spliting variable
Locked
 
|
Dear Listers,
I have a variable named household numer which has a combination of several numbers. The first two digits is the cluster number.
I need to have the cluster number as a separate variable
eg
120081079/00/0 to 12
110081071/00/1 to 11
Any help?
Caleb Ouma Dataworld Research P.O Box 1984-00200 NAIROBI Alt. E-mail: [hidden email] Skype: caleb.ouma |
Calculating polychoric correlations?
|
|
Dear all,
Does someone know a program that can calculate polychoric correlations? I have normality problem with some of my variables and I was told that using polychoric correlations may be a solution. I'm using Amos and it doesn't calculate polychoric correlations. I don't have access to Lisrel or Mplus. Any help will be appreciated. Mbaye, Commander un cadeau en ligne discrètement? Voir la solution offerte par Internet Explorer8 |
|
In reply to this post by Caleb Ouma
You really need to supply more information about
your data file.
If the first part of your variable is numeric (ie
not the slashes. 00 etc.), try
compute newnum = trunc
(oldnum / 1000000) .
If it's a string with the slashes included, someone
else will have a solution.
If your raw data are in fixed format, and the
housenum is at the beginning of the first data line, you could read in the data
as
data list file = <file address>
records =
<number of records>
/1
housnum 1-2 .
If your data are on a different file format, you
should be able to output a *.txt file as a preliminary step.
|
Re: Calculating polychoric correlations?
Administrator
|
In reply to this post by Mbaye Fall Diallo
I've not used it, so cannot vouch for it, but here's one: http://www.john-uebersax.com/stat/xpc.htm
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
Re: Calculating polychoric correlations?
|
|
In reply to this post by Mbaye Fall Diallo
The HETCOR extension command available with Statistics V17 or later does this (polyserial, polychoric, and Pearson). It can be downloaded from SPSS Developer Central (www.spss.com/devcentral). Requirements are in the readme file in the download. HTH, Jon Peck Jon Peck SPSS, an IBM Company [hidden email] 312-651-3435
Dear all, Does someone know a program that can calculate polychoric correlations? I have normality problem with some of my variables and I was told that using polychoric correlations may be a solution. I'm using Amos and it doesn't calculate polychoric correlations. I don't have access to Lisrel or Mplus. Any help will be appreciated. Mbaye, Commander un cadeau en ligne discrètement? Voir la solution offerte par Internet Explorer8 |
Re: Calculating polychoric correlations?
|
|
In reply to this post by Mbaye Fall Diallo
Mbaye, You will find the Polycor plug-in here. I recall that it is pretty easy to install but you first have to install R. http://www.spss.com/devcentral/index.cfm?pg=downloadDet&dId=191 *************************************************************************************************************************************************************** Mark A. Davenport Ph.D. Senior Research Analyst Office of Institutional Research The University of North Carolina at Greensboro 336.256.0395 [hidden email] 'An approximate answer to the right question is worth a good deal more than an exact answer to an approximate question.' --a paraphrase of J. W. Tukey (1962)
Dear all, Does someone know a program that can calculate polychoric correlations? I have normality problem with some of my variables and I was told that using polychoric correlations may be a solution. I'm using Amos and it doesn't calculate polychoric correlations. I don't have access to Lisrel or Mplus. Any help will be appreciated. Mbaye, Commander un cadeau en ligne discrètement? Voir la solution offerte par Internet Explorer8 |
Re: Calculating polychoric correlations?
|
|
In reply to this post by Jon K Peck
Jon, I thought Polycor superceeded HETCOR. Which do you recommend? Do they work with v18? *************************************************************************************************************************************************************** Mark A. Davenport Ph.D. Senior Research Analyst Office of Institutional Research The University of North Carolina at Greensboro 336.256.0395 [hidden email] 'An approximate answer to the right question is worth a good deal more than an exact answer to an approximate question.' --a paraphrase of J. W. Tukey (1962) |
Re: Calculating polychoric correlations?
