BUSI 3014 Week 6 Descriptive Statistics Summary Example

Most New Customers Waited About Two Business Days, but One in Five Waited More Than Four: A Descriptive Summary of 597 Cleaned Requests From the Second Quarter of 2027 at a Pest Control Company

Student Name

Walden University

BUSI 3014: Data Driven Decision Making Workplace Certificate

Instructor Name

Month Day, Year

Most New Customers Waited About Two Business Days, but One in Five Waited More Than Four: A Descriptive Summary of 597 Cleaned Requests From the Second Quarter of 2027 at a Pest Control Company

The Headline

In the second quarter of 2027, most new residential customers of Kestrel Pest Solutions, a composite pest control company in Knoxville, Tennessee, waited about two business days for a first treatment, but about one in five waited more than four.

That sentence summarizes 597 requests logged from April through June 2027 that survived last week's data audit. Each wait is measured as first-visit wait, from the logged request, adjusted to the call date where the request was logged late, to the first completed treatment, in business days.

The Figures

Table 1

First-Visit Wait, New Residential Requests, April to June 2027

Kestrel Pest Solutions and its data are composite. The adjusted column uses the call date in the dispatcher's notes for requests logged more than a day late.

Why the Median Leads

The mean wait, 3.9 business days, is nearly double the median of 2. The reason is a small group of very long waits. Twenty-three requests waited more than fifteen business days. Most were termite inspections, which only the company's one licensed termite inspector can perform, and bed bug heat treatments, which need a two-person crew and a full day. A few month-long waits are enough to pull the mean up, while the median, the wait of the middle customer, is unaffected by how long they ran. For describing what a typical caller experienced, the median is the better figure. Tukey (1977) built exploratory data analysis around summaries such as the median and quartiles precisely because they resist distortion by a few extreme values.

Spread

The middle half of customers waited between one and four business days. That range is narrow compared with the full range of zero to 31, which says most customers had a similar experience and a minority had a very different one. The most useful single number for the Saturday route question is the share above the four-day threshold: 21.4 percent, or 128 of the 597 requests.

How the Quarter Changed Month by Month

Table 2

Share of Requests Waiting More Than Four Business Days, by Month

Adjusted figures; months total 597 requests and 128 long waits.

The share rose every month as the season built. By June, one new customer in four waited more than four business days. That pattern matters for the owner's decision because it shows the problem growing as demand rises, which is what a capacity shortfall would produce.

The Outliers, Kept and Labeled

The 23 waits over fifteen days were not removed. They are real customers, and removing them would make the company look better than it was. They are reported separately because their cause, specialist services with one qualified person, is different from the cause of most long waits, ordinary treatments in a full schedule. A Saturday route staffed by a general technician would shorten ordinary waits but not these.

Which Requests Waited Longest

The long waits were not spread evenly across request types. Of the 128 requests that waited more than four business days, 71 were general pest treatments, 24 were termite inspections, 19 were rodent exclusions and 14 were bed bug treatments. General treatments can be done by any technician, which is the work a Saturday route would absorb. The specialist services would still depend on the people qualified to do them.

Why Both Columns Are Shown

The audit found that about one request in eight was logged a day or more after the customer called, mostly in busy months. Using the logged timestamp alone lowers the long-wait share from 21.4 to 17.8 percent. Showing both lets the owner see that the adjustment matters but does not change the picture: either way, roughly a fifth of new customers waited more than four days in the spring quarter. Wang and Strong (1996) describe data quality in terms of fitness for use; reporting both versions keeps the owner informed about how the data were made fit.

What the Summary Does Not Say

This summary covers one quarter, and it describes waits only for requests that ended in a visit. The 23 requests held aside for missing visit dates and the 96 canceled before any visit were not measured. If many canceled customers were waiting a long time when they gave up, the true share of long waits is higher. Kaplan and Meier (1958) showed how to estimate durations when some observations end without the event of interest, and that approach could fold cancellations into a future version.

References

Kaplan, E. L., & Meier, P. (1958). Nonparametric estimation from incomplete observations. Journal of the American Statistical Association, 53(282), 457-481. https://doi.org/10.1080/01621459.1958.10501452

Tukey, J. W. (1977). Exploratory data analysis. Addison-Wesley.

Wang, R. Y., & Strong, D. M. (1996). Beyond accuracy: What data quality means to data consumers. Journal of Management Information Systems, 12(4), 5-33. https://doi.org/10.1080/07421222.1996.11518099