-3

Here is a table of decreasing negative percentages, how would this be interpreted?

I think more items were shipped on time in July than in May or June. Is this correct ?

Metric May June July
% of items shipped on time -12% -8% -5%
Thomas Andrews
  • 177,126
m0ng00se
  • 9
  • 2
  • 2
    Ask the author of the spreadsheet. It makes no mathematical sense to have the number negative-there are a positive number that are late out of a positive total number, so the percentage late should be positive. I suspect the author found it easier to have this negative for some calculation of expected revenue, but that is a wild guess. In any case it is not a mathematical question. – Ross Millikan May 16 '21 at 01:49
  • Depending upon context, this could simply be the mathematical equivalent of "poor grammar". (It's also "poor grammar" to label a value "% of [whatever]" and then also include the "%" sign on the value.) – Blue May 16 '21 at 01:56
  • 2
    The fact that we have changed from late payments to on time shipments indicates that you are not asking the real question. -1 – Ross Millikan May 16 '21 at 02:30

1 Answers1

1

There's not a definite mathematical convention for what this should mean Two possibilities that immediately spring to mind are:

  • The dashes are just typos, introduced while copy-pasting the data between different software or formats.

  • The table is actually comparing how many late payments there were in each month with the corresponding number from the previous year, and fewer people are late now. Business analysts seem to love such comparisons. But even if that is the case there are two possible things it could be:

    • Differences in percentage points. For example, perhaps last year 250 of 1000 payments in May were late, and this year 260 of 2000 payments were late -- a decrease from 25% of all payments to 13% of all payments, or 12 percentage points.

    • Differences in percent of the raw number of late payments. For example, perhaps last year there were 250 late payments in May, and this year there were only 220, which is a decrease by 12% of 250.

In order to be sure you may need to ask the source you've gotten the numbers from.

Hmm, the question was edited so the numbers that used to be about late payments are now about on-time shipments. That sounds strange -- if you're not even sure what the numbers are about, how much can you trust the numbers themselves?

Troposphere
  • 7,158