**Gain insight into your models!**

When fitting any statistical model, there are many useful pieces of information that are simultaneously calculated and stored beyond coefficient estimates and general model fit statistics. Although there exist some generic functions to obtain model information and data, many package-specific modelling functions do not provide such methods to allow users to access such valuable information.

**insight** is an R-package that fills this important
gap by providing a suite of functions to support almost any model (see a
list of the many models supported below in the **List of Supported
Packages and Models** section). The goal of
**insight**, then, is to provide tools to provide
*easy*, *intuitive*, and *consistent* access to
information contained in model objects. These tools aid applied research
in virtually any field who fit, diagnose, and present statistical models
by streamlining access to every aspect of many model objects via
consistent syntax and output.

The *insight* package is available on CRAN, while its latest
development version is available on R-universe (from *rOpenSci*)
or GitHub.

Type | Source | Command |
---|---|---|

Release | CRAN | `install.packages("insight")` |

Development | r-universe | `install.packages("insight", repos = "https://easystats.r-universe.dev")` |

Development | GitHub | `remotes::install_github("easystats/insight")` |

Once you have downloaded the package, you can then load it using:

`library("insight")`

TipInstead of

`library(insight)`

, use`library(easystats)`

. This will make all features of the easystats-ecosystem available.To stay updated, use

`easystats::install_latest()`

.

Built with non-programmers in mind, **insight** offers a
broad toolbox for making model and data information easily accessible.
While **insight** offers many useful functions for working
with and understanding model objects (discussed below), we suggest users
start with `model_info()`

, as this function provides a clean
and consistent overview of model objects (e.g., functional form of the
model, the model family, link function, number of observations,
variables included in the specification, etc.). With a clear
understanding of the model introduced, users are able to adapt other
functions for more nuanced exploration of and interaction with virtually
any model object.Please visit https://easystats.github.io/insight/ for
documentation.

The functions from **insight** address different
components of a model. In an effort to avoid confusion about specific
“targets” of each function, in this section we provide a short
explanation of **insight**’s definitions of regression
model components.

The dataset used to fit the model.

Values estimated or learned from data that capture the relationship
between variables. In regression models, these are usually referred to
as *coefficients*.

**response**: the outcome or response variable (dependent variable) of a regression model.**predictor**: independent variables of (the*fixed*part of) a regression model. For mixed models, variables that are only in the*random effects*part (i.e. grouping factors) of the model are not returned as predictors by default. However, these can be included using additional arguments in the function call, treating predictors are “unique”. As such, if a variable appears as a fixed effect and a random slope, it is treated as one (the same) predictor.

Any unique variable names that appear in a regression model, e.g.,
response variable, predictors or random effects. A “variable” only
relates to the unique occurence of a term, or the term name. For
instance, the expression `x + poly(x, 2)`

has only the
variable `x`

.

Terms themselves consist of variable and factor names separated by
operators, or involve arithmetic expressions. For instance, the
expression `x + poly(x, 2)`

has *one* variable
`x`

, but *two* terms `x`

and
`poly(x, 2)`

.

**random slopes**: variables that are specified as random slopes in a mixed effects model.**random or grouping factors**: variables that are specified as grouping variables in a mixed effects model.

*Aren’t the predictors, terms and parameters the same
thing?*

In some cases, yes. But not in all cases. Find out more by **clicking
here to access the documentation**.

The package revolves around two key prefixes: `get_*`

and
`find_*`

. The `get_*`

prefix extracts
*values* (or *data*) associated with model-specific
objects (e.g., parameters or variables), while the `find_*`

prefix *lists* model-specific objects (e.g., priors or
predictors). These are powerful families of functions allowing for great
flexibility in use, whether at a high, descriptive level
(`find_*`

) or narrower level of statistical inspection and
reporting (`get_*`

).

In total, the **insight** package includes 16 core
functions: get_data(),
get_priors(),
get_variance(),
get_parameters(),
get_predictors(),
get_random(),
get_response(),
find_algorithm(),
find_formula(),
find_variables(),
find_terms(),
find_parameters(),
find_predictors(),
find_random(),
find_response(),
and model_info().
In all cases, users must supply at a minimum, the name of the model fit
object. In several functions, there are additional arguments that allow
for more targeted returns of model information. For example, the
`find_terms()`

function’s `effects`

argument
allows for the extraction of “fixed effects” terms, “random effects”
terms, or by default, “all” terms in the model object. We point users to
the package documentation or the complementary package website, https://easystats.github.io/insight/, for a detailed
list of the arguments associated with each function as well as the
returned values from each function.

