Movatterモバイル変換

Introduction to the 'memisc' Package

Description

This package collects an assortment of tools that are intended to makework withR easier for the author of this packageand are submitted to the public in the hope that they will be also be useful to others.

The tools in this package can be grouped into four major categories:

• Data preparation and management

• Data analysis

• Presentation of analysis results

• Programming

Data preparation and management

Survey Items

memisc provides facilities to work with what users from otherpackages like SPSS, SAS, or Stata know as ‘variable labels’, ‘value labels’and ‘user-defined missing values’. In the context of this package theseaspects of the data are represented by the"description","labels", and"missing.values" attributes of adata vector. These facilities are useful, for example, if you work withsurvey data that contain coded items like vote intention thatmay have the following structure:

Question: “If there was a parliamentary election next tuesday, which party would you vote for?”

A statistical package like SPSS allows toattach labels like ‘Conservative Party’, ‘Labour Party’, etc.to the codes 1,2,3, etc. and to markmark the codes 96, 97, 98, 99as ‘missing’ and thus to exclude these variables from statisticalanalyses.memisc provides similar facilities.Labels can be attached to codes by calls likelabels(x) <- somethingand expendanded by calls likelabels(x) <-labels(x) + something,codes can be marked as ‘missing’ bycalls likemissing.values(x) <- something andmissing.values(x) <-missing.values(x) + something.

memisc defines a class called "data.set", which is similar to the class "data.frame".The main difference is that it is especially geared toward containing survey item data.Transformations of and within "data.set" objects retain the information aboutvalue labels, missing values etc. Usingas.data.frame sets the data up forR's statistical functions, but doing this explicitely is seldom necessary.Seedata.set.

More Convenient Import of External Data

Survey data sets are often relative large and contain up to a few thousand variables.For specific analyses one needs however only a relatively small subset of these variables.Although modern computers have enough RAM to load such data sets completely into an R session,this is not very efficient having to drop most of the variables after loading. Also, loadingsuch a large data set completely can be time-consuming, because R has to allocate space foreach of the many variables. Loading just the subset of variables really needed for an analysisis more efficient and convenient - it tends to be much quicker. Thus this package providesfacilities to load such subsets of variables, without the need to load a complete data set.Further, the loading of data from SPSS files is organized in such a way that all informationsabout variable labels, value labels, and user-defined missing values are retained.This is made possible by the definition ofimporter objects, for whichasubset method exists.importer objects contain onlythe information about the variables in the external data set but not the data.The data itself is loaded into memory when the functionssubset oras.data.setare used.

Recoding

memisc also contains facilities for recodingsurvey items. Simple recodings, for example collapsing answercategories, can be done using the functionrecode. Morecomplex recodings, for example the construction of indices frommultiple items, and complex case distinctions, can be doneusing the functioncases. This function may alsobe useful for programming, in so far as it is a generalization ofifelse.

Code Books

There is a functioncodebook which produces a code book of anexternal data set or an internal "data.set" object. A codebook contains in aconveniently formatted way concise information about every variable in a data set,such as which value labels and missing values are defined and some univariate statistics.

An extended example of all these facilities is contained in the vignette "anes48",and indemo(anes48)

Data Analysis

Tables and Data Frames of Descriptive Statistics

genTable is a generalization ofxtabs:Instead of counts, also descriptive statistics like means or variancescan be reported conditional on levels of factors. Also conditionalpercentages of a factor can be obtained using this function.

In addition anAggregate function is provided, which has the same syntax asgenTable, butgives a data frame of descriptive statistics instead of atableobject.

Per-Subset Analysis

By is a variant of thestandard functionby: Conditioning factorsare specified by a formula and areobtained from the data frame the subsets of which are to be analysed.Therefore there is no need toattach the data frameor to use the dollar operator.

Presentation of Results of Statistical Analysis

Publication-Ready Tables of Coefficients

Journals of the Political and Social Sciences usually requirethat estimates of regression models are presented in the followingform:

    ==================================================                    Model 1     Model 2     Model 3    --------------------------------------------------    Coefficients    (Intercept)     30.628***    6.360***   28.566***                    (7.409)     (1.252)     (7.355)    pop15           -0.471**                -0.461**                    (0.147)                 (0.145)    pop75           -1.934                  -1.691                    (1.041)                 (1.084)    dpi                          0.001      -0.000                                (0.001)     (0.001)    ddpi                         0.529*      0.410*                                (0.210)     (0.196)    --------------------------------------------------    Summaries    R-squared         0.262       0.162       0.338    adj. R-squared    0.230       0.126       0.280    N                50          50          50    ==================================================

Such tables of coefficient estimates can be producedbymtable. To see some of the possibilities ofthis function, useexample(mtable).

