Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: creating a bar chart with overlayed symbols


From   "Data Analytics Corp." <[email protected]>
To   [email protected]
Subject   Re: st: creating a bar chart with overlayed symbols
Date   Sat, 22 Jun 2013 22:28:29 -0400

Good morning,

Thanks for the help with this problem. But I have it solved. Actually, my solution was along the lines of your solution -- comforting to know that I had the right idea.

The chart is very simple. The bars are suppose to show the mean importance of a set of product attributes, but over all consumers in a survey. The symbols represent the mean importance rating of the same consumers, but by the segments they are classified in. This way, we can see which segments show more emphasis on each attribute, and which show less. This type of chart is used frequently when there are importance and performance ratings on each attribute, with the performance ratings by different competitors being the symbols. The whole diagram would show gaps (and opportunities) for any one competitor to dominant an attribute. The spider diagram is sometimes used, but it becomes difficult to dicipher when there are many attributes -- too confusing. Hope this helps.

Thanks,

Walt

________________________

Walter R. Paczkowski, Ph.D.
Data Analytics Corp.
44 Hamilton Lane
Plainsboro, NJ 08536
________________________
(V) 609-936-8999
(F) 609-936-3733
[email protected]
www.dataanalyticscorp.com
_____________________________________________________

On 6/21/2013 12:00 PM, Sergiy Radyakin wrote:
Walt, I still can't visualize exactly what you want to obtain. But
here is something to play with. Sergiy

sysuse auto, clear
keep if rep78>2
label define lrep78 3 "Feature A" 4 "Feature B" 5 "Feature C"
label values rep78 lrep78

separate price, by(foreign)
collapse (mean) price (mean) weight, by(rep78 foreign)
label variable price "Average price"
label variable weight "Importance of feature"
label variable foreign "Car origin"

twoway bar price foreign, by(rep78, rows(1)) fc(teal) lc(teal) ///
        yscale(r(0(1000)8000)) ///
        xlabel(0 "domestic" 1 "foreign") ||  ///
        scatter weight foreign if foreign==0, by(rep78) ms(T) mc(red) || ///
        scatter weight foreign if foreign==1, by(rep78) ms(S) mc(green) ///
        legend(rows(1))

On Fri, Jun 21, 2013 at 8:31 AM, Data Analytics Corp.
<[email protected]> wrote:
Hi,

Thanks for the suggestion.  I started to head in that direction. Now to do
it.


Thanks,

Walt

________________________

Walter R. Paczkowski, Ph.D.
Data Analytics Corp.
44 Hamilton Lane
Plainsboro, NJ 08536
________________________
(V) 609-936-8999
(F) 609-936-3733
[email protected]
www.dataanalyticscorp.com
_____________________________________________________


On 6/21/2013 4:57 AM, Nick Cox wrote:
You need to translate your problem to -twoway-, e.g.  -twoway bar-
Then overlaid symbols are just a -scatter- or -scatteri- on top.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index