Compute Variable

[MBAstudent]

Hi All,

It's the first time I'm using SPSS and I could use some help.

I'm measuring the effectiveness of transformational leadership style using 20 items, which are subdivided into 5 different behaviours. The effectiveness is measured using 9 items, which are subdivided into 3 outcomes. (USING THE instrument MLQ5X )

In order for me to apply a regression analysis, I need to first create a new variable for each effectiveness outcome and then create the dependant variable (9 items measuring effectiveness- all 3 outcomes). And to bring together the items measuring each leadership behaviour in order to create independent variables (which should create 5 variables).

When trying to create an independent variable for example.
 I have tried to do so in a few ways:

1) independent variable
COMPUTE idealized_influenc_a=SUM(IIA1,IIA2,IIA3,IIA4).
EXECUTE

2)COMPUTE idealized_influenc_a2=IIA1 + IIA2 + IIA3 + IIA4.
EXECUTE

However, as I have missing values (which are defined ) I get a . in part of them, which holds me back with my analysis.

What is it that i need to do in order to create independent and dependent variables from my research items ?

Thank you all so VERY MUCH

[Art Kendall]

functions such as SUM or MEAN, etc. can have a suffix which says the minimum number of variables in the list that must be valid to allow the calculation, e.g.,

COMPUTE idealized_influenc_a=mean.3(IIA1,IIA2,IIA3,IIA4).
would say that if there are 3 valid values calculate the mean other wise return a missing value.

A score using the MEAN function is on  the same scale as the response scale of the items. 
It does not need to be corrected.
It correlates with other variables the same as a  SUM score corrected for the number of valid items.

-----
Not always, but very commonly missing items within scales are due to poor administration of the test.

-----
You do not need EXECUTE after every compute.

On 12/14/2013 12:29 PM, MBAstudent [via SPSSX Discussion] wrote:

Hi All,

It's the first time I'm using SPSS and I could use some help.

I'm measuring the effectiveness of transformational leadership style using 20 items, which are subdivided into 5 different behaviours. The effectiveness is measured using 9 items, which are subdivided into 3 outcomes. (USING THE instrument MLQ5X )

In order for me to apply a regression analysis, I need to first create a new variable for each effectiveness outcome and then create the dependant variable (9 items measuring effectiveness- all 3 outcomes). And to bring together the items measuring each leadership behaviour in order to create independent variables (which should create 5 variables).

When trying to create an independent variable for example.
 I have tried to do so in a few ways:

1) independent variable
COMPUTE idealized_influenc_a=SUM(IIA1,IIA2,IIA3,IIA4).
EXECUTE

2)COMPUTE idealized_influenc_a2=IIA1 + IIA2 + IIA3 + IIA4.
EXECUTE

However, as I have missing values (which are defined ) I get a . in part of them, which holds me back with my analysis.

What is it that i need to do in order to create independent and dependent variables from my research items ?

Thank you all so VERY MUCH

---

If you reply to this email, your message will be added to the discussion below:

http://spssx-discussion.1045642.n5.nabble.com/Compute-Variable-tp5723637.html

To start a new topic under SPSSX Discussion, email [hidden email]
To unsubscribe from SPSSX Discussion, click here.
NAML

[John F Hall]

Art is right, but I would add:

 

Why are you using such long variable names?  What’s wrong with a1 (much less typing)?  You can put longer descriptions in the Labels column.

 

Compute a1  = sum.4 (IIA1,IIA2,IIA3,IIA4)

Var lab a1 ‘Idealised influence’

Freq a1 .                      [plus optional /bar for barchart or /his for histogram]

Others might suggest you use imputed values where these are missing, but I’ve never done this.  Some argue for, others against.

 

Sometimes findings are easier to present as tables of %% or as charts.  Why stick to regression?

 

I take it this is assessment for MBA?  My experience of MBAs is that you’ll neither learn nor retain much about SPSS, research methods or statistics once your assessment is in, but I could be wrong.