LaTeX Representation of R Objects

Output produced bymtable can be transformed intoLaTeX tables by an appropriate method of the generic functiontoLatex which is defined in the packageutils. In addition,memisc definestoLatex methodsfor matrices andftable objects. Note thatresults produced bygenTable can be coerced intoftable objects. Also, a default methodfor thetoLatex function is defined which coerces itsargument to a matrix and applies the matrix method oftoLatex.

Programming

Looping over Variables

Sometimes users want to contruct loops that run over variables rather than values.For example, if one wants to set the missing values of a battery of items.For this purpose, the package contains the functionforeach.To set 8 and 9 as missing values for the itemsknowledge1,knowledge2,knowledge3, one can use

    foreach(x=c(knowledge1,knowledge2,knowledge3),        missing.values(x) <- 8:9)

Changing Names of Objects and Labels of Factors

R already makes it possible to change the names of an object.Substituting thenames ordimnamescan be done with some programming tricks. This package definesthe functionrename,dimrename,colrename, androwrenamethat implement these tricks in a convenient way, so that programmers(like the author of this package) need not reinvent the weel inevery instance of changing names of an object.

Dimension-Preserving Versions oflapply andsapply

If a function that is involved in a call tosapply returns a result an array or a matrix, thedimensional information gets lost. Also, if a list object to whichlapply orsapply are appliedhave a dimension attribute, the result looses this information.The functionsLapply andSapply defined in this package preserve suchdimensional information.

Combining Vectors and Arrays by Names

The generic functioncollect collects several objects of thesame mode into one object, using their names,rownames,colnames and/ordimnames. There are methods foratomic vectors, arrays (including matrices), and data frames.For example

  a <- c(a=1,b=2)  b <- c(a=10,c=30)  collect(a,b)

leads to

     x  y  a  1 10  b  2 NA  c NA 30

Reordering of Matrices and Arrays

Thememisc package includes areordermethod for arrays and matrices. For example, the matrixmethod by default reorders the rows of a matrix according the resultsof a function.

Conditional Evaluation of an Expression

Description

The functionBy evaluates an expression within subsets ofa data frame, where the subsets are defined by a formula.

Usage

By(formula,expr,data=parent.frame())

Arguments

Value

A list of class "by", giving the results for each combination of valuesof variables informula.

Examples

berkeley <- Aggregate(Table(Admit,Freq)~.,data=UCBAdmissions)(Bres <- By(~Dept,glm(cbind(Admitted,Rejected)~Gender,family="binomial"),data=berkeley))# The results all have 'data' componentsstr(Bres[[1]]$data)attach(berkeley)(Bres <- By(~Dept,glm(cbind(Admitted,Rejected)~Gender,family="binomial")))detach(berkeley)

Vectors of Univariate Sample Statistics

Description

Descriptives(x) gives a vector of sample statisticsfor use incodebook.

Usage

Descriptives(x,...)## S4 method for signature 'atomic'Descriptives(x, weights = NULL, ...)## S4 method for signature 'item.vector'Descriptives(x, weights = NULL, ...)

Arguments

Value

A numeric vector of sample statistics, containing the range, themean, the standard deviation, the skewness and the (excess) kurtosis.

Examples

x <- rnorm(100)Descriptives(x)

Operate on grouped data in data frames and data sets

Description

Group creates a grouped variant of an object ofclass "data.frame" or of class "data.set", for which methods forwith andwithin are defined, so that these well-knownfunctions can be applied "groupwise".

Usage

# Create an object of class "grouped.data" from a# data frame or a data set.Groups(data,by,...)## S3 method for class 'data.frame'Groups(data,by,...)## S3 method for class 'data.set'Groups(data,by,...)## S3 method for class 'grouped.data'Groups(data,by,...)# Recombine grouped data into a data fame or a data setrecombine(x,...)## S3 method for class 'grouped.data.frame'recombine(x,...)## S3 method for class 'grouped.data.set'recombine(x,...)# Recombine grouped data and coerce the result appropriately:## S3 method for class 'grouped.data'as.data.frame(x,...)## S4 method for signature 'grouped.data.frame'as.data.set(x,row.names=NULL,...)## S4 method for signature 'grouped.data.set'as.data.set(x,row.names=NULL,...)# Methods of the generics "with" and "within" for grouped data## S3 method for class 'grouped.data'with(data,expr,...)## S3 method for class 'grouped.data'within(data,expr,recombine=FALSE,...)# This is equivalent to with(Groups(data,by),expr,...)withGroups(data,by,expr,...)# This is equivalent to within(Groups(data,by),expr,recombine,...)withinGroups(data,by,expr,recombine=TRUE,...)

