This function attempts to return, or compute, p-values of mixed models.

```
# S3 method for cpglmm
p_value(model, method = "wald", ...)
# S3 method for glmmTMB
p_value(
model,
component = c("all", "conditional", "zi", "zero_inflated", "dispersion"),
verbose = TRUE,
...
)
# S3 method for lmerMod
p_value(model, method = "wald", ...)
# S3 method for merMod
p_value(model, method = "wald", ...)
# S3 method for MixMod
p_value(
model,
component = c("all", "conditional", "zi", "zero_inflated"),
verbose = TRUE,
...
)
# S3 method for mixor
p_value(model, effects = "all", ...)
```

## Arguments

model |
A statistical model. |

method |
For mixed models, can be `"wald"()` (default), `"ml1"()` , `"betwithin"()` , `"satterthwaite"()` or `"kenward"()` . For models that are supported by the sandwich or clubSandwich packages, may also be `method = "robust"` to compute p-values based ob robust standard errors. |

... |
Arguments passed down to `standard_error_robust()` when confidence intervals or p-values based on robust standard errors should be computed. Only available for models where `method = "robust"` is supported. |

component |
Should all parameters, parameters for the conditional model,
or for the zero-inflated part of the model be returned? Applies to models
with zero-inflated component. `component` may be one of `"conditional"` ,
`"zi"` , `"zero-inflated"` , `"dispersion"` or `"all"`
(default). May be abbreviated. |

verbose |
Toggle warnings and messages. |

effects |
Should standard errors for fixed effects or random effects be
returned? Only applies to mixed models. May be abbreviated. When standard
errors for random effects are requested, for each grouping factor a list of
standard errors (per group level) for random intercepts and slopes is
returned. |

## Value

A data frame with at least two columns: the parameter names and the p-values. Depending on the model, may also include columns for model components etc.

## Details

By default, p-values are based on Wald-test approximations (see `p_value_wald()`

). For certain situations, the "m-l-1" rule might be a better approximation. That is, for `method = "ml1"`

, `p_value_ml1()`

is called. For `lmerMod`

objects, if `method = "kenward"`

, p-values are based on Kenward-Roger approximations, i.e. `p_value_kenward()`

is called, and `method = "satterthwaite"`

calls `p_value_satterthwaite()`

.

## Note

`p_value_robust()`

resp. `p_value(method = "robust")`

rely on the sandwich or clubSandwich package (the latter if
`vcov_estimation = "CR"`

for cluster-robust standard errors) and will
thus only work for those models supported by those packages.

## Examples

```
if (require("lme4")) {
data(iris)
model <- lmer(Petal.Length ~ Sepal.Length + (1 | Species), data = iris)
p_value(model)
}
#> Parameter p
#> 1 (Intercept) 9.561737e-01
#> 2 Sepal.Length 1.273781e-44
```