I suspect you need to use the GGRAPH version of the box-plot and add an ELEMENT line to plot the weighted mean.  One of the regulars on the SPSS mailing list is an expert on GGRAPH, and will probably know how to do this.  I suggest you post your question there (link given below).  HTH.

http://spssx-discussion.1045642.n5.nabble.com/

Figured that out quite some time ago, thinking it is best to give this an update. The following is the syntax that I used to solved that problem. This is used to plot box plot overlaid with dotted plot (with mean symbol indicated in red).

DATASET ACTIVATE DataSet1.

EXAMINE VARIABLES=Yourvariable BY Treatment

/PLOT NONE

/STATISTICS DESCRIPTIVES

/CINTERVAL 95

/MISSING LISTWISE

/NOTOTAL.

T-TEST GROUPS=Treatment(0 1)

/MISSING=ANALYSIS

/VARIABLES=Yourvariable

/CRITERIA=CI(.95).

GGRAPH

/GRAPHDATASET NAME="graphdataset" VARIABLES=Treatment Yourvariable MISSING=LISTWISE REPORTMISSING=NO

/GRAPHSPEC SOURCE=INLINE.

BEGIN GPL

SOURCE: s=userSource(id("graphdataset"))

DATA: Treatment=col(source(s), name("Treatment"), unit.category())

DATA: PYYBase_pM=col(source(s), name("Yourvariable"))

DATA: id=col(source(s), name("$CASENUM"), unit.category())

GUIDE: axis(dim(1), label("Treatment"))

GUIDE: axis(dim(2), label("Yourvariable"))

SCALE: cat(dim(1), include("0", "1"))

SCALE: linear(dim(2), include(0))

ELEMENT: point.dodge.symmetric(position(Treatment*Yourvariable))

ELEMENT: schema(position(bin.quantile.letter(Treatment*Yourvariable)), label(id))

ELEMENT: point.dodge.symmetric(position(summary.mean(Treatment*Yourvariable)), color(color.red))

END GPL.

The example of the result is attached.

Change "Yourvariable" and "Treatment" according to your variable name of course.

Hope it helps anybody who need this.

Dear Jia Jiet Lim,

This is so helpful! Thanks!

Kerstin

thank you so much.

if we need to relate the mean vlues by a line, which command do I add?

best regards