Arguments

Details

When applied to a data frameGroups returns an object with class attributes"grouped.data.frame", "grouped.data", and "data.frame", when applied do an object with class"data.set", it returns an object with class attributes "grouped.data.set","grouped.data", and "data.set".

When applied to objects with class attributed"grouped.data", both the functionswith() amdwithin()evaluateexpr separately for each group defined byGroups.with() returns an array composed of the resultsofexpr, whilewithin() returns a modified copy of itsdata argument, which will be a "grouped.data" object("grouped.data.frame" or "grouped.data.set"), unless the argumentrecombine=TRUE is set.

The expressionexpr may contain references to the variablesn_,N_, andi_.n_ is equal to the size ofthe respective group (the number of rows belonging to it), whileN_ is equal to the total number of observations in allgroups. The variablei_ equals to the indices of the rowsbelonging to the respective group of observations.

Examples

some.data <- data.frame(x=rnorm(n=100))some.data <- within(some.data,{    f <- factor(rep(1:4,each=25),labels=letters[1:4])    g <- factor(rep(1:5,each=4,5),labels=LETTERS[1:5])    y <- x + rep(1:4,each=25) +  0.75*rep(1:5,each=4,5)})# For demonstration purposes, we create an# 'empty' group:some.data <- subset(some.data,                       f!="a" | g!="C")some.grouped.data <- Groups(some.data,                           ~f+g)    # Computing the means of y for each combination f and ggroup.means <- with(some.grouped.data,                    mean(y))group.means# Obtaining a groupwise centered variant of ysome.grouped.data <- within(some.grouped.data,{    y.cent <- y - mean(y)},recombine=FALSE)# The groupwise centered variable should have zero mean# whithin each groupgroup.means <- with(some.grouped.data,                    round(mean(y.cent),15))group.means# The following demonstrates the use of n_, N_, and i_# An external copy of yy1 <- some.data$ygroup.means.n <- with(some.grouped.data,                      c(mean(y),  # Group means for y                        n_,       # Group sizes                        sum(y)/n_,# Group means for y                        n_/N_,    # Relative group sizes                        sum(y1)/N_,# NOT the grand mean                        sum(y1[i_])/n_)) # Group mean for y1group.means.n# Names can be attached to the groupwise resultswith(some.grouped.data,     c(Centered=round(mean(y.cent),15),       Uncentered=mean(y)))some.data.ungrouped <- recombine(some.grouped.data)str(some.data.ungrouped)# It all works with "data.set" objectssome.dataset <- as.data.set(some.data)some.grouped.dataset <- Groups(some.dataset,~f+g)with(some.grouped.dataset,     c(Mean=mean(y),       Variance=var(y)))# The following two expressions are equivalent:with(Groups(some.data,~f+g),mean(y))withGroups(some.data,~f+g,mean(y))# The following two expressions are equivalent:some.data <- within(Groups(some.data,~f+g),{    y.cent <- y - mean(y)    y.cent.1 <- y - sum(y)/n_})some.data <- withinGroups(some.data,~f+g,{    y.cent <- y - mean(y)    y.cent.1 <- y - sum(y)/n_})# Both variants of groupwise centred varaibles should# have zero groupwise means:withGroups(some.data,~f+g,{    c(round(mean(y.cent),15),      round(mean(y.cent.1),15))})

Convert Annotations, and Value Labels between Encodings

Description

This function uses the base package functioniconvto translate variable descriptions (a.k.a variable labels) andvalue labels ofitem,data.set,andimporter objects into a specified encoding.

It will be useful in UTF-8 systems when data file come in some ancientencoding like 'Latin-1' as long used by Windows systems.

Usage

Iconv(x,from="",to="",...) ## S3 method for class 'character'Iconv(x,from="",to="",...) ## S3 method for class 'annotation'Iconv(x,from="",to="",...) ## S3 method for class 'data.set'Iconv(x,from="",to="",...) ## S3 method for class 'importer'Iconv(x,from="",to="",...) ## S3 method for class 'item'Iconv(x,from="",to="",...) ## S3 method for class 'value.labels'Iconv(x,from="",to="",...)

