The Graphs randomly deletes some cases

[Ana Budilivski]

Dear SPSS and Statistics gurus,

I had to make a clustered bar chart that would show me how many people fall under the Category A for the 1st, 2nd and 3rd Trimester, how many fall in Category B for the 3 Trimesters and so on.

In order to be able to write down the exact number of cases for each Category for each Trimester, I made a Crosstabs analysis (Here I have a different question: Do I really have to write down the number of cases by hand, or is there a possibility for it to be done automatically?) which showed me how many people fall under each Category (Picture 1).

Then, for reasons that I cannot explain, I decided to check these data again (and I hate myself for that, because God protects the ones that know nothing!). I decided to split the file according to the 3 Trimesters and see how many cases fall under each Category separately for each Trimester.
This showed me that there are actually more cases in my data than the Graphs and the crosstabs showed me (Picture 1 and 2). So the Bar chart and the crosstab analysis did not include some cases in some of the Categories.
Could you please tell me why  SPSS does not include all  the cases in the bar chart and crosstabs? Or at least tell me what other books should I read in order to stop asking these kind of questions anymore.

I kindly thank you all for your future help and patience.

Sunny regards from Switzerland!

Ana

[John F Hall]

Ana

The extra cases are probably missing for one or both variables.  You can
check by asking for a frequency count of the variables of interest.  This
will give you a summary table of valid and user-defined or system-missing
cases for each variable.

John F Hall (Mr)
[Retired academic survey researcher]

Email:   [hidden email]  
Website: www.surveyresearch.weebly.com  
SPSS start page:  www.surveyresearch.weebly.com/1-survey-analysis-workshop





-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ana
Budilivski
Sent: 19 October 2015 15:20
To: [hidden email]
Subject: The Graphs randomly deletes some cases

Dear SPSS and Statistics gurus,

I had to make a clustered bar chart that would show me how many people fall
under the Category A for the 1st, 2nd and 3rd Trimester, how many fall in
Category B for the 3 Trimesters and so on.

In order to be able to write down the exact number of cases for each
Category for each Trimester, I made a Crosstabs analysis (Here I have a
different question: Do I really have to write down the number of cases by
hand, or is there a possibility for it to be done automatically?) which
showed me how many people fall under each Category (Picture 1).

Then, for reasons that I cannot explain, I decided to check these data again
(and I hate myself for that, because God protects the ones that know
nothing!). I decided to split the file according to the 3 Trimesters and see
how many cases fall under each Category separately for each Trimester.
This showed me that there are actually more cases in my data than the Graphs
and the crosstabs showed me (Picture 1 and 2). So the Bar chart and the
crosstab analysis did not include some cases in some of the Categories.
Could you please tell me why  SPSS does not include *all * the cases in the
bar chart and crosstabs? Or at least tell me what other books should I read
in order to stop asking these kind of questions anymore.

I kindly thank you all for your future help and patience.

Sunny regards from Switzerland!

Ana

<http://spssx-discussion.1045642.n5.nabble.com/file/n5730782/Picture1.png>
<http://spssx-discussion.1045642.n5.nabble.com/file/n5730782/Picture_2.png>



--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/The-Graphs-randomly-deletes-so
me-cases-tp5730782.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Ana Budilivski]

Dear Mr. Hall

Indeed! After I performed the frequency count, the system told me than I have 100 valid answers instead of 98 that was used by the bar chart.
So these two variables are not missing, but for some unknown for me reason is not included in te bar chart graph.

Could there be some explanation for this?

Best regards,

Ana

[Bruce Weaver]

Please describe the layout of your data file (e.g., whether it is a wide or long file, and the relevant variable names), and provide the syntax used to create the graph.  (If you do not use syntax regularly, go through the steps you used in the GUI, and exit via the PASTE button to generate the syntax.)  

HTH.

Ana Budilivski wrote

Dear Mr. Hall

Indeed! After I performed the frequency count, the system told me than I have 100 valid answers instead of 98 that was used by the bar chart.
So these two variables are not missing, but for some unknown for me reason is not included in te bar chart graph.

Could there be some explanation for this?

Best regards,

Ana

[Ana Budilivski]

So the syntax for the Bar chart graph would be (Where ALL_Cathegory includes each cathegory in one single variable):

DATASET ACTIVATE DataSet1.
* Chart Builder.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=ALL_Cathegory COUNT()[name="COUNT"] Year_2014
    MISSING=LISTWISE REPORTMISSING=NO
  /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: ALL_Cathegory=col(source(s), name("ALL_Cathegory"), unit.category())
  DATA: COUNT=col(source(s), name("COUNT"))
  DATA: Year_2014=col(source(s), name("Year_2014"), unit.category())
  COORD: rect(dim(1,2), cluster(3,0))
  GUIDE: axis(dim(3), label("ALL_Cathegory"))
  GUIDE: axis(dim(2), label("Count"))
  GUIDE: legend(aesthetic(aesthetic.color.interior), label("Year_2014"))
  SCALE: cat(dim(3), include("1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12"))
  SCALE: linear(dim(2), include(0))
  SCALE: cat(aesthetic(aesthetic.color.interior), include("1", "2", "3"))
  SCALE: cat(dim(1), include("1", "2", "3"))
  ELEMENT: interval(position(Year_2014*COUNT*ALL_Cathegory), color.interior(Year_2014),
    shape.interior(shape.square))
END GPL.

Syntax for the crosstab (very simple.. I only needed the count):

CROSSTABS
  /TABLES=Year_2014 BY ALL_Cathegory
  /FORMAT=AVALUE TABLES
  /CELLS=COUNT
  /COUNT ASIS.

Then I would Split the file

SORT CASES  BY Year_2014.
SPLIT FILE SEPARATE BY Year_2014.


.. and make a frequency analysis for each cathegory separately (As oposed to ALL_Cathegory above)

FREQUENCIES VARIABLES=Cat1 Cat2 Cat3 Cat4 Cat5 Cat6
    Cat7 Cat8 Cat9 Cat10 Cat11 Cat12
  /ORDER=ANALYSIS.

Thank you for your effort and help!