Could any one please provide me with the SAS code I should use for this task? The rate of a type of disease in the general population is 50%, and I am giving a group of randomly selected people a type of vitamin to observe if the vitamin can decrease the rate of disease in this treatment group. I want to be able to identify the significant effect as long as there is 40% rate of this disease in my treatment group at alpha .05 and power .80. What sample size minimally I should have? Thank you so much! ... View more

Hi there, I used R to get a sample size I need for a logistic regression study.  > wp.logistic(n = NULL, p0 = 0.5, p1 = 0.6, alpha = 0.05, + power = 0.8, family = "normal", parameter = c(0,1)) R said I need a sample N=214 if I want to discover a significant difference of .1 (p2=.6 and p1=.5, p2-p1=.1), at alpha=.05, power=.8.    However, when I use SAS to find the sample size based on the covariates I would like to have , it gives me smaller sample sizes (N=168, N=175) than N=214 when I want alpha=.05 and power=.8. Who can help me figure out what is going on? R said N at least should be 214 even without covariates. So I think adding covariates the required sample size should be at least more than N=214. proc power; logistic vardist("recidivism") = BINOMIAL (0.5, 1) vardist("gender") = BINOMIAL (0.5, 1) vardist("self-control") = ordinal((2 4 6) : (0.4 0.4 0.2)) vardist("age") = normal(24, 6) vardist("income") = normal(2400, 600) testpredictor = "recidivism" covariates = "gender"|"self-control" "age" "income" responseprob = 0.5 0.6 .07 testoddsratio = 2.4 alpha = 0.05 power = 0.8 ntotal = .; run; ... View more

Hi folks,  I have a question. I know people can use multiple comparisons such as Hsu's MCB to find out among several groups, which pair of group have the biggest contrast of the treatment mean, and which groups do not differ in their means of treatment effects.  Now I have a data for survival analysis in which there are several groups, I want to compare their survival risk curve in the observation time  to see which groups do not differ and which group different using MCB. Is that possible? I know using Kaplan-Meier curve we can only compare each two groups to see if they differ in survival risk. But I hope to have something like Hsu's MCB so that I can lump those groups that do not differ in their survival risk together into one group.  ... View more

Hi folks, I am running a proc calis procedure based on several latent factors. However I have a problem. I noticed that during the creation of latent factors using proc factor, the cases with missing values on the items used to create a latent factor could not be used. therefore, originally there were 260 cases, after creating some latent variables based on items with missing values, there only 150 cases to be used in SEM because the latent variables have missing values.  MY QUESTION: Should I impute missing values on a latent factor (after I create the latent factor) before running the SEM in proc calis? If so, how to impute missing values of a latent factor? Or I should first impute the values on the items to be used to create a latent factor, and then after imputation on the values of the items, I use proc factor to create a latent factor so that this factor would not have missing values? Thank you!   ... View more

Hi folks, I found myself lost in setting a baseline reference group for a categorical variable in my model of parametric survival analysis. There are 6 categories in priorarrests variable, coded in value 1 to 6. I want to use 2, the second group as my baseline.  This is my code proc lifereg data= roads.onlyrecordslessthan12 outest=weiboutest; class race_group gender_group marital_group famprb0new priorarrests (ref="2"); model survivaltime*censor_1(1) =priorarrests impulsivetemperment school0 age gender_group race_group famprb0new marital_group/distribution=lnormal; title 'Weibull regression for the data'; output out=weibsurv xbeta=weib_xb; run; The log shows 10 proc lifereg data= roads.onlyrecordslessthan12 outest=weiboutest; 11 class race_group gender_group marital_group famprb0new priorarrests (ref="2"); - 22 200 ERROR 22-322: Syntax error, expecting one of the following: a name, ;, -, /, :, _ALL_, _CHARACTER_, _CHAR_, _NUMERIC_. ERROR 200-322: The symbol is not recognized and will be ignored. 12 model survivaltime*censor_1(1) =priorarrests impulsivetemperment school0 age 13 gender_group race_group famprb0new marital_group/distribution=lnormal; 14 title 'Weibull regression for the data'; 15 output out=weibsurv xbeta=weib_xb; 16 run;   I tried using no quotation marks as this priorarrests (ref=2); But the log still gives error message. 17 proc lifereg data= roads.onlyrecordslessthan12 outest=weiboutest; 18 class race_group gender_group marital_group famprb0new priorarrests (ref=2); - 22 200 ERROR 22-322: Syntax error, expecting one of the following: a name, ;, -, /, :, _ALL_, _CHARACTER_, _CHAR_, _NUMERIC_. ERROR 200-322: The symbol is not recognized and will be ignored. 19 model survivaltime*censor_1(1) =priorarrests impulsivetemperment school0 age 20 gender_group race_group famprb0new marital_group/distribution=lnormal; 21 title 'Weibull regression for the data'; 22 output out=weibsurv xbeta=weib_xb; 23 run;   Anyone could help me? Thank you!     ... View more

Hi folks, I created several latent factors using proc factor. The items under each factor loads well, and I created the value of the composite using factor scores.  However, when I am about to use the factors as my independent variables in a model and I need to report descriptive statistics, I found generally the standard deviation of the facotrs are bigger than their means. Usually people would see a redflag when S.D. is bigger than the mean...But I think this rule may not apply to factors created based on factor scores? How can I explain it? Mean        Std Dev           Minimum      Maximum 0.158052   0.9290338     -3.4236783    1.781888 Mean                 Std Dev         Minimum    Maximum -0.0573974    0.9146756    -0.6069258     5.511616 Mean              Std Dev            Minimum      Maximum -0.1035039    0.8218769      -0.3771253     9.7391213 ... View more

Hi all, Can anyone help me with a question on using complex data? I know that we need to use strata and cluster statement in order to use weighted data. But I could not find any example that these statement could be used in manova... If I write my model like this, it will run. But I can only have one dependent variable.    proc surveyreg data = wave1.cro_section_neededIV; cluster psuscid; strata region; weight gswgt1; class BIO_SEX; model nonagressivedelin =BIO_SEX parentattchemnt schoolattachment delinqpeerw1 parentalcontrol BIO_SEX*cparentattchemnt BIO_SEX*cparentalcontrol age eduw1/ solution ; run;     If I add " manova h= _ALL_ ;" after model statement, it will not run... So it means when using weighted data, there is no way to use multivariate method? I also tried to use proc glm, not working with strata statement, either. Only proc reg is allowed to go with "survey" and strata statement. WHY? ... View more