[EMAIL PROTECTED] wrote:
> John,
>
> I stand corrected. I really guess I misunderstood what Mark wanted.
>
> I now realize he is an experienced programmer.
Well I have no idea if he is or not. I suspect he may have underestimated the
difficulty of programming this in Basic for a Palm Pilot. Good luck Mark.!
I can remember doing a repeated measures ANOVA in FORTRAN on a main frame as a
class project ( It finally worked but totally exhausted the class computing
budget :) Prof was annoyed, I was amused. Don't give a project unless you
understand what you are really doing :) We are talking a long time ago.
Possibly even pre-SAS!
> I'm sure that the correct
> BASIC routines can be found in a number of books on programming. It seems to
> me I remember that Apple produced a book quite some time ago and it had in it
> the method of determining the probability of a Z value.
Probably all Mark wanted.
>
> I looked back at some programming I did quite a while back and came up with
> the following code. I hop that Mark can use it. I know it works in my
> software but it is only one of about six or seven things the user might want
> to know when dealing with Z values.
Err, are you talking about the distribution or the implications. Two different
things I believe. The implications are more important , so to speak, in
decision making but if all Mark wants is the z-distribution that is another
matter. It is up to him to know how to use it.
> Depending on the actual basic language that is used (GWBASIC, QUICKBasic,
> TurboBasic) this routine should find the probability of a given Z value to at
> least the fourth decimal place.
I think that this is exactly what he wants. I get the feeling he does not want
the table although as you sat below he could scan it into a Word document and
upload it. His question was about actually calculating the value. I personally
do not really see the value of the exact figure at least at the undergrad
level ( or the grad level in most psych studies) but that is quite another
question.
>
> Of course Mark could scan a copy of the Z table and then put it into a Word
> document and load that onto the PalmPilot. That way he could quickly refer
> to the table at any time.
Exactly what he did not want although quite possibly more immediately useful.
Copyright may apply?
>
> I assume Mark can write the code for entering the values.
>
> Here is the code, I hope it helps.
>
> ZOnly:
> Mean=0
> St.Dev=1
> PROMPT$=" Enter the Z value to compare to the MEAN "
> X$="C"
> PROMPT$=PROMPT$+PAD$ : PROMPT$=LEFT$(PROMPT$,57)
> print PROMPT$;
> input " ",X$
> if X$="" OR X$=" " then X$="0"
> IF X$="-" OR ASC(X$)<45 OR ASC(X$)>57 THEN
> beep
> print
> print " Improper entry. Please try again."
> print
> goto GeneralMenu
> END IF
> XX=VAL(X$)
> Y=XX : Y=ABS(Y)
>
> IF Y=0 THEN
> beep : print
> print " Your entered value must not be zero. Please try again."
> print
> input " Depress <Enter> to continue ",rck$
> goto GeneralMenu:
> END IF
>
> Y=ABS((Y-Mean)/St.Dev)
> StandDevAway=Y
> R=EXP(-(Y^2)/2)/2.5066282746
> goto DoCalc
>
> ZRawData:
> print
> PROMPT$=" Enter the POPULATION MEAN value "
> X$="C"
> PROMPT$=PROMPT$+PAD$ : PROMPT$=LEFT$(PROMPT$,57)
> print PROMPT$;
> input " ",X$
> if X$="" OR X$=" " then X$="0"
> IF X$="-" OR ASC(X$)<45 OR ASC(X$)>57 THEN
> beep
> print
> print " Improper entry. Please try again."
> print
> goto GeneralMenu
> END IF
> XX=VAL(X$)
> Mean=XX
>
> St.Dev=-1
> WHILE St.Dev<=0
> PROMPT$=" Enter the POPULATION STANDARD DEVIATION "
> X$="C"
> PROMPT$=PROMPT$+PAD$ : PROMPT$=LEFT$(PROMPT$,57)
> print PROMPT$;
> input " ",X$
> if VAL(X$)=-999 then goto OnToMenu
> if X$="" or X$=" " then X$="0"
> if X$="-" then X$="0"
> if ASC(X$)<45 or ASC(X$)>57 then X$="0"
> XX=VAL(X$)
> IF XX<=0 THEN
> beep
> print
> print " The Standard Deviation must be a positive number."
> input " Please try again. Depress <Enter> to continue ",rck$
> END IF
> St.Dev=XX
> WEND
> AskStxx:
> St.Dev=St.Dev/(SQR(SSize))
> print
> print " Remember if you are comparing a value to the Mean you may want"
> print " to enter lower class limit of that value. For instance, if you"
> print " are interested in a value of 13 you might want to enter 12.5."
> print
> PROMPT$=" Enter the value to compare to the MEAN "
> X$="C"
> PROMPT$=PROMPT$+PAD$ : PROMPT$=LEFT$(PROMPT$,57)
> print PROMPT$;
> input " ",X$
> if VAL(X$)=-999 then goto OnToMenu
> if X$="" or X$=" " then X$=STR$(Mean)
> if ASC(X$)<45 or ASC(X$)>57 then X$=STR$(Mean)
> XX=VAL(X$)
> Y=XX ': Y=ABS(Y)
> IF Y=Mean THEN
> beep : print
> print " Your entered a value that is equal to the Mean. This results"
> print " in a Z value of 0.
> goto DoneShowing
> END IF
> Interest=Y
> Y=ABS((Y-Mean)/St.Dev)
> Top=Interest-Mean
> StandDevAway=Y
> R=EXP(-(StandDevAway^2)/2)/2.5066282746
>
> DoCalc:
> cls : locate 4,1
>
> if Top<0 then
> StandDevAway=StandDevAway*(-1)
> if SSize=1 then
> kk=abs(Top)
> print " The score entered is ";
> print using fnform$(kk);abs(Top);
> print " less than the Mean."
> end if
> end if
>
> print " For a Z value of "; : print using FORM14$;StandDevAway
> print
> print " (y Ordinate at Z) FREQUENCY = ";
> print using FORM14$; R
> Z=Y
> Y=1/(1+.33267*ABS(Y))
> T=1-R*(.4361836*Y-.1201676*Y^2+.937298*Y^3)
> if Z<0 then T=1-T
> print
> print " (Area in Larger Portion) PROBABILITY = ";
>
> print using FORM14$; T
> print " (Area in Smaller Portion one-tail = ";
> print using FORM14$; 1-T
> print " (Area from Z to Mean one-tail = ";
> print using FORM14$; .5-(1-T)
>
> print " (Area in Smaller Portions two-tail = ";
> print using FORM14$; (1-T)*2
> print
>
> DoneShowing:
Needs cleaning up for today ?? :) Mayhap a few to many Print's ? :)
>
> Seems as though this is like re-inventing the wheel and I'm not sure if he
> helps to understand Z values or probability.
Meaningless in term of the question. He just wanted to program it. It may or
may not be of any use in understanding Z-values or probability. My guess it it
probably is since he has to have an understanding of what he is programming
unless he is mindlessly typing and I doubt that even given the code you have
provided.
And yes re-inventing the wheel. Nothing wrong with that. Shows he's interested
( or lazy ?) either of which leads to advances in the area. I have heard it
suggested that all innovation came from people to lazy to do things 'this' way.
Any way, thanks for the reply .
------------------
John Kane
The Rideau Lakes, Ontario Canada
PS. May be those westerners are not as bad as we have heard.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
