Two Step Cluster Analysis

[waqasjeral]

hi

I have a data files that contain not more than 12 columns but around 3000+ entries in each column. The first four are names of locations while next eight column contain numeric data for entry and exit. When I perform Two Step Cluster Analysis in SPSS with first four variables as categorical and rest variables as numeric, I get only one cluster with all 100% values and there is no good interpretation that I can make.

My supervisor told me to change few string variables to numeric one, but how and what variables while only the name of locations is shown as string variable.

Please help.

[Art Kendall]

look up AUTORECODE. It changes string variables to numeric variables.   the /GROUP allows you to apply the same recoding scheme to columns that contain the same kinds of strings.  As a side benefit the highest numeric value assigned is the number of unique values.

If the string variables were entered by people keying data, AUTORECODE sorts the strings alphabetically so if you use the /PRINT option you can spot variations BWI BWi bWI  BIW will be near each other in the list of values.

IF capitalization or lack of it is an error you can transform the data to uppercase first.

Without you providing  more context and perhaps posting a small example of your data it is hard to say much else.

On 3/15/2013 7:56 AM, waqasjeral wrote:

hi

I have a data files that contain not more than 12 columns but around 3000+
entries in each column. The first four are names of locations while next
eight column contain numeric data for entry and exit. When I perform Two
Step Cluster Analysis in SPSS with first four variables as categorical and
rest variables as numeric, I get only one cluster with all 100% values and
there is no good interpretation that I can make.

My supervisor told me to change few string variables to numeric one, but how
and what variables while only the name of locations is shown as string
variable.

Please help.



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Two-Step-Cluster-Analysis-tp5718704.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