![]() I was able to show scatter plots of data and we then thought of various ways to label each axis. I know that the plots were not created with data, but the trends were there and we could see patterns of how to analyze scatter plots. It was great! When the students took off their blindfolds we had a bunch of scatter plots to observe and analyze. The key was that these trends needed to be made up of a series of dots – a scatter plot. Other students could prompt their partner to try to make a group of dots in a line going up, or a line that started high and the dots traveled down. Some guiding students told the blinded student to put dots all over the place, completely random. The other student could direct them in case they went off the paper, but the idea was to place a bunch of dots on the graph. This student had to place dots on the graph. Then I told one of the students to uncap a marker and put on a blindfold. ![]() Most relevant data for people occur in this quadrant, and it provides a nice large area to create the plot. So I gave my students a full sized piece of graph paper, told them to work in pairs and had them draw the x and y axes to show quadrant 1. So one time I figured I would try something new something to do BEFORE looking at the different types of trends something the opposite of visual, something blind. Students can really start talking about what the data means when they realize that some things are associated, and others are more random. Once the data in the cells is dates with times, you can apply what I have written above.When it comes to analyzing scatter plots, I like to introduce scatter plots visually at first. You may want to convert your text to real date/times with the Text to Columns tool or another technique of your choice. Formatted as times, all integers will show up as midnight. Excel cannot recognize the text in column A as dates, so it just numbers the data points from first to last and plots these numbers on the X axis. Format the X axis labels to "General" and you may find that the labels are integer numbers. If you change your mind, edit your question and post a comment to let me know what you want.Įdit: Looks like your "dates" may actually be text. Not pretty, but that's what you asked for. That will cause all the date/time stamps to overlap and become unreadable, so you may want to revise that decision or make your chart very, very wide if you have more than a few hours of data or turn the X axis labels sideways which will look like this: make the major interval on the horizontal axis 1/12 of a day.Ī day is 1 in Excel, so the major interval needs to be 1 divided by 12, which is 0.08333333.If you want the time and date printed every two hours, you will need to Rows go left to right, columns go top to bottom) (By the way: The row with the time is not a row, but a column. It looks like the format for the axis labels is using only the time without the date. You can clearly see on the X axis that the markers are for 0:00 AM and these are consecutive days.įormat the X axis and you can see the settings for the Axis Options > Units will be 1 (for 1 day). As a last resort, you could reshape the timestamps in your logfile to some format that Excel can handle.Įxcel does not plot the time randomly. I'm guessing that your logfile is written by a device with a 'locale' that differs from the Internationalization settings on your machine, which makes it harder for Excel to properly recognize the timestamps. That option allows you to further specify the DMY order. "General" is the default, which does a fine job of recognizing dates, but you may need to nudge the process in the right direction by explicitly choosing "Date" type. Note how the Text Import Wizard offers the feature to customize each of the imported columns for data type. So your actual problem is getting the proper timestamps imported from some logfile. Based on no input, Excel has made up a series of consecutive dates starting with the awesome Zeroeth of January, 1900 ! I got exactly the same 7 labels "12:00:00 AM" like you had, but the ultimate indication of what is going on, is to put the axis labels in date format. I had to seriously tweak a CSV import to get this wrong on purpose - Excel seems to be getting frighteningly smart at recognizing import data. As noted by the error stands out because the text data are left-justified. I reproduced your problem by intentionally entering text instead of time data in the first column. When entering proper timestamps in Excel, the scatterplot comes out right immediately (even though the axis labels need some tailoring to taste).
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |