Forum Replies Created
-
Posts
-
Yes, you’ve encountered a recently introduced bug. 😞 Fortunately, it is fixed in the next release which, if it passes QA, will come out by end of day April 17.
Thanks for reporting it, and sorry for the inconvenience.
Bill
I agree that it might improve quality of life, at least for some. It could be a dialog box that says something like “Open with auto save enabled?” The downside would be that causes the user to have to make a (possibly ill-understood) decision before they get to their document.
But it’s a good suggestion and I’ll add it to our list of feature requests. Thank you!
I’m very glad you see CODAP as a “wonderful tool.”
Bill
Thank you so much for bringing this to our attention. This seems to be a bug that has crept in rather recently. I’ll put it at the top of the list!
Bill
Hi Kathy,
Thanks for this question. It comes up fairly often.
John Tukey, father of exploratory data analysis, invented box plots in 1970, long before computers with graphical user interfaces were available, let alone commonplace. Datasets were often small enough to be dealt with by hand. Tukey wanted graphical representations that could easily be created and understood with just paper and pencil. Thus his method for constructing a box plot didn’t require any computation, facilitated by the “removal” of the median value in a distribution when finding the “Q2 or Q4” values as the median of what remains.
But you probably knew all that! 😉
Tukey’s method comes up with an approximation to the 25th or 75th percentile of the distribution. CODAP (and Fathom) instead calculate the actual 25th or 75th percentile for their box plots. (See Wikipedia for computational methods.) This seems appropriate in an age when doing things by hand is rarely called for. And, of course, with any reasonably sized dataset, the two methods yield nearly, if not exactly, the same result.
As an aside, I don’t find box plots useful for characterizing a single distribution. They come into their own when comparing two or, especially, many distributions.
Bill
Ah, you have identified a previously unknown bug!Kudos!
I’ll log it an get it on a fast track to be fixed.
Se me ocurren dos formas de calcular los valores de los quintiles de una distribución: una usando fórmulas en una tabla de casos; y un segundo en el que los valores móviles se ajustan manualmente para estar cerca de los valores deseados. Ambos se ilustran en el documento CODAP vinculado a continuación.
Lo siento, pero no estoy seguro de entender tu pregunta. Aquí hay algunas cosas para probar:
1. Muestre un diagrama de caja en el gráfico. Esto le dará acceso a los valores para los percentiles 25, 50 y 75 que puede obtener al pasar el mouse sobre partes del diagrama de caja.
2. Use la habilidad “Plot Value” para trazar, por ejemplo, percentile (edad, 33).
Espero que sea de ayuda. Si no, intente ser más específico sobre lo que le gustaría lograr.
(Disculpas por mi español. Estoy usando Google Translate.)
There are no built-in regression models in CODAP other than linear (yet). This is a consequence of no funded collaborating projects having yet requested such a feature.
Bill
Hello Daichi,
Let’s say you have a plot with x, y, and/or legend attributes. Now you can split this plot by the categories of a categorical attribute by dropping that attribute on the top axis area or the right axis area.
Dropping the categorical attribute on the top axis splits the plot horizontally so that you have the sub-plots laid out left to right. Dropping the categorical attribute on the right axis lays them out top to bottom.
You can actually drop categorical attributes on both the top and bottom!
I’ve attached a screenshot of these three possibilities.
Hope this helps,
Bill
Attachments:
Hi Kathy,
Thanks for this! You and your students found a bug in the abs function. The difference of the missing value and the mean is correctly evaluated as a missing value, but abs incorrectly reports zero instead of missing. Please convey our thanks to your class!
The bug fix will probably appear in less than two weeks. I’m sorry your teaching moment will be taken away. 😉
Bill
Hi Andy,
First of all, I assume you have set the speed slider in the plug-in to “Fastest?”
If so, then I think the best we can do is to close the case table and/or any plots. (I don’t think making them smaller will help.) Of course you may have reasons for not wanting to do this.
If you like, provide a shared link to the problem situation and we’ll take a look at it.
Bill
Hi Joachim,
Yes, it is possible to stack the data. You use a handy little plugin that Jonathan Sandoe wrote called the Attribute Stacker which you can find on the CODAP Data Interactives page. I’ve made a little movie showing how to do it. (In it I have way fewer years, of course.)
Here is the movie. I hope it helps!
Bill
