In dtkaplan/MultipleChoice: Support for Multiple Choice Questions in Shiny

library(mosaic)
library(statisticalModeling)
library(MultipleChoice)
tutorial::go_interactive()

Given data and a model design, the computer will find the model function and model values for you. As an example, consider the Current Population Survey data mosaicData::CPS85. Suppose you want to build a model with wage as a response variable and age and sex as explanatory variables incorporated as main terms. Also include the intercept term, as usual.

The two arguments to lm() are:

. the model design, expressed as a formula: wage ~ 1 + age + sex.

. the data to be used: data = CPS85

data(CPS85, package = "mosaicData")
mod1 <- lm( wage ~ 1 + age + sex, data = CPS85)

The mod1 <-... part of the command simply gives the model a name so that you can use it later on. If you construct more than one model, it makes sense to give them different names.

In making a graph of the function, the model values will always be plotted on the vertical axis. But you have a choice of what to put on the horizontal axis. This plot puts the quantitative variable age on the x-axis, and uses color for sex.

fmodel(mod1, ~ age + sex)

You could arrange things the other way as well.

fmodel(mod1, ~ sex + age)

Note that the line in this plot is merely to guide the eye. The sex variable is categorical and so it's meaningless to interpolate between values.

Your task ... re-create each of the above graphs using the fmodel() command. The two arguments to fmodel() are

1. The model object itself, in this case mod1.
2. A formula describing which roles the explanatory variable will play in the plot, e.g. ~ age + sex or ~ sex + age.

b <- 5

# Create a variable a, equal to 5


# Print out a

# Create a variable a, equal to 5
a <- 5

# Print out a
a

test_object("a")
test_output_contains("a", incorrect_msg = "Make sure to print `a`.")
success_msg("Great!")

dtkaplan/MultipleChoice documentation built on May 15, 2019, 4:58 p.m.