need help with SPSS
Locked
 
|
All, Hi I am currently using SPSS 18 and would like to compute Pearson correlation . I have 265 variables (these are questionnaire items) and there is a limit of 100 variables in SPSS for both Spearman's and Pearson's. I could not perform the Pearson due to the mentioned limitation, I also failed to compute Factor analysis on my data due to this problem. Please do let me know if anyone know of any way of increasing the amount of variables SPSS can handle through a normal bivariate correlation? Or alternatively is there some other freely available easy to use tool that will allow me to do this? Many thanks Abia |
|
|
In v22 the limit seems to be 500. You can use FACTOR analysis command to output a (Pearsons') correlations matrix (without any limits, well, tested for 999 variables). /*Generate dummy data*/. set seed=10. input program. vector x (999,f3). loop id = 1 to 30. loop #p = 1 to 999. compute x(#p) = rnd(rv.normal(50,10)). end loop. end case. end loop. end file. end input program. exe. dataset name Data. /* Try with 501 variables produces error on variable limit of 500*/. *cor x1 to x501. cor x1 to x500. *cor x501 to x999. /* Pearsons' matrix beyond variable limit of CORRELATIONS command. factor variables=x1 TO x999 /matrix=out(cor='C:\Temp\FactorMatrixData.sav') /analysis x1 TO x999 /method=correlations . /* Spearmans' matrix beyond variable limit of CORRELATIONS command. dataset activate Data. set mxloops=1000. MATRIX. * Replace "x1 TO x999" by your variable names *. GET Data /VAR=x1 TO x999 /MISSING=OMIT. COMPUTE n=NROW(Data). COMPUTE k=NCOL(Data). * Rank data to get Spearman's correlation coefficients *. COMPUTE Ranks=MAKE(n,k,0). LOOP i=1 TO k. . COMPUTE Ranks(:,i)=RNKORDER(Data(:,i)). END LOOP. RELEASE Data,k. * MATRIX algebra to compute a correlation matrix *. COMPUTE Means=CSUM(Ranks)/n. COMPUTE StdDevs=SQRT(CSSQ(Ranks)/n-Means&**2). COMPUTE ZScores=(Ranks-MAKE(n,1,1)*Means)/(MAKE(n,1,1)*StdDevs). COMPUTE CorrMat=T(ZScores)*ZScores/n. * Export correlation matrix *. MSAVE CorrMat /TYPE=CORR /OUTFILE="C:\Temp\MatrixData.sav" /VARIABLES=x1 TO x999. END MATRIX. /* Check results match COR output*/. get file="C:\Temp\MatrixData.sav". dataset name MatrixData. /* Check results match COR output*/. get file="C:\Temp\FactorMatrixData.sav". dataset name FactorMatrixData. On 28 November 2014 at 06:57, abia nazim <[hidden email]> wrote:
|
|
|
You could of course, RANK the variables first and run the FACTOR analysis approach on the rank variables, as Marta describe also: http://spssx-discussion.1045642.n5.nabble.com/Bivariate-Correlation-Variable-Limit-tp1083152p1083158.html On 28 November 2014 at 08:31, Jignesh Sutar <[hidden email]> wrote:
|
|
In reply to this post by abia nazim
Why on earth are you trying to put so many variables into your analysis? Do you not have any theoretical basis for using sub-sets of questions that you think may be related? John F Hall (Mr) [Retired academic survey researcher] Email: [hidden email] Website: www.surveyresearch.weebly.com SPSS start page: www.surveyresearch.weebly.com/1-survey-analysis-workshop From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of abia nazim All, Hi I am currently using SPSS 18 and would like to compute Pearson correlation . I have 265 variables (these are questionnaire items) and there is a limit of 100 variables in SPSS for both Spearman's and Pearson's. I could not perform the Pearson due to the mentioned limitation, I also failed to compute Factor analysis on my data due to this problem. Please do let me know if anyone know of any way of increasing the amount of variables SPSS can handle through a normal bivariate correlation? Or alternatively is there some other freely available easy to use tool that will allow me to do this? Many thanks Abia ===================== 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 |
|
|
I totally agree. Reduce the number of variables first. For example, why don’t you compute scale scores first and use these as predictors? From: SPSSX(r) Discussion [mailto:[hidden email]]
On Behalf Of John F Hall Why on earth are you trying to put so many variables into your analysis? Do you not have any theoretical basis for using sub-sets of questions
that you think may be related? John F Hall (Mr) [Retired academic survey researcher] Email:
[hidden email] Website:
www.surveyresearch.weebly.com
SPSS start page:
www.surveyresearch.weebly.com/1-survey-analysis-workshop From: SPSSX(r) Discussion [[hidden email]]
On Behalf Of abia nazim All, Hi I am currently using SPSS 18 and would like to compute Pearson correlation . I have 265 variables (these are questionnaire items) and there is a limit of 100 variables in SPSS
for both Spearman's and Pearson's. I could not perform the Pearson due to the mentioned limitation, I also failed to compute Factor analysis on my data due to this problem. Please do let me know if anyone know of any way of increasing the amount of variables SPSS can handle through a normal bivariate correlation? Or alternatively is there some other freely available easy to use tool that will allow me to do this? Many thanks Abia ===================== 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 |
|
In reply to this post by abia nazim
Do you have a student version of SPSS?
Are the variables items from pre-existing scales? Are you trying to explore the items or just check whether you have the scoring key correct? How many cases do you have? What is the response scale for the items?
Art Kendall
Social Research Consultants |
| Free forum by Nabble | Edit this page |