|
|
Hi all,
The following steps have been used to undertake q-sort analysis via hierarchical analysis. If I have missed anything could someone please let me know, it would be much appreciated.
Step 1
Generate a co-occurence matrix with tallys out of 60 (60 respondents combined into a single matrix with co-occurrence of tallys- item 1 with item 2 ETC)
Step 2
Generate a proximity matrix to be entered into the cluster analysis (co-occurrence matrix needs to be pre-processed rather than using items directly in cluster analysis)
Matrix Set-Up
RowType_ VarName_
U1 U2 U3
U1 0 45 37
U2 45 0 28
Syntax
PROXIMITIES
U1 U2 U3 U4 U5 U6 U7 U8 U9 U10 U11 V1 V2 V3 V4 V5 V6 V7 V8 V9 V10
V11 V12 V13 V14 EG1 EG2 EG3 EG4 EG5 EG6 EG7 EG8 EG9 EG10 C1 C2 C3 C4 C5 C6
C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 EV1 EV2 EV3 EV4 EV5 EV6 EV7 EV8 EV9
EV10
/MATRIX OUT (file name)
/VIEW=VARIABLE
/MEASURE=CHISQ (because of count data)
/PRINT PROXIMITIES
/STANDARDIZE=NONE .
Step 3 Use the proximity matrix as data input into the hierarchical cluster analysis
CLUSTER
/MATRIX IN (file name)
/METHOD BAVERAGE
/PRINT SCHEDULE CLUSTER(3,6)
/PLOTS DENDROGRAM.
Thanks in advance for any corrections
Regards Paul
|
Locked