We now would like to provide examples of use cases of the
**insight** package. These examples probably do not cover
typical real-world problems, but serve as illustration of the core idea
of this package: The unified interface to access model information.
**insight** should help both users and package developers
in order to reduce the hassle with the many exceptions from various
modelling packages when accessing model information.

Say, the goal is to make predictions for a certain term, holding
remaining co-variates constant. This is achieved by calling
`predict()`

and feeding the `newdata`

-argument
with the values of the term of interest as well as the “constant” values
for remaining co-variates. The functions `get_data()`

and
`find_predictors()`

are used to get this information, which
then can be used in the call to `predict()`

.

In this example, we fit a simple linear model, but it could be replaced by (m)any other models, so this approach is “universal” and applies to many different model objects.

```
library(insight)
<- lm(
m ~ Species + Petal.Width + Sepal.Width,
Sepal.Length data = iris
)
<- get_data(m)
dat <- find_predictors(m, flatten = TRUE)
pred
<- lapply(pred, function(x) {
l if (is.numeric(dat[[x]])) {
mean(dat[[x]])
else {
} unique(dat[[x]])
}
})
names(l) <- pred
<- as.data.frame(l)
l
cbind(l, predictions = predict(m, newdata = l))
#> Species Petal.Width Sepal.Width predictions
#> 1 setosa 1.2 3.1 5.1
#> 2 versicolor 1.2 3.1 6.1
#> 3 virginica 1.2 3.1 6.3
```

The next example should emphasize the possibilities to generalize
functions to many different model objects using
**insight**. The aim is simply to print coefficients in a
complete, human readable sentence.

The first approach uses the functions that are available for some, but obviously not for all models, to access the information about model coefficients.

```
<- function(model) {
print_params paste0(
"My parameters are ",
toString(row.names(summary(model)$coefficients)),
", thank you for your attention!"
)
}
<- lm(Sepal.Length ~ Petal.Width, data = iris)
m1 print_params(m1)
#> [1] "My parameters are (Intercept), Petal.Width, thank you for your attention!"
# obviously, something is missing in the output
<- mgcv::gam(Sepal.Length ~ Petal.Width + s(Petal.Length), data = iris)
m2 print_params(m2)
#> [1] "My parameters are , thank you for your attention!"
```

As we can see, the function fails for *gam*-models. As the
access to models depends on the type of the model in the R ecosystem, we
would need to create specific functions for all models types. With
**insight**, users can write a function without having to
worry about the model type.

```
<- function(model) {
print_params paste0(
"My parameters are ",
toString(insight::find_parameters(model, flatten = TRUE)),
", thank you for your attention!"
)
}
<- lm(Sepal.Length ~ Petal.Width, data = iris)
m1 print_params(m1)
#> [1] "My parameters are (Intercept), Petal.Width, thank you for your attention!"
<- mgcv::gam(Sepal.Length ~ Petal.Width + s(Petal.Length), data = iris)
m2 print_params(m2)
#> [1] "My parameters are (Intercept), Petal.Width, s(Petal.Length), thank you for your attention!"
```

In case you want to file an issue or contribute in another way to the package, please follow this guide. For questions about the functionality, you may either contact us via email or also file an issue.

