Serious problems with SPSS data analysis!

[MeerieB]

Dear all,
 
I have some problems with analysing some of the results from SPSS. I have added the parts on which i encounter some serious problems below. As you can see in the table belonging to paragraph 5.2 I have made an attempt in assesssing the validity of four different scales that I have made according a questionnaire that I have set up earlier and where different people have responded on. The cronbach's alpha is rather small for the element called 'careercha' (that is careerchange) in order to be called internally consistent so my question is what can I do to make this element be internally consistent? Or can I assume it being consistent since it belongs to a relatively small sample since the data results from a sample of 54 respondents out of 160? I read somewhere that if you have a relatively small sample you can take an alpha > 0.5 but I don't know for sure whether this is true.


§ 5.2    Scale validity

The next step of the data analysis is assessing the validity of the scales used in the questionnaire. Cronbach's alpha is obtained on each scale of the different variable. The Cronbach's alpha is a reliability coefficient that indicates how well the items in a set are positively correlated to one another.

 

Scale

 Cronbach's alpha
 
communic
 0,8
 
orgcult
 0,451
 
psychlead
 0,841
 
careercha
 0,657
 


 

The cut-off for a reliable scale lies at a minimum α > 0.7, which means that the scales communication and psychological leadership are internally consistent. The scale career change has a Cronbach's alpha which is near 0.7 and therefore I consider this scale as being internally consistent as well. For the scale organizational culture a better Cronbach's alpha can be obtained by deleting certain items, this would be question 18 from the questionnaire. When this is done the following (improved) Cronbach's alpha's are obtained:

 

Scale

 Cronbach's alpha
 
communic
 0,8
 
orgcult
 0,834
 
psychlead
 0,841
 
careercha
 0,657
 


 


Another 'issue' that I have encountered is that I don't know how to give a clear explanation on the correlations between the different scales which are set out in paragraph 5.4, my question is whether the manner in which I tried to describe it here is a good description or are there any elements missing and/or should I work this out differently?
 

 

§ 5.4    Correlations

The findings as stated above could be related to each other, the way to detect this is by conducting a Pearson Correlations Matrix as done below.

                                                                           Correlations

 

 

 
 communic
 orgcult
 psychlead
 careercha
 
communic
 Pearson Correlation
 1
 -.828(**)
 .812(**)
 -.627(**)
 
   Sig. (2-tailed)
 
 .000
 .000
 .000
 
   N
54
 54
 54
 54
 
orgcult
 Pearson Correlation
 -.828(**)
 1
 -.836(**)
 .601(**)
 
   Sig. (2-tailed)
.000
 
 .000
 .000
 
   N
54
 54
 54
 54
 
psychlead
 Pearson Correlation
 .812(**)
 -.836(**)
 1
 -.609(**)
 
   Sig. (2-tailed)
.000
 .000
 
 .000
 
 N
54
 54
 54
 54
 
careercha
 Pearson Correlation
 -.627(**)
 .601(**)
 -.609(**)
 1
 
   Sig. (2-tailed)
.000
 .000
 .000
 
 
   N
54
 54
 54
 54
 


**  Correlation is significant at the 0.01 level (2-tailed).

 

 

Negative correlations:

Organizational Culture - Communication.

The relatively strong negative relationship between organizational culture and communication could be explained by the fact that whenever there is more cultural convergence less communication is needed and thus less communication will occur. While in case of very little cultural convergence more communication will be necessary in order for an organization to function.

 

Organizational Culture - Psychological Leadership.

Also between the variables organizational culture and psychological leadership there appears to be a negative relationship. An explanation for this outcome could be that a clear continuity and visibility of leadership is needed when there is no cultural convergence between the different groups within an organization.

 

Psychological Leadership - Career Change

With a visible and continuous leadership the employees will perceive less career change, while in case of no visible and continuous leadership the employees will perceive more career change during the conduction of a merger.

 

Communication  - Career Change.

Less communication will eventually arouse more feelings of radical changes in the personal situation. When there is clear communication about the changes that will be effectuated due to the merger the employees know what they could expect and anticipate and therefore will perceive less career changes.

 

Positive correlations:

Communication - Psychological Leadership

In case of clear continuity and visibility of leadership the employees within an organization are better able to communicate with each other.

 

Organizational Culture - Career Change

It seems as if feelings of change in the career have a positive effect on the cultural convergence within an organization.


 
I really need your help with this because I think I am doing it wrong!

 
Thank you!
 
Kind regards Meerie
 

[Burleson,Joseph A.]

Meerie:

You are correct in that you have a number of issues to deal with here. In fact, you have a whole course's worth of issues, seemingly none of them having to do with SPSS per see. So I do not see anywhere where there are "serious problems with SPSS data analysis."