Arguments

Value

Iconv returns a copy of its first argument with re-encoded character data or attributes.

See Also

iconv,iconvlist

Examples

## Not run: # Locate an SPSS 'system' file and get info on variables, their labels etc.ZA5302 <- spss.system.file("Daten/ZA5302_v6-0-0.sav",to.lower=FALSE)# Convert labels etc. from 'latin1' to the encoding of the current locale.ZA5302 <- Iconv(ZA5302,from="latin1")# Write out the codebookwriteLines(as.character(codebook(ZA5302)),           con="ZA5302-cdbk.txt")# Write out the description of the varialbes (their 'variable labels')writeLines(as.character(description(ZA5302)),            con="ZA5302-description.txt")## End(Not run)

Create a list and conveniently supply names to its elements

Description

List creates a list and names its elements after thearguments given, in a manner analogously todata.frame

Usage

  List(...)

Arguments

Examples

  num <- 1:3  strng <- c("a","b","A","B")  logi <- rep(FALSE,7)  List(num,strng,logi)

Convenience wrappers for common statistical functions

Description

Mean(),Median(), etc. are mere wrappers ofthe functionsmean(),median(), etc. with thena.rm= optional argument setTRUE by default.

Usage

Mean(x, na.rm=TRUE, ...)Median(x, na.rm=TRUE, ...)Min(x, na.rm=TRUE, ...)Max(x, na.rm=TRUE, ...)Weighted.Mean(x, w, ..., na.rm = TRUE)Var(x, na.rm=TRUE, ...)StdDev(x, na.rm=TRUE, ...)Cov(x, y = NULL, use = "pairwise.complete.obs", ...)Cor(x, y = NULL, use = "pairwise.complete.obs", ...)Range(..., na.rm = TRUE, finite = FALSE)

Arguments

Means for groups of observations

Description

The functionMeans() creates a table of groupmeans, optionally with standard errors, confidence intervals, andnumbers of valid observations.

Usage

Means(data, ...)## S3 method for class 'data.frame'Means(data,    by, weights=NULL, subset=NULL,    default=NA,    se=FALSE, ci=FALSE, ci.level=.95,    counts=FALSE, ...)## S3 method for class 'formula'Means(data, subset, weights, ...)## S3 method for class 'numeric'Means(data, ...)## S3 method for class 'means.table'as.data.frame(x, row.names=NULL, optional=TRUE, drop=TRUE, ...)## S3 method for class 'xmeans.table'as.data.frame(x, row.names=NULL, optional=TRUE, drop=TRUE, ...)

Arguments

Value

An array that inherits classes "means.table" and "table". IfMeans was called withse=TRUE orci=TRUEthen the result additionally inherits class "xmeans.table".

Examples

# Preparing example dataUSstates <- as.data.frame(state.x77)USstates <- within(USstates,{    region <- state.region    name <- state.name    abb <- state.abb    division <- state.division})USstates$w <- sample(runif(n=6),size=nrow(USstates),replace=TRUE)# Using the data frame methodMeans(USstates[c("Murder","division","region")],by=c("division","region"))Means(USstates[c("Murder","division","region")],by=USstates[c("division","region")])Means(USstates[c("Murder")],1)Means(USstates[c("Murder","region")],by=c("region"))# Using the formula method# One 'dependent' variableMeans(Murder~1, data=USstates)Means(Murder~division, data=USstates)Means(Murder~division, data=USstates,weights=w)Means(Murder~division+region, data=USstates)as.data.frame(Means(Murder~division+region, data=USstates))# Standard errors and countsMeans(Murder~division, data=USstates, se=TRUE, counts=TRUE)drop(Means(Murder~division, data=USstates, se=TRUE, counts=TRUE))as.data.frame(Means(Murder~division, data=USstates, se=TRUE, counts=TRUE))# Confidence intervalsMeans(Murder~division, data=USstates, ci=TRUE)drop(Means(Murder~division, data=USstates, ci=TRUE))as.data.frame(Means(Murder~division, data=USstates, ci=TRUE))# More than one dependent variableMeans(Murder+Illiteracy~division, data=USstates)as.data.frame(Means(Murder+Illiteracy~division, data=USstates))# Confidence intervalsMeans(Murder+Illiteracy~division, data=USstates, ci=TRUE)as.data.frame(Means(Murder+Illiteracy~division, data=USstates, ci=TRUE))# Some 'non-standard' but still valid usages:with(USstates,     Means(Murder~division+region,subset=region!="Northeast"))with(USstates,     Means(Murder,by=list(division,region)))

Reshape data frames or data sets

Description

Reshape is a conveniencewrapper aroundreshape with a somewhat simplersyntax.

