Adding Trendlines and other data analysis for science

So I have a couple questions. So I have 2 columns, each containing numbers that correlate with each other that I want to graph. More specifically one column, say A, is pH and B is Volume Added. When I highlighted these and selected a scatter plot, it placed the pH on the x-axis and the Volume on the y-axis. This doesn't make any significant sense, since it needs to be the other way around (pH on the y-axis). I spent about an hour trying to figure out how to switch this, that is to directly specify which column is on which axis rather than simply having the first column be the x-axis. It seems as though this is trivial, and that it might just be better to arrange the columns manually, but the bigger issue is that there will be instances when two columns need to be graphed and they are not even next to each other. Seems like this is a pretty annoying issue with ease of use, since even if I did take those 2 said columns and copied them to be next to each other this is only cluttering and disorganizing the tables.

Another issue I have, which is much larger and could possibly make me not use numbers past the trial is chart and data analysis ability. Normally when you create a scatter plot you can select to add a Trendline, namely a line which seeks to match the pattern of your data, giving you an equation for this line. I can't figure out how to do this in numbers, and it is absolutely essential for almost every reason I use spreadsheets. Is this possible and I'm missing something? Or is numbers really not that useful to me?

Also how would I take this line, which generally represent the trend of my data, and find things such as the derivative or tangent equation of a range of data. Even still how can I use numbers to detect points of inflection or other significant mathematical occurances in graphs.

Thanks in advance

White MacBook 2.0ghz, Mac OS X (10.4.10), 1gb Ram, 80gb HD

Posted on Aug 9, 2007 8:01 PM

Reply
12 replies

Aug 12, 2007 5:45 AM in response to AndrewG7862

I had the same problem yesterday. I found out that when you hold the apple key, you can select any other row. You first have to select the X data though otherwise it will reverse the data.
So select the first column first, then select the second one using the apple key.

However, I haven't figure it out for trend lines or to change the type of graph in the same chart to display for instance a mean horizontal line or std along the X axis. When you choose a type of chart, all series seems to be of the same type. You can't have scattered points and lines for instance... Or I haven't figure it out yet..

Aug 12, 2007 7:51 AM in response to Mark Stockwell

I think its pretty safe to say that very little of iWork, is designed for scientific use. I have bought every version of iWork, because I get it at the educational rate, and have seen it slowly improve, but it still lacks a lot of basic functionality required by scientists. I thought I would be able to show some really interesting trends to my students using a table and dynamic graph, but the lack of a curve fit is a major problem.

Aug 16, 2007 7:00 PM in response to AndrewG7862

As a chemistry undergrad I use the trendline functions and corresponding equations for those lines on a regular basis using Excel. After getting iWork '08 and finding out that this feature was non-existent, I knew that this program was of no use to me. Its a shame that such a basic feature was not included. I would really appreciate an Excel analog that was made by Apple in order to do all of my data processing. Very sad.

Aug 24, 2007 1:59 PM in response to AndrewG7862

Sounds like you are trying to make a regression line -- a line which represents the correlation between x and y which you can use visually to determine how close your data approximates that prediction.

This can be done relatively easy in Numbers.

First make two columns of your data (X and Y).

Then, in the next column (or a separate table if you choose) use the functions INTERCEPT (y range, x range) for one cell and SLOPE (y range, x range) in the next cell. These 2 values will create the intercept and slope of your regression line (Y = slope * X + intercept).

Third, go back to your data and create two more columns adjacent to the first ones. The first additional column should be a copy of your "X" column. The second will be the "predicted Y values" based on the slope and intercept.
In each cell under the "predicted Y values" column, multiply the data in the "X" column times the value in the "SLOPE" cell and add the value in the "INTERCEPT" cell.

Finally, select all 4 columns of your data and choose scatter plot and two superimposed graphs should emerge -- one with your data and the other with the regression line.

Aug 24, 2007 3:39 PM in response to AndrewG7862

I'm also a scientist - I spend a great deal of time visualizing spectral data. Unfortunately, as pointed out above, Numbers is not suited to scientific applications at all. Trend lines, error bars, connecting lines, and other simple but important aspects of scientific data plotting are not available in Numbers. I haven't even been able to figure out how to change the symbol size in a line or scatter chart!

Although Excel has some of the features you want, I've always found Excel very unpleasant and painful to use for plotting scientific data. For plotting and simple curve fitting, I'll echo a previous poster and suggest DeltaGraph (I believe it's actually www.redrocksw.com). It is spreadsheet-based, offers a great deal of flexibility, such as saving plot templates (good if you plot the same things over and over), and it permits user-defined functions (so that if it doesn't have a curve fit you want, you can define it yourself). If you are a student or at an academic institution, it is only $199. I use it for creating all of my figures for journal articles.

Digression: IMHO, spreadsheet programs really are not the right kind of programs for scientific data analysis beyond the simplest curve fitting routines; there are many other programs (several free online) that are better suited to that task. I use a lot of command-line based analytical software for data analysis, but such software generally tends not to do the visualization tasks well. So I end up doing all my manipulation/analysis using the command line-based software, then when I'm done, output the result to a file that I can then import into a plotting program.

Lastly, I try to keep in mind that Numbers is a version 1.0 product; it's not going to be everything to everyone at the outset. But some polite feedback to Apple might get some of these features implemented in future versions (I'd be surprised if simple curve fitting isn't implemented in the next rev). For example, Keynote has improved by leaps and bounds since its introduction based on user feedback, and I expect the same will be true for Numbers.

Update: A demo version of DeltaGraph is available at the website above, as well as a tutorial on curve fitting. Although DG is my preference, I know Kaleidagraph also is a very capable program as well ( http://www.synergy.com) and also offers a free demo.

Message was edited by: sandrift

Sep 6, 2007 7:38 PM in response to Arthur Busbey

I also have Aabel, but it is pretty heavy-duty stuff, and I didn't find it very intuitive to use. And the Educational license is $345 ($445 for a regular license).

That said, I just noticed that on the Gigawiz site they are offering something called Citrin, which they call "Gem Quality Charting", and a license is $120. It looks like a pared-down version of Aabel, and might be worth a look.

This thread has been closed by the system or the community team. You may vote for any posts you find helpful, or search the Community for additional answers.

Adding Trendlines and other data analysis for science

Welcome to Apple Support Community
A forum where Apple customers help each other with their products. Get started with your Apple Account.