Hector and Jon,

Thanks for your replies and the heads-up on more efficient syntax. WRT to Jon's comment, I get the same singularity message even when I drop all the other predictors, i.e., I use political affiliations as the only predictors.

I have three related issues that might together help me clear this up.

1. With Hector's coding below, I imagine that I would still need to declare those cases where democrat = 0. Would I still use an IF statement of some sort? Using your set up, would the following be correct for democrats or is there some "non-IF" statement that would be more efficient.

MISSING VALUES q78a (4,9). COMPUTE DEMOCRAT=(q78a=1). IF (q78a = 2 | q78a = 3 | q78a = 5) democrat = 0 . EXECUTE.

2. Thanks for the info on how to set up missing values. I was about to ask how to do that. However, do I need to declare missing values? When I create the new variables, e.g., for Democrat, I would *not* have declared a value for when the raw data (q78a) = 4 or 9.

In other words, when q78a = 4, democrat contains a period. Is this period not equivalent to a missing value.

(I realize that you are declaring missing values for q78a and I am talking about missing values for the field Democrat. In any case, I do not see why I need to declare missing values for either q78a or for the field Democrat. Moreover, suppose I want to perform subsequent analysis on the "don't knows" would declaring those don't knows as missing values is q78a not cause me to lose information that I might need for subsequent analyses. I suspect that an answer to this question would help me understand just what a missing value is and why it might be important to declare it.)

3. On a related note, how does one write syntax for a period. E.g., suppose the interviewer fails to collect any information for certain cell and the cell turns up as a period in SPSS. I usually assume that SPSS treats this as missing data and that I do not have to worry about it -- that case is simply left out the analysis. However, suppose I wanted to write syntax to manipulate this period, how do I treat this period?

Thanks much.

Rayman

---- Original message ---- >Date: Fri, 17 Jun 2005 01:01:33 -0300 >From: "Hector Maletta" <hmaletta@fibertel.com.ar> >Subject: RE: Unexpected singularities >To: <ar7347@wayne.edu>, <SPSSX-L@LISTSERV.UGA.EDU> > >Rayman, >You should be aware that singularities can arise even when you do everything >right, just by chance. Entering all the dummies assures redundancy, but >leaving one category aside does not ensure you do not get it. > >Besides, your syntax is essentially correct in my view, though somewhat >inefficient. Each EXECUTE necessitares reading the file. You can put just >one at the end, saving time. > >Also you can create each dummy in a single command each (but define DON'T >KNOWS or other excluded values as missing before you do that, so that the >dummies are system missing for them). I assume 9 and 4 are to be ignored as >missing. > >MISSING VALUES q78a (4,9). >COMPUTE DEMOCRAT=(q78a=1). >COMPUTE REPUBLIC=(q78a=2). >COMPUTE IND_OTHER=(q78a=3 OR q78a=5). >EXECUTE. > >Hector > >> -----Original Message----- >> From: SPSSX(r) Discussion [mailto:SPSSX- L@LISTSERV.UGA.EDU] >> On Behalf Of Rayman Mohamed >> Sent: Friday, June 17, 2005 12:03 AM >> To: SPSSX-L@LISTSERV.UGA.EDU >> Subject: Unexpected singularities >> >> >> Hi all, >> >> I have been doing some multinomial logistic regressions. >> When I use two particular categorical variables (in >> appropriate dummy form -- see later) in my models I get the >> following message. >> >> "Unexpected singularities in the Hessian matrix are >> encountered. This indicates that either some predictor >> variables should be excluded or some categories should be merged." >> >> This implies that I have redundancy in my dummy variables (I >> think). However, I have checked and rechecked but I cannot >> see where I am going wrong. One of the categorical variables >> for which I am getting this message is political affiliation. >> I created three dummy variables (democrat, republican, and >> independent/no affiliation) and entered two into the >> regression model. (There are other categories that I ignored >> when I created the dummy variables, e.g., the "don't knows".) >> >> Could anyone think of another reason why I am getting this >> message. Or do you think that perhaps I just looked at this >> too much so I cannot see some rather obvious mistake in my >> syntax to create the dummy variables? >> >> Here is the syntax that I used to create the dummy variables >> for political affiliation: >> >> IF (q78a = 1) democrat_1 = 1 . >> EXECUTE . >> IF (q78a = 2 | q78a = 3 | q78a = 5) democrat_1 = 0 . >> EXECUTE . >> >> IF (q78a = 2) republican_1 = 1 . >> EXECUTE . >> IF (q78a = 1 | q78a = 3 | q78a = 5) republican_1 = 0 . >> EXECUTE . >> >> IF (q78a = 3 | q78a = 5) Ind_other_1 = 1 . >> IF (q78a = 1 | q78a = 2) ind_other_1 = 0 . >> EXECUTE . >> >> Thanks for your suggestions, >> >> R. >> Rayman Mohamed >> Assistant Professor >> Wayne State University >> College of Urban, Labor, and Metropolitan Affairs >> Department of Geography and Urban Planning >> 227 State Hall >> 5143 Cass Avenue >> Detroit, MI 48202 >> http://www.culma.wayne.edu >> > > Rayman Mohamed Assistant Professor Wayne State University College of Urban, Labor, and Metropolitan Affairs Department of Geography and Urban Planning 227 State Hall 5143 Cass Avenue Detroit, MI 48202 http://www.culma.wayne.edu