How to fix "variable lengths differ" problem in linear regression model?

[Chul-Kyoo Jung]

Hi, I don't understand why the following codes generate error message:

library(wooldridge)
library(jtools)
library(huxtable)
data('card')


fit1 <- lm(card$educ~card$nearc4)
fit2 <- lm(card$educ~card$nearc4 + card$motheduc + card$fatheduc)
fit3 <- lm(card$lwage~card$nearc4)


if (requireNamespace("huxtable")) {
  # Export all 3 regressions with "Model #" labels,
  # standardized coefficients, and robust standard errors
  export_summs(fit1, fit2, fit3,
               model.names = c("Model 1","Model 2","Model 3"),
               scale = TRUE, robust = TRUE)
}

And the error is:

Error in model.frame.default(formula = card$educ ~ `card$nearc4` + `card$motheduc` +  :   variable lengths differ (found for 'card$nearc4')

I checked the length of each variable, which is all the same. Each fitted model works without any issue, but `export_summs()` seems to generate a problem. Is there any mistake in my code?

Thanks for help.

[Rich Goldstein]

this is Statalist and deals with the Stata program; your code is for R and thus this is the wrong place to ask a question about R; please go to an R list

[Ken Chui]

I'm new here so I am not sure if R questions are accepted (this is a Stata forum). I'll answer it anyway and beware that this thread may be deleted.

That error is often caused by missing data. If you check the summary:

summary(fit1)
summary(fit2)
summary(fit3)

You'll see that model 2 has about 800 fewer cases in it. Along the same vein, if you remove "fit2" and "Model 2" from the huxtable code chunk, you'll find it working.

[Chul-Kyoo Jung]

So sorry. Apology for my mistake. I was totally confused with Stata and r. I will remove this post right away. Thanks for all responses!