Matched store pairs, a twelve-week test and a simulation that sets the smallest detectable change define the experiment design memo written for GB794 Unit 3. Searches like "gb 794 unit 3 assignment example", "gb794 unit 3 sample" and "gb794 unit 3 example" land here.
What a finished GB794 Unit 3 experiment design memo looks like
Five pages go to one decision maker, the script held in an appendix. Page one states the decision, whether to cut the bar set from 140 items to 90 chain-wide, and the answer the test can give by a named date. Page two describes assignment: 44 stores ranked by last year's bar sales, paired neighbor with neighbor, one store in each pair drawn by seeded code to carry the reduced set. The outcome section names category dollar sales per store-week as primary, plus two secondary measures, total basket value and the share of loyalty baskets containing any bar. A power table follows, reporting simulated power of 41, 63, 81 and 92 percent at changes of 3, 4, 5 and 6 percent. A last page covers risks, the stopping rule and slotting terms.
How a GB794 Unit 3 example is structured
What the approver must accept sets the memo's order. Decision and date come first because a merchandising executive approves tests that answer a question on the calendar. Assignment follows, and the memo explains the store as the unit plainly: shelves are physical, so shoppers cannot be randomized, and 22 pairs is the whole sample the chain can supply. The effect size of interest comes from economics rather than convention; savings on inventory and labor would be erased by a category decline beyond about 4 percent, so the design has to detect a change near that size. The power table then admits that a 3 percent shift would usually go unseen. Treatment is defined as the reduced set as it would actually be merchandised, wider facings included, because that bundle is what the chain would roll out.
The decision on page one
A chain-wide cut from 140 bars to 90, approved or rejected by a stated week. Leading with the decision tells the reader what the experiment is for and lets every later choice be judged by whether it helps answer that particular question.
Pairs, a seeded draw and the whole chain
Stores are sorted by prior bar sales and paired in order, then assignment within each pair comes from a seeded random draw recorded in the appendix. Pairing buys precision that a sample of 44 could not otherwise afford.
Power from a simulation, not a formula
The script resamples two years of store-week sales, imposes a chosen effect on treated stores and reruns the planned analysis 2,000 times. Store-level variation and week-to-week correlation come from real data, which standard calculators would have had to assume.
Defining the treatment honestly
Fewer items arrive with wider facings for the survivors, and the memo treats the two as one treatment. Separating them would need a second arm the chain cannot staff, and the rollout decision concerns the bundle anyway.
A stopping rule before launch
If treated stores trail their pairs by more than 8 percent in category sales for three consecutive weeks, the full set returns. Fixing the rule before any data arrive keeps one nervous early week from ending the test on impulse.
Where marks go in GB794 Unit 3
Approval is built into the task in most sections, so a design nobody could sign starts behind. A memo proposing to randomize shoppers to shelves, or to recruit a hundred stores the chain does not have, has not engaged the setting. Power asserted without a calculation draws comment, and power computed from a textbook effect size with no link to the business decision reads as ritual. Scripts pasted into the body instead of an appendix, or reported without their assumptions, cost clarity. Readers also check how treatment is defined; memos pretending that fewer items and wider facings are separable, without a design that separates them, overclaim. A stopping rule, secondary outcomes chosen to catch spillover and a named decision date mark the memos that read like work a merchandiser would sign.
Get a GB794 Unit 3 example written to your instructions
Describe the decision your experiment should inform, the units you could realistically randomize and any historical data you hold, together with the Unit 3 instructions and rubric. The memo returned sizes the test by simulation, defines treatment as it would really run and sets a stopping rule before launch. No charge for a first sample; allow 24-48h.
GB794 Unit 3 questions, answered
What does power by script mean in a design memo?
Instead of plugging assumed values into a formula, a short program simulates the planned experiment many times using real historical data, adds a chosen effect and records how often the analysis detects it. It suits designs like store pairs, where clustering and correlation over time make textbook formulas unreliable. Instructors usually want the code attached and its assumptions stated in the body.
Is twenty-two pairs enough for a doctoral design?
It is enough if the memo is honest about what it can detect. Here the simulation shows about 80 percent power for a 5 percent change and far less for 3 percent. A design that states its detectable effect, and ties that figure to a business threshold, is stronger than one claiming a large sample it could never assemble.
Does the memo need ethics review language?
A shelf test analyzed through aggregate store sales usually raises few human-subjects questions, but sections often ask the memo to say so explicitly and to note what review the university requires before any data enter a dissertation. Loyalty-card data are a different matter, and the memo flags them as needing a data agreement and whatever review the program specifies.