Currently, 221 model classes are supported.

```
supported_models()
#> [1] "aareg" "afex_aov"
#> [3] "AKP" "Anova.mlm"
#> [5] "anova.rms" "aov"
#> [7] "aovlist" "Arima"
#> [9] "averaging" "bamlss"
#> [11] "bamlss.frame" "bayesQR"
#> [13] "bayesx" "BBmm"
#> [15] "BBreg" "bcplm"
#> [17] "betamfx" "betaor"
#> [19] "betareg" "BFBayesFactor"
#> [21] "bfsl" "BGGM"
#> [23] "bife" "bifeAPEs"
#> [25] "bigglm" "biglm"
#> [27] "blavaan" "blrm"
#> [29] "bracl" "brglm"
#> [31] "brmsfit" "brmultinom"
#> [33] "btergm" "censReg"
#> [35] "cgam" "cgamm"
#> [37] "cglm" "clm"
#> [39] "clm2" "clmm"
#> [41] "clmm2" "clogit"
#> [43] "coeftest" "complmrob"
#> [45] "confusionMatrix" "coxme"
#> [47] "coxph" "coxph.penal"
#> [49] "coxr" "cpglm"
#> [51] "cpglmm" "crch"
#> [53] "crq" "crqs"
#> [55] "crr" "dep.effect"
#> [57] "DirichletRegModel" "draws"
#> [59] "drc" "eglm"
#> [61] "elm" "epi.2by2"
#> [63] "ergm" "feglm"
#> [65] "feis" "felm"
#> [67] "fitdistr" "fixest"
#> [69] "flac" "flexsurvreg"
#> [71] "flic" "gam"
#> [73] "Gam" "gamlss"
#> [75] "gamm" "gamm4"
#> [77] "garch" "gbm"
#> [79] "gee" "geeglm"
#> [81] "glht" "glimML"
#> [83] "glm" "Glm"
#> [85] "glmm" "glmmadmb"
#> [87] "glmmPQL" "glmmTMB"
#> [89] "glmrob" "glmRob"
#> [91] "glmx" "gls"
#> [93] "gmnl" "HLfit"
#> [95] "htest" "hurdle"
#> [97] "iv_robust" "ivFixed"
#> [99] "ivprobit" "ivreg"
#> [101] "lavaan" "lm"
#> [103] "lm_robust" "lme"
#> [105] "lmerMod" "lmerModLmerTest"
#> [107] "lmodel2" "lmrob"
#> [109] "lmRob" "logistf"
#> [111] "logitmfx" "logitor"
#> [113] "logitr" "LORgee"
#> [115] "lqm" "lqmm"
#> [117] "lrm" "manova"
#> [119] "MANOVA" "marginaleffects"
#> [121] "marginaleffects.summary" "margins"
#> [123] "maxLik" "mblogit"
#> [125] "mclogit" "mcmc"
#> [127] "mcmc.list" "MCMCglmm"
#> [129] "mcp1" "mcp12"
#> [131] "mcp2" "med1way"
#> [133] "mediate" "merMod"
#> [135] "merModList" "meta_bma"
#> [137] "meta_fixed" "meta_random"
#> [139] "metaplus" "mhurdle"
#> [141] "mipo" "mira"
#> [143] "mixed" "MixMod"
#> [145] "mixor" "mjoint"
#> [147] "mle" "mle2"
#> [149] "mlm" "mlogit"
#> [151] "mmclogit" "mmlogit"
#> [153] "mmrm" "mmrm_fit"
#> [155] "mmrm_tmb" "model_fit"
#> [157] "multinom" "mvord"
#> [159] "negbinirr" "negbinmfx"
#> [161] "ols" "onesampb"
#> [163] "orm" "pgmm"
#> [165] "plm" "PMCMR"
#> [167] "poissonirr" "poissonmfx"
#> [169] "polr" "probitmfx"
#> [171] "psm" "Rchoice"
#> [173] "ridgelm" "riskRegression"
#> [175] "rjags" "rlm"
#> [177] "rlmerMod" "RM"
#> [179] "rma" "rma.uni"
#> [181] "robmixglm" "robtab"
#> [183] "rq" "rqs"
#> [185] "rqss" "rvar"
#> [187] "Sarlm" "scam"
#> [189] "selection" "sem"
#> [191] "SemiParBIV" "semLm"
#> [193] "semLme" "slm"
#> [195] "speedglm" "speedlm"
#> [197] "stanfit" "stanmvreg"
#> [199] "stanreg" "summary.lm"
#> [201] "survfit" "survreg"
#> [203] "svy_vglm" "svychisq"
#> [205] "svyglm" "svyolr"
#> [207] "t1way" "tobit"
#> [209] "trimcibt" "truncreg"
#> [211] "vgam" "vglm"
#> [213] "wbgee" "wblm"
#> [215] "wbm" "wmcpAKP"
#> [217] "yuen" "yuend"
#> [219] "zcpglm" "zeroinfl"
#> [221] "zerotrunc"
```

**Didn’t find a model?**File an issue and request additional model-support in*insight*!

If this package helped you, please consider citing as follows:

Lüdecke D, Waggoner P, Makowski D. insight: A Unified Interface to Access Information from Model Objects in R. Journal of Open Source Software 2019;4:1412. doi: 10.21105/joss.01412

Please note that the insight project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.