Usage

Reshape(data,...,id,within_id,drop,keep,direction)

Arguments

Examples

example.data.wide <- data.frame(    v  = c(35,42),    x1 = c(1.1,2.1),    x2 = c(1.2,2.2),    x3 = c(1.3,2.3),    x4 = c(1.4,2.4),    y1 = c(2.5,3.5),    y2 = c(2.7,3.7),    y3 = c(2.9,3.9))example.data.wide# The following two calls are equivalent:example.data.long <- Reshape(data=example.data.wide,                             x=c(x1,x2,x3,x4),                             # N.B. it is possible to                             # specify 'empty' i.e. missing                             # measurements                             y=c(y1,y2,y3,),                             t=1:4,                             direction="long")example.data.long <- Reshape(data=example.data.wide,                             list(                                 x=c(x1,x2,x3,x4),                                 # N.B. it is possible to                                 # specify 'empty' i.e. missing                                 # measurements                                 y=c(y1,y2,y3,)                             ),                             t=1:4,                             direction="long")example.data.long# Since the data frame contains an "reshapeLong" attribute# an id variable is already specified and part of the data# frame.example.data.wide <- Reshape(data=example.data.long,                             x=c(x1,x2,x3,x4),                             y=c(y1,y2,y3,),                             t=1:4,                             direction="wide")example.data.wide# Here we examine the case where no "reshapeLong" attribute# is present:example.data.wide <- Reshape(data=example.data.long,                             x=c(x1,x2,x3,x4),                             y=c(y1,y2,y3,),                             t=1:4,                             id=v,                             direction="wide")example.data.wide# Here, an "automatic" time variable is created. This works# only if there is a single argument other than the data=# and direction= argumentsexample.data.long <- Reshape(data=example.data.wide,                             list(                                 x=c(x1,x2,x3,x4),                                 y=c(y1,y2,y3,)                             ),                             direction="long")example.data.longexample.data.wide <- Reshape(data=example.data.long,                             list(                                 x=c(x1,x2,x3,x4),                                 y=c(y1,y2,y3,)                             ),                             direction="wide")example.data.wide

A Dimension Preserving Variant of "sapply" and "lapply"

Description

Sapply is equivalent tosapply, exceptthat it preserves the dimension and dimension names of theargumentX. It also preserves the dimension ofresults of the functionFUN.It is intended for application to results e.g.of a call toby.Lapply is an analogtolapply insofar as it does not try to simplifythe resultinglist of results ofFUN.

Usage

Sapply(X, FUN, ..., simplify = TRUE, USE.NAMES = TRUE)Lapply(X, FUN, ...)

Arguments

Value

IfFUN returns a scalar, then the result has the same dimensionasX, otherwise the dimension of the result is enhanced relativetoX.

Examples

berkeley <- Aggregate(Table(Admit,Freq)~.,data=UCBAdmissions)berktest1 <- By(~Dept+Gender,                glm(cbind(Admitted,Rejected)~1,family="binomial"),                data=berkeley)berktest2 <- By(~Dept,                glm(cbind(Admitted,Rejected)~Gender,family="binomial"),                data=berkeley)sapply(berktest1,coef)Sapply(berktest1,coef)sapply(berktest1,function(x)drop(coef(summary(x))))Sapply(berktest1,function(x)drop(coef(summary(x))))sapply(berktest2,coef)Sapply(berktest2,coef)sapply(berktest2,function(x)coef(summary(x)))Sapply(berktest2,function(x)coef(summary(x)))

Substitutions in Language Objects

Description

Substitute differs fromsubstitutein so far as its first argument can be a variable thatcontains an object of mode "language". In that case,substitutions take place inside this object.

Usage

Substitute(lang,with)

Arguments

Details

The function body is justdo.call("substitute",list(lang,with)).

Value

An object of storage mode "language" or "symbol".

