reg1 template: survey weights


It is common in the epidemiology research to report odd ratios or risk-ratios and their p-values from logistic regression or other regression models. reg1 template of asdocx makes it extremely easy to build such tables with much needed flexibility. In my previous post, I showed how to use reg1 template [see post ].

In this post, I want to show you how to use the survey prefix with the reg1 template. The syntax of using the svy prefix with asdocx is given below.

Syntax

asdocx svy: reg depvar indepvars, title(text) dec(#) dect(#) ///
       template(reg1) reg_options 

 

Syntax Details


The syntax starts with the keywords asdocx followed by the svy: The word reg can be any Stata command for regressions such reg for ols, logistic for logistic regressions, etc.

The depvar is the dependent or the explained variable .

The indepvars are all other independent or explanatory variables in the regression model.

 

An Example


asdocx svy: logistic heartatk  tcresult tgresult age race sex houssiz smsa region, 
       template(reg1) dec(4) dect(2) replace

In the above example, heartatk is the dependent variable and all other variables are independent. After comma, option template(reg1) causes asdocx to use the reg1 template.  The above code  generates a table that looks like this.

 

Table: Regression Results – heartatk
Variable OR [95% CI] P-value
Serum cholesterol (mg/dL) 1.0001 [0.9958, 1.0045] 0.95
Serum triglycerides (mg/dL) 1.0023 [1.0007, 1.0038] 0.00
Age (years) 1.0941 [1.0826, 1.1056] 0.00
Race 0.8727 [0.5552, 1.3717] 0.54
Sex 0.4100 [0.2791, 0.6024] 0.00
Number of people in household 1.0034 [0.8641, 1.1651] 0.96
SMSA type 0.9938 [0.8426, 1.1721] 0.94
Region 1.0563 [0.8870, 1.2581] 0.53
Constant 0.0007 [0.0002, 0.0026] 0.00
Observations 5049

 

subpop() option


Subpopulation estimation focuses on part of the population. The svy prefix command’s subpop() option performs subpopulation estimation. If we were to estimate the above regression only for female population, we would use the subpop(if female == 1) options. See this example:

asdocx svy, subpop(if female == 1): logistic heartatk tcresult tgresult age race sex houssiz smsa region,
      template(reg1) dec(4) dect(2) replace 
Table: Regression Results – heartatk
Variable OR [95% CI] P-value
Serum cholesterol (mg/dL) 0.9983 [0.9922, 1.0043] 0.56
Serum triglycerides (mg/dL) 1.0081 [1.0050, 1.0112] 0.00
Age (years) 1.0783 [1.0511, 1.1062] 0.00
Race 0.9048 [0.4112, 1.9913] 0.80
o 1.0000 [ ., .] .
Number of people in household 0.9465 [0.7054, 1.2699] 0.71
SMSA type 0.8413 [0.6615, 1.0699] 0.15
Region 1.4284 [1.0825, 1.8849] 0.01
Constant 0.0001 [0.0000, 0.0009] 0.00
Observations 7523

Order asdocx


Yearly license of asdocx is available at $9.99. Its life-time license is available at $49.99. With the asdocx membership, you get :

  • Life-time license to use
  • All future updates
  • All premium templates / plugins.

 

Order asdocx