPSS types earns little, because the software already displays that; credit goes to the column explaining what each variable means and whether its stored level is defensible. Markers read the measurement section closely, and treating a set of rating items as interval without comment is the error most drafts make. Taking a position on it, with a source on each side, earns the analytic share. Missing values carry substantial weight: an undeclared code that would enter an average as a real score is exactly the kind of flaw a careful audit catches, and silently dropping incomplete rows draws comment. The closing plain-language paragraph is scored separately, since this course measures translation from the first week.
Get a DDBA 8307B Week 1 example written to your instructions
Attach the Week 1 prompt and rubric, plus the codebook or variable list that came with your assigned dataset. An audit with its full variable table returns within 24-48h, and the first one is free. Nothing needs to be run for it; the sample audits a look-alike file, and your own data stays on your machine.
DDBA 8307B Week 1 questions, answered
Does the audit include descriptive statistics?
Only counts of valid and missing cases per variable, and perhaps the categories present. Means, standard deviations and distributions belong to the next week's work, and reporting them here blurs the purpose, which is understanding the file before summarizing it. An audit that opens with a table of means usually skips the measurement questions that decide whether those means are meaningful.
Can a Likert-type index be treated as interval data?
Many analysts do, and the practice has defenders, but the audit has to argue it rather than assume it. Stevens's scheme would call a single rating item ordinal; Velleman and Wilkinson argued that rigid level rules mislead, and summed multi-item scores often behave well enough for interval methods. State the position, cite both sides, and name the risk if the choice proves wrong.
What should the audit say about missing values it cannot explain?
That they exist, how many, and on which variables, with any visible pattern described in words. Unexplained missingness is a finding, not an embarrassment, and it limits later claims. The audit also notes whether incomplete rows will be excluded listwise or pairwise, since that choice changes the sample each procedure uses and a reader deserves to know which applies.