# asef: Find associated effects In hohenstein/remef: Remove Partial Effects

## Description

This function finds the labels of coefficients associated with interactions or main effects in the results of various model fitting functions.

## Usage

 `1` ```asef(model, ef, order = c("higher", "lower", "all"), include.base = FALSE) ```

## Arguments

 `model` a model object. `ef` a character string (a coefficent label). `order` a character string vector giving the order of the terms to be returned. One of `"higher"` (default), `"lower"`, or `"all"`, can be abbreviated. `include.base` logical. Should the labels in `ef` be included in the output?

## Details

`model` is an object returned by a model fitting function (e.g., `lm`, `aov`).

The character string passed to `ef` must not include the intercept (`"(Intercept)"`) since it cannot be part of an interaction. All non-intercept labels of any order are allowed.

If `order` is `"higher"`, all terms of a higher order are returned. If `order` is `"lower"`, all terms of a lower order are returned. If `order` is `"all"` then both terms of lower and higher order are returned. For example, if a model fit has the terms `A`, `B`, `C`, `A:B`, `B:C`, `A:C`, and `A:B:C`, terms of higher order relative to `A` are `A:B`, `A:C`, and `A:B:C`. Terms of lower order relative to `A:B` are `A` and `B`.

If `include.base` is `FALSE` (the default), the term label in `ef` is not part of the returned string vector.

## Value

`asef` returns vectors of character strings (coefficent labels).

`terms` for extracting model terms.
 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13``` ```require(utils) data(iris) fit <- lm(Sepal.Length ~ Sepal.Width * Petal.Length * Petal.Width, data = iris) asef(fit, "Petal.Length:Petal.Width", order = "higher") asef(fit, "Sepal.Width:Petal.Length:Petal.Width", order = "lower") asef(fit, "Petal.Length:Petal.Width", order = "all") data(warpbreaks) fit2 <- lm(breaks ~ wool * tension, data = warpbreaks) # Since wool is a factor, the coefficient has another name, "woolB" asef(fit2, "woolB", order = "higher") asef(fit2, "woolB", order = "higher", include.base = TRUE) ```