SPSSX Discussion

Need Syntax for Combining Data from 3 Vars into One-Orig Vars were Contingency Questions

Classic

List

Threaded

3 messages Options

JanM1

Need Syntax for Combining Data from 3 Vars into One-Orig Vars were Contingency Questions

Using SPSS V 16 at the moment. Have situation in dissertation data that I
need to resolve before I can proceed. Situation follows.
Have VAR named ROLE (that people answering survey picked from the beginning,
which determined the questions they would see for the rest of the survey.
The ROLE responses were: Student (1) Faculty (2) Field Instructor (3)
Have VA named STRESPONSE (which represents responses to an opinion question
with YES/NO options (coded 1, 2) that only Students answered.
Have VAR named FARESPONSE (which represents responses to the same opinion
question with YES/NO options that only Faculty answered.
Have VAR named FIRESPONSE (which represents responses to the same opinion
question with YES/NO options that only Field Instructors answered.

I need to get the answers from all three of those variables into a single
variable - yet to be created - and named OPINION which will also have the
same YES/NO options.
That variable should end up with having the data from all 3 of the other
variables in it, and mapped to the role that each subject chose at the
beginning.
This is a situation that occurs throughout my survey so I will need to apply
the solution to multiple situations. THANK YOU THANK YOU THANK YOU in
advance!! I realize I created a headache for myself by having these
questions listed separately. I should have had one, that everyone answered.

--
Sent from: http://spssx-discussion.1045642.n5.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

Rich Ulrich

Re: Need Syntax for Combining Data from 3 Vars into One-Orig Vars were Contingency Questions

David Marso and I each answered this in the Usenet group, where your example

used Gender rather than Role.

His answer uses MAX rather than SUM.1 which I used --

COMPUTE OPINION= MAX(stresponse, firesponse, faresponse).

Rich Ulrich

From: SPSSX(r) Discussion <[hidden email]> on behalf of JanM1 <[hidden email]>
Sent: Saturday, February 3, 2018 3:14:35 PM
To: [hidden email]
Subject: Need Syntax for Combining Data from 3 Vars into One-Orig Vars were Contingency Questions

===================== 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

John F Hall

Re: Need Syntax for Combining Data from 3 Vars into One-Orig Vars were Contingency Questions

Here's an alternative solution using MULT RESPONSE

mult response groups

rolegroups (stresponse, firesponse, faresponse (1,2)).

Tested example based on national survey (Urban GB: dataset7) replicated simultaneously in two towns (Stoke-on-Trent:datataset2 and Sunderland: dataset3): new variables sexbg sexstoke and sexsund created from sex to mirror same problem.

add files file = dataset7 /file = dataset2 /file = dataset3

/keep sample sex sexgb sexstoke sexsund.

freq sample sex to sexsund.

sample
					Frequency			Percent			Valid Percent			Cumulative Percent
Valid	Urban GB				753			30.3			30.3			30.3
	Stoke				966			38.8			38.8			69.1
	Sunderland				770			30.9			30.9			100.0
	Total				2489			100.0			100.0
Sex of respondent
			Frequency			Percent			Valid Percent			Cumulative Percent
Valid	Men		1169			47.0			47.0			47.0
	Women		1320			53.0			53.0			100.0
	Total		2489			100.0			100.0
sexgb
				Frequency			Percent			Valid Percent			Cumulative Percent
Valid		Men		442			17.8			45.8			45.8
		Women		524			21.1			54.2			100.0
		Total		966			38.8			100.0
Missing		System		1523			61.2
Total				2489			100.0
sexstoke
				Frequency			Percent			Valid Percent			Cumulative Percent
Valid		Men		389			15.6			51.7			51.7
		Women		364			14.6			48.3			100.0
		Total		753			30.3			100.0
Missing		System		1736			69.7
Total				2489			100.0
sexsund
				Frequency			Percent			Valid Percent			Cumulative Percent
Valid		Men		338			13.6			43.9			43.9
		Women		432			17.4			56.1			100.0
		Total		770			30.9			100.0
Missing		System		1719			69.1
Total				2489			100.0

mult response groups = sexgroups (sexgb sexstoke sexsund (1,2))

/freq sexgroups.

sexgroups Frequencies
		Responses		Percent of Cases
		N	Percent
sexgroups^a	Men	1169	47.0%	47.0%
	Women	1320	53.0%	53.0%
Total		2489	100.0%	100.0%
a. Group

John F Hall MA (Cantab) Dip Ed (Dunelm)

[Retired academic survey researcher]

Email: [hidden email]

Website: Journeys in Survey Research

Course: Survey Analysis Workshop (SPSS)

Research: Subjective Social Indicators (Quality of Life)

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Rich Ulrich
Sent: 03 February 2018 22:49
To: [hidden email]
Subject: Re: Need Syntax for Combining Data from 3 Vars into One-Orig Vars were Contingency Questions

David Marso and I each answered this in the Usenet group, where your example

used Gender rather than Role.

His answer uses MAX rather than SUM.1 which I used --

COMPUTE OPINION= MAX

Rich Ulrich