getPlotSummary {iNZightPlots}

iNZight Plot Summary and Inference

Description

Generate summary or inference information for an iNZight plot

Usage

getPlotSummary(
  x,
  y = NULL,
  g1 = NULL,
  g1.level = NULL,
  g2 = NULL,
  g2.level = NULL,
  varnames = list(),
  colby = NULL,
  sizeby = NULL,
  data = NULL,
  design = NULL,
  freq = NULL,
  missing.info = TRUE,
  inzpars = inzpar(),
  summary.type = "summary",
  table.direction = c("horizontal", "vertical"),
  hypothesis.value = 0,
  hypothesis.alt = c("two.sided", "less", "greater"),
  hypothesis.var.equal = FALSE,
  hypothesis.use.exact = FALSE,
  hypothesis.test = c("default", "t.test", "anova", "chi2", "proportion"),
  hypothesis.simulated.p.value = FALSE,
  hypothesis = list(value = hypothesis.value, alternative = match.arg(hypothesis.alt),
    var.equal = hypothesis.var.equal, use.exact = hypothesis.use.exact, test =
    match.arg(hypothesis.test), simulated.p.value = hypothesis.simulated.p.value),
  survey.options = list(),
  width = 100,
  epi.out = FALSE,
  privacy_controls = NULL,
  html = FALSE,
  ...,
  env = parent.frame()
)

Arguments

x

a vector (numeric or factor), or the name of a column in the supplied data or design object

y

a vector (numeric or factor), or the name of a column in the supplied data or design object

g1

a vector (numeric or factor), or the name of a column in the supplied data or design object. This variable acts as a subsetting variable.

g1.level

the name (or numeric position) of the level of g1 that will be used instead of the entire data set

g2

a vector (numeric or factor), or the name of a column in the supplied data or design object. This variable acts as a subsetting variable, similar to g1

g2.level

same as g1.level, however takes the additional value "_MULTI", which produces a matrix of g1 by g2

varnames

a list of variable names, with the list named using the appropriate arguments (i.e., list(x = "height", g1 = "gender"))

colby

the name of a variable (numeric or factor) to colour points by. In the case of a numeric variable, a continuous colour scale is used, otherwise each level of the factor is assigned a colour

sizeby

the name of a (numeric) variable, which controls the size of points

data

the name of a data set

design

the name of a survey object, obtained from the survey package

freq

the name of a frequency variable if the data are frequencies

missing.info

logical, if TRUE, information regarding missingness is displayed in the plot

inzpars

allows specification of iNZight plotting parameters over multiple plots

summary.type

one of "summary" or "inference"

table.direction

one of 'horizontal' (default) or 'vertical' (useful for many categories)

hypothesis.value

H0 value for hypothesis test

hypothesis.alt

alternative hypothesis (!=, <, >)

hypothesis.var.equal

use equal variance assumption for t-test?

hypothesis.use.exact

logical, if TRUE the exact p-value will be calculated (if applicable)

hypothesis.test

in some cases (currently just two-samples) can perform multiple tests (t-test or ANOVA)

hypothesis.simulated.p.value

also calculate (where available) the simulated p-value

hypothesis

either NULL for no test, or missing (in which case above arguments are used)

survey.options

additional options passed to survey methods

width

width for the output, default is 100 characters

epi.out

logical, if TRUE, then odds/rate ratios and rate differences are printed when appropriate (y with 2 levels)

privacy_controls

optional, pass in confidentialisation and privacy controls (e.g., random rounding, suppression) for microdata

html

logical, it TRUE output will be returned as an HTML page (if supported)

...

additional arguments, see inzpar

env

compatibility argument

Details

Works much the same as iNZightPlot

Value

an inzight.plotsummary object with a print method

Author(s)

Tom Elliott

Examples

getPlotSummary(Species, data = iris)
getPlotSummary(Species, data = iris,
    summary.type = "inference", inference.type = "conf")

# perform hypothesis testing
getPlotSummary(Sepal.Length, data = iris,
    summary.type = "inference", inference.type = "conf",
    hypothesis.value = 5)

# if you prefer a formula interface
inzsummary(Sepal.Length ~ Species, data = iris)
inzinference(Sepal.Length ~ Species, data = iris)

## confidentialisation and privacy controls
# random rounding and suppression:
HairEyeColor_df <- as.data.frame(HairEyeColor)
inzsummary(Hair ~ Eye, data = HairEyeColor_df, freq = Freq)
inzsummary(Hair ~ Eye, data = HairEyeColor_df, freq = Freq,
    privacy_controls = list(
        rounding = "RR3",
        suppression = 10
    )
)