Which is Best for Your Business?
ADVISORY
By Dr. Kenneth Smith, PMP
Honolulu, Hawaii
& Manila, The Philippines
A former participant of one of my Program/Project Monitoring & Evaluation (PM&E) workshops who monitors sales at multiple store outlets recently contacted me with this concern about outliers: “I want to get the average daily sales per asset type. To get a more realistic average sales per asset type, I need to identify and exclude stores with outlier performance. May I know which approach I should take?”
After replying, I thought this issue would be of sufficient interest to warrant a review of basics.
There are three ‘averages’ in general use for measuring and comparing performance by ‘cohorts’ — i.e. groups in which individual members share a common characteristic. Since each average has a different focus, it is important to use the most appropriate one for your purpose, so as not to misrepresent the results to stakeholders; or deceive yourself – with unintended consequences!
The three averages, and the essential distinction between them is as follows:
1. The MEAN – more precisely the Arithmetic Mean – is probably the average most widely known, generally understood and used. The Mean is obtained by adding the values of all the data items in the group, then dividing that sum by the number of items. For example:
Given 11 items with the values 2 3 3 4 5 12 13 20 25 28 60 respectively,
the sum is 175, which divided by 11 results in an Arithmetic Mean of 15.91
2. The MEDIAN is simply the value of the item at the midpoint of a range of data, ranked from low to high, or high to low, with no other consideration of their values.
In that same range, 2 3 3 4 5 12 13 20 25 28 60, the Median is 12.
[Given an even number of items, the Median is the Mean of the two middlemost values.]
3. The MODE is the most frequently recurring value (if any) in a data series.
In the foregoing range, 2 3 3 5 12 13 20 25 28 60 the Mode is 3.
So, between options 15.91, 12, or 3, which would be the most appropriate one to use as a realistic average to represent this group? Or, perhaps, ‘None of the above!’
Well, as William Shatner says in his TV show “The UNXPLAINED” on the History Channel,
“That’s what we’ll try and find out!”
To better appreciate the utility of each average, I’ll use this example to take a closer look at each, in turn. Then, to address & assess the problem my inquiring former participant posed, I’ll apply all three averages on a pertinent data set, to clarify the issue for your consideration and future use.
The Mean
While the Mean of 15.91 was the largest result — and it certainly is precise — that may also tempt you to use it as the most impressive of the three options. Furthermore, the formula is readily available & accessible — in cell phones, calculators, and computers. However, if you intend to measure and compare performance levels within the group, establish standards, or set targets, note that in this example 15.91 is not an actual performance level within the cohort. Consequently, although arithmetically correct & statistically factual, it would be disputed by many as an arbitrary value. Furthermore, despite being statistically within 2 standard deviations – i.e. 95% probability — it can clearly be seen in Figure 1, that 15.91 is way beyond the reach of at least 5 members in the cohort. So if 15.91is used as a criterion, standard or target, it would be unreasonable for them.
More…
To read entire article, click here
How to cite this work: Smith, K. F. (2025). Given Several Extant Averages; Which is Best for Your Business? PM World Journal, Vol. XIV, Issue XII, December Available online at https://pmworldlibrary.net/wp-content/uploads/2025/12/pmwj159-Dec2025-Smith-which-is-best-average-to-use-advisory.pdf
About the Author
Dr. Kenneth Smith
Honolulu, Hawaii
& Manila, The Philippines
Initially a US Civil Service Management Intern, then a management analyst & systems specialist with the US Defense Department, Ken subsequently had a career as a senior foreign service officer — management & evaluation specialist, project manager, and in-house facilitator/trainer — with the US Agency for International Development (USAID). Ken assisted host country governments in many countries to plan, monitor and evaluate projects in various technical sectors; working ‘hands-on’ with their officers as well as other USAID personnel, contractors and NGOs. Intermittently, he was also a team leader &/or team member to conduct project, program & and country-level portfolio analyses and evaluations.
Concurrently, Ken had an active dual career as Air Force ready-reservist in Asia (Japan, Korea, Vietnam, Indonesia, Philippines) as well as the Washington D.C. area; was Chairman of a Congressional Services Academy Advisory Board (SAAB); and had additional duties as an Air Force Academy Liaison Officer. He retired as a ‘bird’ colonel.
After retirement from USAID, Ken was a project management consultant for ADB, the World Bank, UNDP and USAID.
He earned his DPA (Doctor of Public Administration) from the George Mason University (GMU) in Virginia, his MS from Massachusetts Institute of Technology (MIT Systems Analysis Fellow, Center for Advanced Engineering Study), and BA & MA degrees in Government & International Relations from the University of Connecticut (UCONN). A long-time member of the Project Management Institute (PMI) and IPMA-USA, Ken is a Certified Project Management Professional (PMP®) and a member of the PMI®-Honolulu and Philippines Chapters.
Ken has two KENBOOKS: 1. Project Management PRAXIS which includes many innovative project management tools & techniques; and describes a “Toolkit” of related templates, and 2. MUSINGS on Project Management — a compilation of contemporary concerns in project planning, monitoring & evaluation, with some tools & techniques suggested for their solution. Either or both books are available from Amazon, and their related templates are available directly from him at kenfsmith@aol.com on proof of purchase.
To view other works by Ken Smith, visit his author showcase in the PM World Library at https://pmworldlibrary.net/authors/dr-kenneth-smith/
