BUS 352 Topic 6 two sample test report example

Business Statistics Grand Canyon University Free custom sample in 24 to 48h

This BUS 352 Topic 6 sample paper reports a trial comparing two freight carriers' lead times and on-time rates. It belongs to GCU BUS 352, Business Statistics, the course listed as BUS-352 in the Grand Canyon University BS in Business Management. The samples are independent, lead time is quantitative and on-time delivery is a proportion, so different tests apply. Forty-eight shipments averaged 3.9 days with a standard deviation of 1.1 and 45 averaged 3.4 with 1.6, so the unequal-variance t test is used. The half-day difference has a p-value near 0.085 and an interval from 0.07 days slower to 1.07 faster.

CourseBUS 352 Business Statistics
TopicTopic 6
Paper typeTwo sample test report
LengthAbout 990 words, 2 pages
FormatAPA 7 student paper
SchoolGrand Canyon University (GCU)
ProgramBS in Business Management
UpdatedSeptember 2026

Free sample paper for BUS 352 Topic 6

1

Is the Challenger Carrier Faster? A Half-Day Advantage That Does Not Reach Significance, and Why the Choice of Test Is Part of the Answer

[Author Name]

Colangelo College of Business, Grand Canyon University

BUS-352: Business Statistics

Topic 6 Assignment

[Instructor Name]

[Date]

The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.

What this page is doingThe title states the finding and the methodological point the report makes. The company and all figures are illustrative; the sources are real.
2

Is the Challenger Carrier Faster? A Half-Day Advantage That Does Not Reach Significance, and Why the Choice of Test Is Part of the Answer

The Question

Pinecrest Home Supply, a composite regional distributor, ran a trial of a second freight carrier for two months alongside its current carrier. The logistics manager wants to know whether the challenger delivers faster, because faster delivery would let Pinecrest promise shorter lead times to contractors. Two outcomes were recorded for each shipment: lead time in days from pickup to delivery, and whether the shipment arrived by the promised date. All figures are illustrative.

Choosing the Tests

The choice of test is part of the answer, so it comes first. The shipments on each carrier were different shipments, assigned by the dispatch team according to which carrier had a truck available, so the two samples are independent rather than paired. A paired test would be appropriate only if the same shipments had been sent both ways, which is impossible here.

The two outcomes also need different procedures. Lead time is quantitative, measured in days, so the comparison is of two means. On-time delivery is a yes-or-no outcome, so the comparison is of two proportions (Levine et al., 2020).

For lead time, the two samples have noticeably different spreads: a standard deviation of 1.1 days for the current carrier and 1.6 for the challenger. The pooled t test assumes equal variances, which these data do not support. The report therefore uses Welch's unequal-variance t test instead, since it drops the equal-spread assumption and adjusts the degrees of freedom accordingly (Welch, 1947).

The Data

Summary statistics for the trial period are below.

Table 1

Trial Summaries

Current carrierChallenger
Shipments4845
Mean lead time3.9 days3.4 days
Standard deviation1.1 days1.6 days
Delivered on time40 of 48, 83.3%39 of 45, 86.7%

Checking the Conditions

The t test relies on the sample means being approximately normal. Lead times are right-skewed, since a delivery cannot take fewer than one day but can occasionally take seven, and a histogram of each carrier's lead times shows that lean. With samples of 45 and 48, the central limit theorem makes the sample means close enough to normal for the test to be reliable; neither sample has extreme outliers that would dominate its mean. The two-proportion test requires at least about ten successes and ten failures in each group. On-time counts of 40 and 39 meet that easily, but late counts of 8 and 6 are small, which makes the proportion comparison less precise and is one reason its result is so inconclusive.

The Results

Each outcome was tested at the 0.05 significance level, with a two-sided alternative, since the manager did not rule out the challenger being slower.

Table 2

Test Results

ComparisonTestResultp-value95% interval for the difference
Mean lead timeWelch two-sample t testChallenger 0.5 days faster; t = 1.75, about 77 degrees of freedom0.0850.07 days slower to 1.07 days faster
On-time shareTwo-proportion z testChallenger 3.3 points higher; z = 0.450.65Not significant

Reading the Lead Time Result

The challenger averaged half a day faster in the trial. With a standard error near 0.29 days, the t statistic comes to about 1.75 on roughly 77 degrees of freedom, and the chance of a gap this large or larger if the carriers were equal is about 0.085. The difference does not reach significance at the 0.05 level. The 95 percent confidence interval shows why: the true difference could plausibly be anything from the challenger being 0.07 days slower to about 1.07 days faster. The data are consistent with a real advantage of about half a day, and also with no advantage at all (Anderson et al., 2020).

Why Not a Pooled Test

Had the report used the pooled t test, which assumes both carriers have the same spread, it would have combined the two standard deviations into one and produced a slightly different result. With samples of similar size, the difference in the p-value would be small here, but the choice would rest on an assumption the data contradict: the challenger's standard deviation is nearly half again as large as the current carrier's. Welch's version costs little when the spreads happen to be equal and protects against misleading results when they are not, which is why many analysts use it by default for comparing two independent means.

Reading the On-Time Result

The challenger delivered 86.7 percent of shipments on time against 83.3 percent for the current carrier. With samples this size, a difference of 3.3 points is well within what chance alone would produce; the p-value is about 0.65. The trial gives no evidence that either carrier is more reliable at meeting promised dates.

