AC570 Unit 8 Data Interpretation Seminar Reflection Example

Triple the Median, and Still Not a Problem Site: A Unit 8 Seminar Reflection on Reading a Ratio

[Student Name]

Purdue University Global

AC570: Data Analytics for Accountants

Unit 8 Seminar Reflection

[Instructor Name]

[Date]

Stowfield Storage, the classmates and every figure are composite; the textbook, book and guide cited are real.

The Session

The facilitator projected the facility-month credit file from the regression exercise and asked which sites deserved a field visit. Each of us posted an answer in the chat before discussion.

My Position

I wrote: "Facility 9 and Facility 22. Both are above twice the median credit ratio, and Facility 9's residuals are positive in 10 of 12 months, $8,810.65 over the year against the refit." I thought Facility 9 was the stronger case of the two because its ratio, 19.51 percent of billed rent, was the highest in the company by far, and its pattern was steady rather than a single bad month.

The Question

Naveen asked what Facility 9 was doing in 2025 that other facilities were not. I did not know. The facilitator pulled the facility list: Facility 9 opened in late 2024 and spent 2025 in lease-up, filling empty units. Almost all its move-ins came on the one-dollar promotion, and its occupancy rose from 31 to 74 percent.

That changed both numbers I had relied on. The ratio was high because billed rent was low, since most units were empty or in their first, credited month, and credits were high because nearly every tenant was new. A lease-up facility will always have a ratio several times the median. And the residuals were positive partly because Facility 9's units are newer and climate-controlled, with higher rents, so its promotion credit per move-in is larger than the model's average of about $98. Naveen estimated that the rent difference alone would account for most of the $8,810.65.

Comparing the Two Facilities

Facility 22 credits 11.8 percent of billed rent. Facility 9 credits 19.51 percent. But Facility 22 is a mature site at 92 percent occupancy, its credits run $36,730.50 over what the model expects, and the exception tests found refunds paid twice, gaps in receipt numbers, evening manual credits and amounts clustered just below the approval threshold there. Facility 9 has none of those. On every measure except the raw ratio, Facility 22 is the problem and Facility 9 is a new facility doing what new facilities do.

I had read the ratio as if every facility were in the same situation, when the same 19.51 percent means something entirely different at a site in its first year of lease-up, and the question that exposed it was simply what the facility was doing that the others were not.

What I Felt in the Room

I was a little embarrassed when the lease-up fact came out, because it was in the facility list we had all been given and I had not looked. Naveen did not know the answer either; he just asked the question. That struck me more than the answer did. I had treated the analysis as finished once the numbers ranked the facilities, when the ranking was really the start of a set of questions about each one.

Afterward I reread my chat post and noticed that it contained two numbers and no facts about the facility at all, not even its opening date, which was in the first column of the file.

Why I Got It Wrong

I trusted a ratio because it was large and a residual because it was persistent, and I never asked about the business behind either. The regression could not know that Facility 9 was new, because I had not given it a variable for facility age or rent level. A model's residuals are only as meaningful as its variables are complete, and analytics texts warn that context is what turns a pattern into a finding (Richardson et al., 2023).

What the Room Concluded

The class agreed on Facility 22 for a field visit. For Facility 9, we suggested adding facility age and average rent to the model next year, so lease-up sites are compared with an expectation that fits them (American Institute of Certified Public Accountants, 2017).

The Habit I Now Keep

Before I call any ratio a red flag, I ask what the unit was doing that period that others were not. Here is the four-line comparison I now draw for every flagged site: credit ratio, residual, occupancy trend and exception test results. Facility 9: 19.51 percent, plus $8,810.65, rising from 31 to 74 percent, no exceptions. Facility 22: 11.8 percent, plus $36,730.50, stable at 92 percent, four tests flagged. Laid out that way, the answer was never close (Nigrini, 2012).

References

American Institute of Certified Public Accountants. (2017). Guide to audit data analytics. AICPA.

Nigrini, M. J. (2012). Benford's law: Applications for forensic accounting, auditing, and fraud detection. Wiley.

Richardson, V. J., Teeter, R. A., & Terrell, K. L. (2023). Data analytics for accounting (3rd ed.). McGraw Hill.