XY Scatter

[caycee516]

I need help plotting this data on a XY scatter graph. I can't get it correct! I alreadey posted one thread but wasn't sure how to attach a file.

Thanks!

CYCLE COUNT SHEET.xls

[MrShorty]

I find, before I can make a good plot in Excel, I have to think through what I want the plot to show. I don't see any indication of what you expect this plot to show. On the "graph" sheet, you have a plot that shows column 3 plotted without an associated X range (so Excel defaults to the count numbers 1,2,3,4,...). This plot is essentially the same as plotting column 3 (variance) against column 1 (count).
"Sheet1" has a graph plotting column 3 (variance) against column 2 (tags/count).
Either plot is technically "correct". Perhaps if we step away from the Excel side of the question and look at other questions:

1) What do these quantities represent? Obviously I can read the labels you put at the tops of the columns, but that doesn't really explain what these numbers mean.
2) What kind of relationship between the 3 quantities do we expect?
3) In what ways are the two example graphs given inadequate? What are they showing that is "wrong" and what do you need them to show?

[caycee516]

Yes after trying everything, I know the chart is missing something to compare all the data too. The count is for how many counts were done, the # of tags, are the # of tags on each cycle count, and the variance is the amount of inventory we are over (black) or under (red).

I don't have anything to compare it to and i'm really frustrated.

[MrShorty]

I'm not sure there is an easy way for us to know what is missing. Once we figure out what to show on the plot, I expect it will be easy to put it into the chart. The difficult step at this point is knowing what to put into the chart so it is meaningful. Perhaps the following questions will help you think through what you want to show.

1) I still don't know what a "count" is. It sounds like something to do with inventory management. Is a count nothing more than an I.D. number? Is there a time element hidden inside the count value (count 1 is the first count in the history of the company, count 5 was performed 5 minutes/hours/days/weeks/months/years later)?
2) What is a tag? What is different about a count that has 1 tag as compared to a count that has 200+ tags/count? Are all tags created equal, or are some tags larger/smaller than others?
3) Variance is an over/under kind of number. Over/under what? What is different about a variance of 0 compared to something with a variance of 2000? Do these values represent dollar/monetary amounts? There seems to be little relationship between variance and tags. I notice that many entries with tags=0 have a variance of 0, but some have a large variance. What does this difference mean?
4) Overall, what is the goal of a project like this? Descriptive (to describe what has happened)? Are you trying to identify any trends in the data (over time, we are doing worse/better/nochange)? Identify anomalous behavior (most of the time, variance is 0, but there are several entries with large variances. what is going on when the variance is large?)?