Examples

lang <- quote(sin(x)+z)substitute(lang,list(x=1,z=2))Substitute(lang,list(x=1,z=2))

One-Dimensional Table of Frequences and/or Percentages

Description

Table is a generic function thatproduces a table of counts or weighted countsand/or the corresponding percentages of an atomic vector,factor or"item.vector" object.This function is intended for use withAggregate orgenTable.The"item.vector" method is the workhorseofcodebook.

Usage

## S4 method for signature 'atomic'Table(x,weights=NULL,counts=TRUE,percentage=FALSE,...)## S4 method for signature 'factor'Table(x,weights=NULL,counts=TRUE,percentage=FALSE,...)## S4 method for signature 'item.vector'Table(x,weights=NULL,counts=TRUE,percentage=(style=="codebook"),              style=c("table","codebook","nolabels"),              include.missings=(style=="codebook"),              missing.marker=if(style=="codebook") "M" else "*",...)

Arguments

Value

The atomic vector and factor methods return either a vectorof counts or vector of percentages or a matrix of counts and percentages.The same applies to the"item.vector" vector method unlessinclude.missing=TRUE andpercentage=TRUE,in which case total percentages and percentages of valid valuesare given.

Examples

  with(as.data.frame(UCBAdmissions),Table(Admit,Freq))  Aggregate(Table(Admit,Freq)~.,data=UCBAdmissions)  A <- sample(c(1:5,9),size=100,replace=TRUE)  labels(A) <- c(a=1,b=2,c=3,d=4,e=5,dk=9)  missing.values(A) <- 9  Table(A,percentage=TRUE)

Named Lists, Lists of Items, and Atomic Vectors

Description

The classes "named.list" and "item.list" are merely some'helper classes' for the construction of the classes "data.set"and "importer".

Class "named.list" extends the basic class "list" by an additionalslot "names". Itsinitialize method assures that the namesof the list are unique.

Class "item.list" extends the class "named.list", but does notadd any slots. From "named.list" it differs only by theinitialize method, which calls that for "named.list"and makes sure that all elements of the list belong to class "item".

Classes "atomic" and "double" are merely used formethod selection.

Examples

new("named.list",a=1,b=2)# This should generate an error, since the names# are not unique.try(new("named.list",a=1,a=2))# Another error, one name is missing.try(new("named.list",a=1,2))# Also an error, the resulting list would be unnamed.try(new("named.list",1,2))new("item.list",a=1,b=2)# Also an error: "item.list"s are "named.lists",# and here the names would be non-unique.try(new("item.list",a=1,a=2))

Write Codebooks and Variable Descriptions into a Text File

Description

This is a convenience function to facilitate the creation of data set documentsin text files.

Usage

Write(x,...)## S3 method for class 'codebook'Write(x,file=stdout(),...)## S3 method for class 'descriptions'Write(x,file=stdout(),...)

Arguments

Adding Annotations to Objects

Description

Annotations, that is, objects of class"annotation",are character vectors with all their elements named.Only one method is defined for this subclass of character vectors,a method forshow, that shows the annotation ina nicely formatted way. Annotations of an object can be obtainedvia the functionannotation(x) and can be set viaannotation(x)<-value.

Elements of an annotation with names"description"and"wording" have a special meaning.The first kind can be obtained and set viadescription(x) anddescription(x)<-value,the second kind can be obtained viawording(x) andwording(x)<-value."description" elements are used in way the "variable labels"are used in SPSS and Stata."wording" elements of annotationobjects are meant to contain the question wording of a questionnaireitem represented by an"item" objects.These elements of annotations are treated in a special wayin the output of thecoodbook function.

Usage

annotation(x)## S4 method for signature 'ANY'annotation(x)## S4 method for signature 'item'annotation(x)## S4 method for signature 'data.set'annotation(x)annotation(x)<-value## S4 replacement method for signature 'ANY,character'annotation(x)<-value## S4 replacement method for signature 'ANY,annotation'annotation(x)<-value## S4 replacement method for signature 'item,annotation'annotation(x)<-value## S4 replacement method for signature 'vector,annotation'annotation(x)<-valuedescription(x)description(x)<-valuewording(x)wording(x)<-value## S4 method for signature 'data.set'description(x)## S4 method for signature 'importer'description(x)## S4 method for signature 'data.frame'description(x)## S4 method for signature 'tbl_df'description(x)

Arguments

Value

annotation(x) returns an object of class"annotation",which is a named character.description(x) andwording(x) each usually return a character string.Ifdescription(x) is applied to adata.set or animporter object,however, a character vector is returned, which is named after thevariables in the data set or the external file.