What This Means for the Decision

Pinecrest cannot conclude from this trial that the challenger is faster, and it certainly should not promise contractors shorter lead times on the strength of it. Nor can it conclude that the challenger is no faster: a trial of about 45 shipments per carrier is too small to detect a half-day difference reliably when lead times vary by more than a day. Two options follow. If the challenger is cheaper, or equal in cost, Pinecrest could shift some volume to it on cost grounds while continuing to measure. If speed is the main reason to switch, a longer trial is needed: roughly 120 shipments per carrier would give a good chance of detecting a true half-day difference, if one exists.

A Note on the Challenger's Spread

The challenger's larger standard deviation deserves attention in its own right. A carrier that is faster on average but less consistent can be harder to plan around, because contractors care about the worst likely delivery as much as the average. In the trial, the challenger's slowest 10 percent of shipments took more than five days, against about four and a half for the current carrier. Any decision should weigh consistency alongside speed.

References

Anderson, D. R., Sweeney, D. J., Williams, T. A., Camm, J. D., Cochran, J. J., Fry, M. J., & Ohlmann, J. W. (2020). Statistics for business and economics (14th ed.). Cengage Learning.

Levine, D. M., Szabat, K. A., & Stephan, D. F. (2020). Business statistics: A first course (8th ed.). Pearson.

Welch, B. L. (1947). The generalization of "Student's" problem when several different population variances are involved. Biometrika, 34(1-2), 28-35. https://doi.org/10.1093/biomet/34.1-2.28

BUS 352 Topic 6 instructions, in plain terms

The Topic 6 instructions in BUS-352 ask students to compare two groups with a hypothesis test. Weak submissions choose a test without explaining why. The assignment rewards reports that treat the choice of test as part of the answer: independent or paired samples, means or proportions, equal or unequal variances. Students should check the conditions each test relies on, report test statistics, p-values and intervals, read a nonsignificant result correctly as inconclusive rather than as proof of no difference, estimate the sample size needed to detect a meaningful difference and connect the findings to the decision.

How the BUS-352 Topic 6 example is put together

The finished report treats the choice of test as part of the answer. It notes that the shipments on each carrier are different shipments, so the samples are independent rather than paired, and that lead time in days is quantitative while delivered-on-time is a yes-or-no outcome needing a different procedure. Illustrative summaries follow: 48 shipments averaging 3.9 days with a standard deviation of 1.1 for the current carrier, 45 averaging 3.4 days with a standard deviation of 1.6 for the challenger. Because the spreads differ, the report uses the unequal-variance version of the t test and says why. The half-day difference does not reach significance at the 0.05 level, and the interval for it runs from slightly below zero to just over a day. The report says plainly that this does not prove the carriers equal.

Where the marks sit in the BUS 352 Topic 6 rubric

Faculty grading the test report reward the choice of procedure justified from the data. Excellent reports explain why the samples are independent and why Welch's test suits unequal spreads. Credit goes to reporting the confidence interval and reading it as consistent with both a real advantage and none. The rubric values checking normality and success-failure conditions. The highest level estimates the trial size needed and notes the challenger's larger spread as a planning risk. Reports that call a nonsignificant difference proof that the carriers are equal usually lose marks.

Common BUS 352 Topic 6 mistakes, and how to avoid them

This topic can be marked wrong outright, and the first place it happens is the choice of test. Treating the on-time outcome as a measurement and running a t test on it, where the course expects a comparison of two proportions, is typically marked as the wrong procedure. Treating the two carriers' shipments as paired when they are different shipments, or pooling variances that clearly differ, produces a statistic the design does not support. An assumption asserted rather than examined leaves the result resting on faith. The most common interpretive loss is reading a p-value above 0.05 as proof the carriers perform the same, when the interval shows a difference of up to a day is still plausible. Recommendations that switch carriers on a half-day gap the test could not confirm overstate the evidence in the other direction.

Get a BUS-352 Topic 6 example written to your instructions

Send the BUS-352 Topic 6 instructions, the rubric from your classroom and the two data sets or the output you are working with. We write a custom example to those criteria, with the test chosen from the design and data type, every assumption examined and the result translated into the business decision, in 24 to 48 hours. The first is free.

More BUS 352 papers and BS in Business Management samples

BUS 352 Topic 6 questions, answered

When should the two samples be treated as paired?

When each observation in one group is matched to a specific observation in the other, such as the same routes shipped by both carriers or the same stores measured before and after a change. Pairing removes differences between routes or stores from the comparison, which usually sharpens it. When the groups are simply different shipments, as in this example, an independent-samples procedure is the correct one.

Why use the unequal-variance t test?

Because the pooled version assumes both carriers have the same underlying spread, and these samples suggest they do not. The unequal-variance version, often called Welch's test, drops that assumption and adjusts the degrees of freedom, at very little cost when the spreads happen to be similar. Many analysts use it by default for two independent means, and the example states that choice rather than leaving it to the software.

If the result is not significant, should the company keep the current carrier?

Not automatically. A non-significant result says the data do not clearly show a difference in average lead time, and the interval shows the challenger might be up to a day faster. The decision can reasonably turn on other evidence: price, the on-time comparison, and the challenger's less predictable deliveries. The example recommends a longer trial with routes assigned at random, which would answer the question more cleanly.

Where can I find a free BUS 352 Topic 6 sample paper?

Every part is above: the question, choosing the tests, the data, checking the conditions, the results, reading lead time, why not a pooled test, the on-time result, what it means and the challenger's spread. Describe your trial for a free worked report.

Why use Welch's test in BUS 352 Topic 6?

The carriers' standard deviations, 1.1 and 1.6 days, differ noticeably, and Welch's test does not assume equal variances.