Analysis for comparing Likert and checklist

[mizzougrad]

I have been trying to figure out which test to use for this situation. There are two questions on my survey where I ask the respondents to:
1. Select all the media types they consume (entertainment, sports, etc.) in a checklist format. So, the data input is either 1 for yes or blank for no for each type.
2. Rate how often they use several platforms (tablets, newspaper, etc.) in a Likert scale format. The data input is 0 for they don't use it or 1-7 for using least often to most often.

I want to perform an analysis to determine if the kinds of platforms people use have anything to do with the kinds of media they consume. Basically, for example, I want to know if people use tablets more often to consume sports media or use newspapers more often to consume entertainment media, etc. I did a crosstabs but it doesn't make much sense to me. Would anyone have any ideas?

Thanks!

[John F Hall]

Forget testing for the moment, just do some tabulations to look for patterns.  It would also help if you copy/pasted your crosstab syntax and table(s) and told us which version of SPSS you are using and how many cases you have.  If your questionnaire is in Word or pdf format, can you copy/paste the set of questions?

 

For now try running some syntax on your data.  From the Data Editor:

 

File > New > Syntax

 

Then type in the following [UNTESTED]

 

(modify to use the actual variable names in your file):

 

mult response groups

      mediaused 'Media used'(media1 to media10 (1))

      medianotused 'Media not used' (media1 to media10 (0))

/var platform1 to platform10 (0,7)

/freq mediaused medianotused

/tab mediaused by platform1 to platform10 /cel cou row.

 

This should produce summary tables listing, first all YES answers to questions on media use, second all NO answers, third a crosstabulation of all YES answers together with their respective usage distribution (with cell counts and row percentages).

 

Check out the Command Syntax Reference:

 

Help > Command Syntax Reference

 

. . and scroll down to MULT RESPONSE.  Look carefully at page 1287 as there is a keyword PAIRED you can use if your questions on usage frequency exactly match the questions on media usage.  If so the last line of the syntax could be:

 

/tab mediaused by platform1 to platform10 /cel cou row /paired.

 

If you need to see some examples, have a look at the tutorials on my page:

http://surveyresearch.weebly.com/33-multiple-response-mult-response.html  

 

Other SPSS gurus can better answer on testing, but you could always run:

 

means platform1 to platform5 by media1 to media10.

 

Also try:

 

crosstabs media1 to media10 by platform1 to platform5

/cel cou row.

 

If you need further assistance, I’m happy to look at your *.sav file if you send it to me off-list or via nabble.

 

John F Hall (Mr)

[retired academic survey researcher]

 

Email:     [hidden email]

Website: www.surveyresearch.weebly.com

 

 

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of mizzougrad
Sent: 03 April 2013 06:00
To: [hidden email]
Subject: Analysis for comparing Likert and checklist

 

I have been trying to figure out which test to use for this situation. There are two questions on my survey where I ask the respondents to:

1. Select all the media types they consume (entertainment, sports, etc.) in a checklist format. So, the data input is either 1 for yes or blank for no for each type.

2. Rate how often they use several platforms (tablets, newspaper, etc.) in a Likert scale format. The data input is 0 for they don't use it or 1-7 for using least often to most often.

 

I want to perform an analysis to determine if the kinds of platforms people use have anything to do with the kinds of media they consume. Basically, for example, I want to know if people use tablets more often to consume sports media or use newspapers more often to consume entertainment media, etc. I did a crosstabs but it doesn't make much sense to me. Would anyone have any ideas?

 

Thanks!

 

 

 

--

View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Analysis-for-comparing-Likert-and-checklist-tp5719245.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

[Bruce Weaver]

From the OP (emphasis added):
"1. Select all the media types they consume (entertainment, sports, etc.) in a checklist format. So, the data input is either 1 for yes or blank for no for each type."

If those blanks have ended up as system missing values in the data file, then they would have to be recoded from SYSMIS to 0 before John's syntax is tried.  E.g.,

RECODE media1 to media10 (sysmis=0) (else=copy).

HTH.

John F Hall wrote

Forget testing for the moment, just do some tabulations to look for
patterns.  It would also help if you copy/pasted your crosstab syntax and
table(s) and told us which version of SPSS you are using and how many cases
you have.  If your questionnaire is in Word or pdf format, can you
copy/paste the set of questions?

For now try running some syntax on your data.  From the Data Editor:

File > New > Syntax

Then type in the following [UNTESTED]

(modify to use the actual variable names in your file):

mult response groups
      mediaused 'Media used'(media1 to media10 (1))
      medianotused 'Media not used' (media1 to media10 (0))
/var platform1 to platform10 (0,7)
/freq mediaused medianotused
/tab mediaused by platform1 to platform10 /cel cou row.

This should produce summary tables listing, first all YES answers to
questions on media use, second all NO answers, third a crosstabulation of
all YES answers together with their respective usage distribution (with cell
counts and row percentages).

Check out the Command Syntax Reference:

Help > Command Syntax Reference

. . and scroll down to MULT RESPONSE.  Look carefully at page 1287 as there
is a keyword PAIRED you can use if your questions on usage frequency exactly
match the questions on media usage.  If so the last line of the syntax could
be:

/tab mediaused by platform1 to platform10 /cel cou row /paired.

If you need to see some examples, have a look at the tutorials on my page:
http://surveyresearch.weebly.com/33-multiple-response-mult-response.html

Other SPSS gurus can better answer on testing, but you could always run:

means platform1 to platform5 by media1 to media10.

Also try:

crosstabs media1 to media10 by platform1 to platform5
/cel cou row.

If you need further assistance, I'm happy to look at your *.sav file if you
send it to me off-list or via nabble.

John F Hall (Mr)
[retired academic survey researcher]

Email:      <mailto:[hidden email]> [hidden email]
Website:  <http://surveyresearch.weebly.com/> www.surveyresearch.weebly.com


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
mizzougrad
Sent: 03 April 2013 06:00
To: [hidden email]
Subject: Analysis for comparing Likert and checklist

I have been trying to figure out which test to use for this situation. There
are two questions on my survey where I ask the respondents to:
1. Select all the media types they consume (entertainment, sports, etc.) in
a checklist format. So, the data input is either 1 for yes or blank for no
for each type.
2. Rate how often they use several platforms (tablets, newspaper, etc.) in a
Likert scale format. The data input is 0 for they don't use it or 1-7 for
using least often to most often.

I want to perform an analysis to determine if the kinds of platforms people
use have anything to do with the kinds of media they consume. Basically, for
example, I want to know if people use tablets more often to consume sports
media or use newspapers more often to consume entertainment media, etc. I
did a crosstabs but it doesn't make much sense to me. Would anyone have any
ideas?

Thanks!



--
View this message in context:
<http://spssx-discussion.1045642.n5.nabble.com/Analysis-for-comparing-Likert
-and-checklist-tp5719245.html>
http://spssx-discussion.1045642.n5.nabble.com/Analysis-for-comparing-Likert-
and-checklist-tp5719245.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
<mailto:[hidden email]> [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

[mizzougrad]

Wow thank you that gave me exactly what I was looking for. I will let you know if I run into anything as I am testing this hypothesis!