Posted: 5/21/2006 10:28:12 PM EDT
|
Ok, I have a Intro Statistics class and we are just starting out. I have no prior experience in stats. One of my assignements is to draw a cumulative frequency polygon. I have it down, BUT I don't know where to start the actual polygon. Do I start it 0 (origin), or at the very first class interval boundary? The textbook has not been much help. Looks like the Hive Mind can save my ass yet again. Thanks. |
|
That depends ... Do you want to show the stats on an absolute basis ("the big picture") or just relative to the data. You can get higher "resolution" (see more of your data) by starting at the lower boundary but be careful. I've seen waaay too many peole trying to make a mountain out of a mole hill because they lost sight of their data's relevance in the "big picture". |
The assignment doesn't say absolute or relative. The data starts at above 0, but below the first class boundary, though. 
|
|
I think you're referring to an ogive. What kind of data is it? If you're talking about, say, the probability distribution of people's lifetimes, it makes sense to start at 0. If, on the other hand, you're talking about something that doens't clearly have to be positive and start at 0, then I don't see why you'd use 0.\ I dunno... I guess I need to see the problem. |
|
Okay, but what did you do? Plot it in Excel. You can see it either or both ways. All you have to do is change the scale on the X axis to see it one way versus the other. I think the decision has more to do with skills/methods for presenting data than the statistics. |