如何使用broom :: tidy（）从lme4 :: lmer（）创建的线性混合效果模型中计算p值？

Question

I have a built a mixed effect model using the lmer() function from the lme4 package. The lme4 package does not output the p-value of the coefficients for some good philosophical reason. However, I still need p-values to report in my publication. I know there are multiple ways to calculate p-values using the model created by lmer() , eg here .

My problem is: I want to extract p-value using the tidy() function from the broom package. Here, I really want to stick with tidy() because I want to maintain the following pipeline:

data_frame %>% group_by(grouping variables) %>% do(tidy(fitted_model))

One option would be to create a custom function and append it to the pipeline. However, the man page of the broom package ( http://rpackages.ianhowson.com/cran/broom/man/lme4_tidiers.html ) says:

"p.value  P-value computed from t-statistic (may be missing/NA)". 

By this I am assuming a function to calculate p-value from the t-value given by lmer model has already been implemented in broom. So, I am reluctant to reinvent the wheel.

The problem is I don't get the column with name p.value at all. I was expecting a column named p.value with NAs as the worst case scenario.

Code:

library(lme4)
library(broom)
lme <- lmer(Reaction ~ Days + (1 + Days | Subject), sleepstudy)
tidy(lme)
tidy(lme, effects = "fixed")

Output:

> tidy(lme)
                         term     estimate std.error statistic    group
1                  (Intercept) 251.40510485  6.824557 36.838306    fixed
2                         Days  10.46728596  1.545789  6.771485    fixed
3       sd_(Intercept).Subject  24.74045195        NA        NA  Subject
4              sd_Days.Subject   5.92213312        NA        NA  Subject
5 cor_(Intercept).Days.Subject   0.06555113        NA        NA  Subject
6      sd_Observation.Residual  25.59181564        NA        NA Residual
> tidy(lme, effects = "fixed")
         term  estimate std.error statistic
1 (Intercept) 251.40510  6.824557 36.838306
2        Days  10.46729  1.545789  6.771485

Answer 1

You will need package lmerTest to obtain p-values. tidy will not work on the lme object and you will need to append it to your format.

attach(mtcars)
lme <- lmer(mpg ~ cyl + (1 + cyl | carb), mtcars)
summary(lme)

Answer 2

您可以使用sjstats::p_value()从许多不同的模型（包括（通用）混合模型sjstats::p_value()中获取p值。