PlanMaker DEVSQ: Sum Squared Deviations and Derive Variance

• Introduced: Documented by the PlanMaker 2010 manual; the first introduction release is not established.
• Availability: Documented in PlanMaker 2026; this is not a final working-version endpoint.
• Scope: The reference belongs to the 2026/NX manual family. Individual FreeOffice and mobile-build availability remains unverified.
• Checked: October 14, 2026, against the official function reference. Examples use vendor results and independent calculations; no native PlanMaker test was performed.

=DEVSQ(Number1 [, Number2, Number3 ...])

The official DEVSQ reference describes the sum of squared deviations from the sample mean and says empty cells, text and logical values are ignored. Its heading uses the wording average square deviation, but the body and numeric example establish a sum, not an average of the squared deviations.

The current signature requires the first numeric argument and makes further arguments optional. This guide supplies either complete numeric ranges or explicit numbers. It does not extend the short input note into every direct-argument, error or coercion case.

Squaring makes deviations nonnegative and gives the result squared measurement units. A DEVSQ result in minutes squared is not directly comparable to a mean stated in minutes.

Create a blank sheet named SquareDeviations. Put Value, Deviation and Squared deviation in A1:C1. Enter 2, 2, 2, 3 in A2:A5.

Enter =AVERAGE(A2:A5) in E2; expect 2.25. In B2 enter =A2-$E$2 and in C2 =B2^2. Fill B2:C2 through row 5.

In E4 enter =DEVSQ(A2:A5) and in E5 enter =SUM(C2:C5). Both expected values are 0.75, the total of the four squared deviations.

For a separate control, put =DEVSQ(2,4,6) in E7. Expect 8, because the mean is 4 and the squared deviations are 4, 0 and 4.

Keep the same four-value dataset. Put these formulas in G2:G5:

The comparison functions use explicit copies of the four fixture values. Update those literal controls if you change A2:A5.

There are four observations. Population variance divides the square sum by 4; sample variance divides it by 3. DEVSQ itself has not chosen either denominator. Calling the arithmetic mean a sample mean does not imply that DEVSQ already uses n−1.

In G7 enter =STDEV.S(2,2,2,3); expect 0.5, the square root of 0.25. Standard deviation returns to the original measurement units, while DEVSQ and variance remain in squared units.

Create a separate sheet named Processing. Put 20, 20, 20, 30 in B2:B5, with Minutes in B1. Enter =AVERAGE(B2:B5) in D2 and =DEVSQ(B2:B5) in D3. Expected results are 22.5 minutes and 75 minutes squared.

These times are ten times the original 2,2,2,3 values, so the square sum is one hundred times 0.75. In C2 enter =B2/60 and fill to C5; enter =DEVSQ(C2:C5) in D5. The expected result is 0.0208333333 hours squared, equal to 75/3600, not 75/60.

Do not change units merely by relabeling the result. The square in the statistic determines how the conversion factor must be applied.

On a separate sheet named RepeatedData, enter 2, 2, 2, 3, 2, 2, 2, 3 in A2:A9. Put =DEVSQ(A2:A9) in C2; expect 1.5, twice the original total. The mean remains 2.25.

This illustrates why an unnormalized square sum should not be compared across different sample sizes as though it were already variance. Repeated rows receive repeated weight; duplicating a dataset does not create new independent observations.

The 2010 manual prints two arguments before its optional bracket, whereas the current signature makes only the first required. The following historical page gives the same 0.75 example. This is a documented syntax presentation difference, not evidence for a particular old runtime error or a release boundary.

The short reference does not establish every error token, empty-input behavior or floating-point limit. Keep source errors visible and use a tolerance for numerical comparisons. AVEDEV uses absolute rather than squared distances; VAR.S and VAR.P normalize the same square sum for different statistical questions.