Examples

vote <- sample(c(1,2,3,8,9,97,99),size=30,replace=TRUE)labels(vote) <- c(Conservatives         =  1,                    Labour                =  2,                    "Liberal Democrats"   =  3,                    "Don't know"          =  8,                    "Answer refused"      =  9,                    "Not applicable"      = 97,                    "Not asked in survey" = 99                    )missing.values(vote) <- c(97,99)description(vote) <- "Vote intention"wording(vote) <- "If a general election would take place next tuesday,                    the candidate of which party would you vote for?"annotation(vote)annotation(vote)["Remark"] <- "This is not a real questionnaire item, of course ..."codebook(vote)

Apply a Formatting Template to a Numeric or Character Vector

Description

applyTemplate is called internally bymtableto format coefficients and summary statistics.

Usage

applyTemplate(x,template,float.style=getOption("float.style"),                      digits=min(3,getOption("digits")),                      signif.symbols=getOption("signif.symbols"))

Arguments

Details

Character vectors that are used as templates may be arbitrary. However,certain character sequences may formtemplate expressions.A template expression is of the form($<POS>:<Format spec>),where "($" indicates the start of a template expression,"<POS>" stands for either an index or name that selects anelement fromx and "<Format spec>" stands for aformat specifier. It may contain an letter indicating thestyle in which the vector element selected by<POS>will be formatted byformatC, it may containa number as the number of significant digits, a "#"indicating that the number of signifcant digits will be at most that givenbygetOption("digits"), or* that means thatthe value will be formatted as a significance symbol.

Value

applyTemplate returns a character vector in which templateexpressions intemplate are substituted by formatted values fromx.Iftemplate is an array then the return value is also an array ofthe same shape.

Examples

applyTemplate(c(a=.0000000000000304,b=3),template=c("($1:g7#)($a:*)"," (($1:f2)) "))applyTemplate(c(a=.0000000000000304,b=3),template=c("($a:g7#)($a:*)"," (($b:f2)) "))

Converting Data Frames into Arrays

Description

Theas.array for data framestakes all factors in a data frame and uses themto define the dimensions of the resulting array,and fills the array with the values ofthe remaining numeric variables.

Currently, the data frame must contain allcombinations of factor levels.

Usage

## S4 method for signature 'data.frame'as.array(x,data.name=NULL,...)

Arguments

Value

An array

Examples

BerkeleyAdmissions <- to.data.frame(UCBAdmissions)BerkeleyAdmissionsas.array(BerkeleyAdmissions,data.name="Admit")try(as.array(BerkeleyAdmissions[-1,],data.name="Admit"))

Construction of Lists of Symbols

Description

as.symbols andsyms are functions potentially usefulin connection withforeach andxapply.as.symbols produces a list of symbols from a character vector,whilesyms returns a list of symbols from symbols given as arguments,but it can be used to construct patterns of symbols.

Usage

as.symbols(x)syms(...,paste=FALSE,sep="")

Arguments

Value

A list of language symbols (results ofas.symbol - not graphicalsymbols!).

Examples

  as.symbols(letters[1:8])  syms("a",1:3,paste=TRUE)  sapply(syms("a",1:3,paste=TRUE),typeof)

Assign a values to a variable for instances where a condition ismet

Description

The%if% operator allows to assign values to a variable only ifa condition is met i.e. results inTRUE. It is supposed tobe used similar to thereplace ... if construct in Stata.

Usage

expr %if% condition# For example# (variable <- value) %if% (other_variable == 0)

Arguments

Details

The 'value' that is assigned to the variable inexprshould either be a scalar, a vector with as many elements as thecondition vector has, or as many elements as the number of elementsin the condition vector that are equal (or evaluate to)TRUE.

Examples

(test_var <- 1) %if% (1:7 > 3)test_var(test_var <- 2) %if% (1:7 <= 3)test_var(test_var <- 100*test_var) %if% (1:7%%2==0)test_var# This creates a warning about non-matching lengths.(test_var <- 500:501) %if% (1:7 <= 3)test_var(test_var <- 501:503) %if% (1:7 <= 3)test_var(test_var <- 401:407) %if% (1:7 <= 3)test_var

Operators for Setting Annotations and Attributes

Description

The operator%#% can be used to attach adescription annotation to an object.%##% can beused to attach a character vector of annotations to an object.%@% returns the attribute with the name given as secondargument. With%@% it is also possible to assign attributes.

