TY - JOUR
T1 - A latent structure analysis of the positive and negative symptoms in schizophrenia
AU - Brekke, John S.
AU - DeBonis, Judith A.
AU - Graham, John W.
PY - 1994
Y1 - 1994
N2 - Although previous studies have suggested the inadequacy of the two-factor models of positive and negative symptoms in schizophrenia, confirmatory testing of the putative three-factor models is needed. Using a sample of 193 individuals diagnosed with schizophrenia, this study tested the relative goodness-of-fit of one-, two-, and three-factor models of the positive and negative symptoms. Using confirmatory factor analysis (CFA), the three-factor model of Addington et al., Arndt et al., and Liddle and Barnes that specifies positive, negative, and disorganized factors had the best fit with the data. Allowing the factors to co-vary and specifying dimensionality to the negative symptoms substantially improved the fit of the model. The study addressed several other issues. First, whereas the correlation between positive and negative symptoms was modest, the disorganized symptoms were significantly and more strongly related to both the positive and negative symptoms. Second, depression was not correlated with negative symptoms, but was significantly related to both the positive and disorganized symptoms. Third, the relationships between the three factors and levels of global, social, and work functioning in the sample supported the criterion-related validity of the three-factor model.
AB - Although previous studies have suggested the inadequacy of the two-factor models of positive and negative symptoms in schizophrenia, confirmatory testing of the putative three-factor models is needed. Using a sample of 193 individuals diagnosed with schizophrenia, this study tested the relative goodness-of-fit of one-, two-, and three-factor models of the positive and negative symptoms. Using confirmatory factor analysis (CFA), the three-factor model of Addington et al., Arndt et al., and Liddle and Barnes that specifies positive, negative, and disorganized factors had the best fit with the data. Allowing the factors to co-vary and specifying dimensionality to the negative symptoms substantially improved the fit of the model. The study addressed several other issues. First, whereas the correlation between positive and negative symptoms was modest, the disorganized symptoms were significantly and more strongly related to both the positive and negative symptoms. Second, depression was not correlated with negative symptoms, but was significantly related to both the positive and disorganized symptoms. Third, the relationships between the three factors and levels of global, social, and work functioning in the sample supported the criterion-related validity of the three-factor model.
UR - http://www.scopus.com/inward/record.url?scp=0028232917&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0028232917&partnerID=8YFLogxK
U2 - 10.1016/0010-440X(94)90016-7
DO - 10.1016/0010-440X(94)90016-7
M3 - Article
C2 - 7956180
AN - SCOPUS:0028232917
SN - 0010-440X
VL - 35
SP - 252
EP - 259
JO - Comprehensive Psychiatry
JF - Comprehensive Psychiatry
IS - 4
ER -