 

John F Hall (Mr)

[Retired academic survey researcher]

 

Email:   [hidden email] 

Website: www.surveyresearch.weebly.com

SPSS start page:  http://surveyresearch.weebly.com/1-survey-analysis-workshop.html

 

  

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: 14 December 2013 20:01
To: [hidden email]
Subject: Re: Compute Variable

 

functions such as SUM or MEAN, etc. can have a suffix which says the minimum number of variables in the list that must be valid to allow the calculation, e.g.,

COMPUTE idealized_influenc_a=mean.3(IIA1,IIA2,IIA3,IIA4).
would say that if there are 3 valid values calculate the mean other wise return a missing value.

A score using the MEAN function is on  the same scale as the response scale of the items. 
It does not need to be corrected.
It correlates with other variables the same as a  SUM score corrected for the number of valid items.

-----
Not always, but very commonly missing items within scales are due to poor administration of the test.

-----
You do not need EXECUTE after every compute.

Art Kendall

Social Research Consultants

On 12/14/2013 12:29 PM, MBAstudent [via SPSSX Discussion] wrote:

Hi All,

It's the first time I'm using SPSS and I could use some help.

I'm measuring the effectiveness of transformational leadership style using 20 items, which are subdivided into 5 different behaviours. The effectiveness is measured using 9 items, which are subdivided into 3 outcomes. (USING THE instrument MLQ5X )

In order for me to apply a regression analysis, I need to first create a new variable for each effectiveness outcome and then create the dependant variable (9 items measuring effectiveness- all 3 outcomes). And to bring together the items measuring each leadership behaviour in order to create independent variables (which should create 5 variables).

When trying to create an independent variable for example.
 I have tried to do so in a few ways:

1) independent variable
COMPUTE idealized_influenc_a=SUM(IIA1,IIA2,IIA3,IIA4).
EXECUTE

2)COMPUTE idealized_influenc_a2=IIA1 + IIA2 + IIA3 + IIA4.
EXECUTE

However, as I have missing values (which are defined ) I get a . in part of them, which holds me back with my analysis.

What is it that i need to do in order to create independent and dependent variables from my research items ?

Thank you all so VERY MUCH

---

If you reply to this email, your message will be added to the discussion below:

http://spssx-discussion.1045642.n5.nabble.com/Compute-Variable-tp5723637.html

To start a new topic under SPSSX Discussion, email [hidden email]
To unsubscribe from SPSSX Discussion, click here.
NAML

 

Art Kendall
Social Research Consultants

 

---

View this message in context: Re: Compute Variable
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

[David Marso]

I guess you need to better communicate NOW you wish to create these variables!
A SUM when there are missing values is probably NOT appropriate.
A MEAN as Art proposes might be better, but that makes possibly implausible assumptions about what the value of the missing values are (the mean of the others) .
SO, define your desired variables and for crying out loud read the FM (fine manual) about COMPUTE!

---

MBAstudent wrote

Hi All,

It's the first time I'm using SPSS and I could use some help.

I'm measuring the effectiveness of transformational leadership style using 20 items, which are subdivided into 5 different behaviours. The effectiveness is measured using 9 items, which are subdivided into 3 outcomes. (USING THE instrument MLQ5X )

In order for me to apply a regression analysis, I need to first create a new variable for each effectiveness outcome and then create the dependant variable (9 items measuring effectiveness- all 3 outcomes). And to bring together the items measuring each leadership behaviour in order to create independent variables (which should create 5 variables).

When trying to create an independent variable for example.
 I have tried to do so in a few ways:

1) independent variable
COMPUTE idealized_influenc_a=SUM(IIA1,IIA2,IIA3,IIA4).
EXECUTE

2)COMPUTE idealized_influenc_a2=IIA1 + IIA2 + IIA3 + IIA4.
EXECUTE

However, as I have missing values (which are defined ) I get a . in part of them, which holds me back with my analysis.

What is it that i need to do in order to create independent and dependent variables from my research items ?

Thank you all so VERY MUCH