What is the logic of XY (Scatter) charts?

[catbill]

I understand that XY or scatter charts require data on the X-axis as well as the Y-axis.

What I don’t understand is what the Calc 4.2 User Guide says:

“While constructing the chart, if you choose Data Range > Data series in rows, the first row of data represents the X-axis. The rest of the rows of data are then compared against the first of row data. Figure 39 shows a comparison of three currencies with the Japanese Yen. Even though the table presents the monthly series, the chart does not. In fact the Japanese Yen does not appear; it is merely used as the constant series that all the other data series are compared against.”

The example in the user guide shows four rows of data. Only three data series are represented in the chart because they are all compared to the first row of data. The user guide says “the first row of data represents the X-axis.” What does it mean to say that the data is represented in the X-axis?

By contrast, when I create an XY chart (attached), I have not seen the behavior described. When there are three rows of data, three data series appear in the chart.

The guide says that this works for data in rows. But wouldn’t it work the same for data in columns? In the attached file, I created XY charts for data in columns and for data in rows. I saw no difference.

So, I would appreciate some insights about how data is treated in XY charts. Is the description in the user guide correct? If not, what does it say that is incorrect?
Margarine syrup butter chart 5-14-19.ods

[Grantler]

The months are plain text, no numeric values. Therefore XY scatter works like simple line chart.

.

Instead of Jan write 01.01.2019 (format depends on your locale) and transform the other cell contents into numeric values you will see the difference. If there are “irregular”/uneven time spaces the XY scatter will “spread” them in sort of equal/proportional mode…

[robleyd]

Also, this Wikipedia article on Scatter plots may help with the concept.

[catbill]

Thank you, Grantler. I should have noticed that about the dates, especially since I said in the first sentence that the X-axis contains data.

[catbill]

Thank you, robleyd. After reading the other comments, the article makes more sense.

[ve3oat]

Without referring directly to the User Guide myself and going by your quotation, I beieve that what the Guide says is correct. The Yen does not appear as a data series because it is the independent variable against which the other (possibly) dependent variables are compared.

In your chart, you actually do have four series rather than just three, the independent series being Jan, Feb, Mar … , against which the other three (dependent) series are compared. Does this make sense??

And, Yes, it works for both rows and columns, as you have well illustrated.

X-Y scatter charts are very useful, especially when the data has precise values and are complex in nature, not easily represented in simple bar charts or bubble charts. You see them often illustrating scientific and engineering studies.

[catbill]

The idea that the X-axis is independent variables is what I was missing. Now I understand this better. Thank you.

[Regina]

XY-charts are used in cases, which require a coordinate system as you have used it in school mathematics. The data is considered to be given as point with two coordinates. The x-axis is a true number line. That it the difference to a line chart, where the x-axis is a category axis. So you can label an x-axis in a line chart with months, which is not possible in an XY-chart. The label of an x-axis in an XY-chart mark a single point on the axis, in a line-chart the label marks an interval. Only for xy-charts the equation of a regression curve is meaningful.

[catbill]

Thank you for being so clear.