Error in PROC ANOVA

[m_priya]

IM A BEGINNER IN SAS .IN MY PROJECT I NEED TO PERFORM ANALYSIS TO  MY DEPENDENT AND INDEPENDENT VARIABLES TO CHECK FOR HYPOTHESES.

MY DEPENDENT VARIABLE  :   ORDER - CHARACTER VAR

    INDEPENDENT VARIABLES : CITY ,SOURCE-CHARACTER VAR

                                                      PRICE ,QUANTITY,SALES AMT- NUMERIC VARIABLE.

TOTALLY I HAVE SIX VARIABLES. I CONVERTED ALL MY NUMERIC VAR TO CHARACTER VAR AND PERFORMED THE CHISQ TEST.I GOT WARNING MESSAGE AS:(SOME%) OF CELLS HAVE EXPECTED COUNTS LESS THAN 5.CHISQ MAY NOT BE A VALID TEST.

AGAIN I TRIED FOR GLM /ANNOVA MULTIVARIATE TEST.GOT ERROR AS

ERROR: Variable Order  in list does not match type prescribed for this list.

EVEN IF I CHANGE THE POSITION OF DEPENDENT AND INDEPENDENT VAR I GET THE SAME ERROR.

HERE IS MY SCRIPT:

PROC ANOVA;
CLASS SOURCE;
MODEL   ORDER = PRICE QUANTITY SALES AMOUNT CITY SOURCE ;
MANOVA H=_ALL_;
RUN;

COULD SOMEBODY HELP ME WITH WHAT KIND OF  ERROR IS THIS /ANY OTHER WAY TO PERFORM ANALYSIS FOR ALL THESE VARIABLE ALTOGETHER.

THANKS IN ADVANCE.

[ballardw · Accepted solution]

First thing, when you have an error it helps to post the code and error by copying from the log and pasting into a code box opened with the {i} icon in the forum. Error messages often have indicators that help determine where the error occured but the main forum message box reformats the text making the error diagnostics less useful.

This error almost always means that you are using a character variable where the procedure expects a numeric.

ERROR: Variable Order  in list does not match type prescribed for this list.

The Anova procedure requires a numeric for the dependent variable. Character variables do not have any "variance" to analyze in this context

[m_priya]

hi,

yup i understood.and also corrected the error. thank you

[Reeza]

WRITING IN ALL CAPS is yelling, personally I can't read your post. Can you please try and repost your question in a proper format. Use the { i } icon to embed your code in the post. 

[Ksharp]

Error info said your ORDER is character type variable, not numeric , therefore you can't use ANOVA.

[m_priya]

hi,

thakyou.i understood and corrected .