You probably need to start with a text on reliability and validity, or, more basically, a text on research design in which these chapters are included.

I am sure there are any number of people on the listserve that will probably be happy to answer your questions (there always are!).

Joe Burleson

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of MeerieB
Sent: Wednesday, July 30, 2008 1:55 PM
To: [hidden email]
Subject: Serious problems with SPSS data analysis!

Dear all,

I have some problems with analysing some of the results from SPSS. I have added the parts on which i encounter some serious problems below. As you can see in the table belonging to paragraph 5.2 I have made an attempt in assesssing the validity of four different scales that I have made according a questionnaire that I have set up earlier and where different people have responded on. The cronbach's alpha is rather small for the element called 'careercha' (that is careerchange) in order to be called internally consistent so my question is what can I do to make this element be internally consistent? Or can I assume it being consistent since it belongs to a relatively small sample since the data results from a sample of 54 respondents out of 160? I read somewhere that if you have a relatively small sample you can take an alpha > 0.5 but I don't know for sure whether this is true.


§ 5.2    Scale validity

The next step of the data analysis is assessing the validity of the scales used in the questionnaire. Cronbach's alpha is obtained on each scale of the different variable. The Cronbach's alpha is a reliability coefficient that indicates how well the items in a set are positively correlated to one another.



Scale

 Cronbach's alpha

communic
 0,8

orgcult
 0,451

psychlead
 0,841

careercha
 0,657





The cut-off for a reliable scale lies at a minimum α > 0.7, which means that the scales communication and psychological leadership are internally consistent. The scale career change has a Cronbach's alpha which is near 0.7 and therefore I consider this scale as being internally consistent as well.
For the scale organizational culture a better Cronbach's alpha can be obtained by deleting certain items, this would be question 18 from the questionnaire. When this is done the following (improved) Cronbach's alpha's are obtained:



Scale

 Cronbach's alpha

communic
 0,8

orgcult
 0,834

psychlead
 0,841

careercha
 0,657






Another 'issue' that I have encountered is that I don't know how to give a clear explanation on the correlations between the different scales which are set out in paragraph 5.4, my question is whether the manner in which I tried to describe it here is a good description or are there any elements missing and/or should I work this out differently?




§ 5.4    Correlations

The findings as stated above could be related to each other, the way to detect this is by conducting a Pearson Correlations Matrix as done below.


Correlations






 communic
 orgcult
 psychlead
 careercha

communic
 Pearson Correlation
 1
 -.828(**)
 .812(**)
 -.627(**)

   Sig. (2-tailed)

 .000
 .000
 .000

   N
54
 54
 54
 54

orgcult
 Pearson Correlation
 -.828(**)
 1
 -.836(**)
 .601(**)

   Sig. (2-tailed)
.000

 .000
 .000

   N
54
 54
 54
 54

psychlead
 Pearson Correlation
 .812(**)
 -.836(**)
 1
 -.609(**)

   Sig. (2-tailed)
.000
 .000

 .000

 N
54
 54
 54
 54

careercha
 Pearson Correlation
 -.627(**)
 .601(**)
 -.609(**)
 1

   Sig. (2-tailed)
.000
 .000
 .000


   N
54
 54
 54
 54



**  Correlation is significant at the 0.01 level (2-tailed).





Negative correlations:

Organizational Culture - Communication.

The relatively strong negative relationship between organizational culture and communication could be explained by the fact that whenever there is more cultural convergence less communication is needed and thus less communication will occur. While in case of very little cultural convergence more communication will be necessary in order for an organization to function.



Organizational Culture - Psychological Leadership.

Also between the variables organizational culture and psychological leadership there appears to be a negative relationship. An explanation for this outcome could be that a clear continuity and visibility of leadership is needed when there is no cultural convergence between the different groups within an organization.



Psychological Leadership - Career Change

With a visible and continuous leadership the employees will perceive less career change, while in case of no visible and continuous leadership the employees will perceive more career change during the conduction of a merger.



Communication  - Career Change.

Less communication will eventually arouse more feelings of radical changes in the personal situation. When there is clear communication about the changes that will be effectuated due to the merger the employees know what they could expect and anticipate and therefore will perceive less career changes.



Positive correlations:

Communication - Psychological Leadership

In case of clear continuity and visibility of leadership the employees within an organization are better able to communicate with each other.



Organizational Culture - Career Change

It seems as if feelings of change in the career have a positive effect on the cultural convergence within an organization.



I really need your help with this because I think I am doing it wrong!


Thank you!

Kind regards Meerie



--
View this message in context: http://www.nabble.com/Serious-problems-with-SPSS-data-analysis%21-tp18738990p18738990.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