DataGraph Forums › Technical Support › Support Desk › Creating a graph suitable for publication in a scientific journal
Every graph I have ever created has multiple y values for each x, and it is necessary to plot the mean of the y values associated with each x plus error bars showing either the standard error or the standard deviation. I do not see how to do this with DataGraph. Is DataGraph suitable for this task?
This is definitely possible in DataGraph. If you care to provide some example data or an example chart that shows exactly what you’re after, I’m sure people here can give you more specific help.
On the plot there are 3 y values for each x, and the lines are means and the error bars are the standard error of the mean. The graph is from a Prism file, and I am searching for a less expensive alternative, at least for the graphing. I don’t see how to upload a sample data table.
This is no problem for DataGraph. You just need one column containing the x-axis, and columns with the y values for each trace, and the error for each trace. After plotting, you select the column you want to use for the error bars. The user manual provides explicit instructions on these features.
Here is an example from my own collection (the colors and styles of the error bars are entirely user-selectable):
Ms. Hamilton, do you have a page number? Do I use the Pivot command? How do I get the sem across each row? Do I need to enter the formula? This is incomprehensible to me.
If you need to calculate the mean and SEM, the Pivot command can do that for you.
Then you can extract the calculated values back to the data table to graph the points/error bars using the Points command.
We will post a sample file later today.
Here is an example that I think shows you what you would like to see:
Here is an easy way to accomplish such a plot. It helps if you already have your data formatted like this in columns, and all calculations completed. Here is a screen shot of the definitions table:
And here is representative screenshot of the data table representing the columns shown above:
Making the plots is also straight forward using the plot command as the screen shot shows:
Just replicate the settings for each of the three lines, making sure to adjust the Y value and the error in Y for each one. This should provide you all the information you need to get your plot ready. Most things you can do in Prism you can do in DataGraph for a much better price.
The price difference is why I am considering DataGraph. I still do not see how to geet DataGraph to calculate SD or SEM. Jordan, did you use DG or something else to get the mean and SD?
My data usually comes from an Excel spreadsheet that has the calculations built in, but you can definitely do the calculations in DataGraph. This is how I would do it, but it certainly is not the best way.
I anticipate that someone from the dgteam will post a better version of what you see here. I anticipate that you can extract the averages and SD out a pivot plot, but I don’t have time to mess with it right now. I will work it tonight if there is not a reply from Pamela or David. It is important to remember that DataGraph may look like a spreadsheet program, it is not a spreadsheet. It straddles the line between database and spreadsheet.
Jordan, your reply posted this morning while I was still crafting a response/example, so thank you for picking this up.
DataGraph Forums › Technical Support › Support Desk › Creating a graph suitable for publication in a scientific journal