|
|
In reply to this post by Mbaye Fall Diallo
The program FACTOR con do it: http://psico.fcep.urv.cat/utilitats/factor/ Andrés Mg. Andrés Burga León Coordinador de Análisis e Informática Unidad de Medición de la Calidad Educativa Ministerio de Educación del Perú Calle El Comercio s/n (espalda del Museo de la Nación) Lima 41 Perú Teléfono 615-5840
Dear all, Does someone know a program that can calculate polychoric correlations? I have normality problem with some of my variables and I was told that using polychoric correlations may be a solution. I'm using Amos and it doesn't calculate polychoric correlations. I don't have access to Lisrel or Mplus. Any help will be appreciated. Mbaye, Commander un cadeau en ligne discrètement? Voir la solution offerte par Internet Explorer8 |
Creating new data with sample mean
|
|
In reply to this post by Mbaye Fall Diallo
Hello to everybody: I need to create a new dataset containning the means from X random samples taken from a dataset called "normal". Firt, I've made this macro to create a normal distribution of test scores: DEFINE !normal(casos=!TOKENS(1) / media=!TOKENS(1) / de=!TOKENS(1)) INPUT PROGRAM. + LOOP caso = 1 TO !casos. + COMPUTE medida = RND(RV.NORMAL(!media,!de)) . + END CASE. + END LOOP. + END FILE. END INPUT PROGRAM. FORMATS caso (F10.0). FORMATS medida (F3.0). EXECUTE . DATASET NAME normal. !ENDDEFINE. !normal casos=5000 media=23 de=7. Then I draw a random sample of 500 cases (dataset "data") and create a new dataset containing the sample mean ("media"): DATASET ACTIVATE normal. DATASET COPY data. DATASET ACTIVATE data. FILTER OFF. USE ALL. SAMPLE 500 from 5000. EXECUTE. DATASET DECLARE media. AGGREGATE /OUTFILE='media' /BREAK= /medida_media=MEAN(medida) /casos=N. DATASET CLOSE data. Finaly I need to repeat the following syntax 200 times to have 200 diferent means estimates agregated to the "media" dataset, but I'm not sure how to do it: DATASET ACTIVATE normal. DATASET COPY data. DATASET ACTIVATE data. FILTER OFF. USE ALL. SAMPLE 500 from 5000. EXECUTE. DATASET DECLARE temporal. AGGREGATE /OUTFILE='temporal' /BREAK= /medida_media=MEAN(medida) /casos=N. DATASET ACTIVATE media. ADD FILES /FILE=* /FILE='temporal'. EXECUTE. DATASET CLOSE temporal. DATASET CLOSE data. Coul somebody help me tho specify the sintax for repeating the last part 200 times Thank you Andrés Mg. Andrés Burga León Coordinador de Análisis e Informática Unidad de Medición de la Calidad Educativa Ministerio de Educación del Perú Calle El Comercio s/n (espalda del Museo de la Nación) Lima 41 Perú Teléfono 615-5840 |
Re: Creating new data with sample mean
|
you can adapt the compute statement to match your needs. *draw 1000 samples of size 50 without replacement. *create some data. new file. set seed = 20051209. INPUT PROGRAM. loop set = 1 to 1000. leave set. LOOP id=1 TO 50. COMPUTE x = rv.normal(0,1). END CASE. END LOOP. end loop. END FILE. END INPUT PROGRAM. dataset declare rawfile. FORMATS set id (F4.0) x (f6.3). descriptives variables =x/statistics=all. dataset declare aggfile. aggregate outfile=aggfile /break = set /setmean = mean(x). dataset activate aggfile. descriptives variables= setmean /statistics=all. Art Kendall Social Research Consultants On 5/6/2010 12:58 PM, ANDRES ALBERTO BURGA LEON wrote: ===================== 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
Art Kendall
Social Research Consultants |
Re: Calculating polychoric correlations?
|
|
In reply to this post by ANDRES ALBERTO BURGA LEON
Many thanks for all your suggestions. I will try and let you know.
Best, Mbaye, Date: Thu, 6 May 2010 11:48:29 -0500 From: [hidden email] Subject: Re: Calculating polychoric correlations? To: [hidden email] The program FACTOR con do it: http://psico.fcep.urv.cat/utilitats/factor/ Andrés Mg. Andrés Burga León Coordinador de Análisis e Informática Unidad de Medición de la Calidad Educativa Ministerio de Educación del Perú Calle El Comercio s/n (espalda del Museo de la Nación) Lima 41 Perú Teléfono 615-5840
Dear all, Does someone know a program that can calculate polychoric correlations? I have normality problem with some of my variables and I was told that using polychoric correlations may be a solution. I'm using Amos and it doesn't calculate polychoric correlations. I don't have access to Lisrel or Mplus. Any help will be appreciated. Mbaye, Commander un cadeau en ligne discrètement? Voir la solution offerte par Internet Explorer8 Télécharger en toute sécurité sur Internet ? La solution avec Internet Explorer 8 |
|
|
In reply to this post by John F Hall
|
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |