You can do this with the star.cutoffs
and star.char
arguments. Some fake data below to demonstrate:
library(stargazer)
# Generate some fake data
set.seed(10)
x <- rnorm(10)
x1 <- rnorm(10)
e <- rnorm(10)
y <- 10 + x + 2*x1 + e
# Estimate a model
m1 <- lm(y~x + x1)
# We can see that we have three different levels of sig at typical cutoffs
summary(m1)
#>
#> Call:
#> lm(formula = y ~ x + x1)
#>
#> Residuals:
#> Min 1Q Median 3Q Max
#> -1.12552 -0.20126 -0.06919 0.60370 0.76845
#>
#> Coefficients:
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 9.1012 0.3731 24.394 4.95e-08 ***
#> x 0.8389 0.3923 2.138 0.0698 .
#> x1 1.7477 0.4094 4.269 0.0037 **
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Residual standard error: 0.7851 on 7 degrees of freedom
#> Multiple R-squared: 0.8167, Adjusted R-squared: 0.7643
#> F-statistic: 15.59 on 2 and 7 DF, p-value: 0.002639
# We will make the 10% level a plus sign, and stars for .05, .01 and .001
stargazer(m1, type = "text",
star.char = c("+", "*", "**", "***"),
star.cutoffs = c(.1, .05, .01, .001))
#>
#> ===============================================
#> Dependent variable:
#> ---------------------------
#> y
#> -----------------------------------------------
#> x 0.839+
#> (0.392)
#>
#> x1 1.748**
#> (0.409)
#>
#> Constant 9.101***
#> (0.373)
#>
#> -----------------------------------------------
#> Observations 10
#> R2 0.817
#> Adjusted R2 0.764
#> Residual Std. Error 0.785 (df = 7)
#> F Statistic 15.589** (df = 2; 7)
#> ===============================================
#> Note: *p<0.1; **p<0.05; ***p<0.01
Created on 2018-08-08 by the reprex package (v0.2.0).
star.cutoffs = c(.1, .05, .01)
or whichever levels you prefer. If you need to set the symbols, you can use the argumentstar.char = c(".", "*", "**")
to specify the symbol corresponding to each significance level. – Merits