Usage

  x %#% descr  x %##% annot  x %@% nm  x %@% nm <- value

Arguments

Examples

test1 <- 1 %#% "One"# This is equivalent to:# test <- 1# description(test) <- "One"description(test1)# Results in "One"# Not that it makes sense, but ...test2 <- 2 %##% c(                    Precedessor = 0,                    Successor   = 2                 )# This is equivalent to:# test2 <- 2# annotation(test2) <- c(#                    Precedessor = 0,#                    Successor   = 2#                 )annotation(test2)# The following examples are equivalent to# attr(test2,"annotation")test2 %@% annotationtest2 %@% "annotation"test2 %@% another.attribute <- 42# This is equivalent to attr(test2,"another.attribute") <- 42attributes(test2)

Distinguish between Cases Specified by Logical Conditions

Description

cases allows to distinguish several cases defined logicalconditions. It can be used to code these cases into a vector. Thefunction can be considered as a multi-condition generalization ofifelse.

Usage

cases(...,check.xor=c("warn","stop","ignore"),      .default=NA,.complete=FALSE,      check.na=c("warn","stop","ignore"),      na.rm=TRUE)

Arguments

Details

There are two distinct ways to use this function. Either thefunction can be used to construct a factor that representsseveral logical cases or it can be used to conditionallyevaluate an expression in a manner similar toifelse.

For the first use, the... arguments have to be a series oflogical expressions.cases then returns a factorwith as many levels as logical expressions given as... arguments. The resulting factor will attain itsfirst level if the first condition is TRUE, otherwise it will attain itssecond level if the second condition is TRUE, etc.The levels will be named after the conditions or, if name tags areattached to the logical expressions, after the tags of the expressions.Not that the logical expressions all need to evaluate to logical vectorsof the same length, otherwise an error condition is raised.If.complete isTRUE then an additional factor level iscreated for the conditions not satisfied for any of the cases.

For the second use, the... arguments have to be a seriesof assignment expression of the type<expression> <- <logical expression>or<logical expression> -> <expression>. For casesin which the first logical expression is TRUE, the result of first expression thatappears on the other side of the assignment operator become elements of thevector returned bycases, for cases in which the second logical expression is TRUE,the result of the second expression that appears on the other sideof the assignment operator become elements of thevector returned bycases, etc.For cases that do not satisfy any of the given conditions the value ofthe.default argument is used. Note that the logical expressions also here all need to evaluate to logicalvectors of the same length. The expressions on the other side of theassignment operator should also be either vectors of the same lengthand mode or should scalars of the same mode, otherwise unpredictableresults may occur.

Value

If it is called with logical expressions as ... arguments,cases returns a factor, if it is called withassignment expressions the function returns a vector with thesame mode as the results of the "assigned" expressionsand with the same length as the logical conditions.

Examples

# Examples of the first kind of usage of the function#df <- data.frame(x = rnorm(n=20), y = rnorm(n=20))df <- df[do.call(order,df),](df <- within(df,{  x1=cases(x>0,x<=0)  y1=cases(y>0,y<=0)  z1=cases(    "Condition 1"=x<0,    "Condition 2"=y<0,# only applies if x >= 0    "Condition 3"=TRUE    )  z2=cases(x<0,(x>=0 & y <0), (x>=0 & y >=0))  }))xtabs(~x1+y1,data=df)dd <- with(df,  try(cases(x<0,            x>=0,            x>1,            check.xor=TRUE)# let's be fussy            )  )dd <- with(df,  try(cases(x<0,x>=0,x>1))  )genTable(range(x)~dd,data=df)# An example of the second kind of usage of the function:# A construction of a non-smooth function#fun <- function(x)  cases(    x==0      -> 1,    abs(x)> 1 -> abs(x),    abs(x)<=1 -> x^2  )x <- seq(from=-2,to=2,length=101)plot(fun(x)~x)# Demo of the new .default and .complete argumentsx <- seq(from=-2,to=2)cases(a = x < -1,      b = x > 1,      .complete = TRUE)cases(x < -1,      x > 1,      .complete = TRUE)cases(1 <- x < -1,      3 <- x > 1,      .default = 2)threshhold <- 5d <- c(1:10, NaN)d1 <- cases(  d > threshhold -> 1,  d <= threshhold -> 2)d2 <- cases(  is.na(d) -> 0,  d > threshhold -> 1,  d <= threshhold -> 2)# Leads to missing values because some of the conditions result in missing# even though they could be 'captured'd3 <- cases(  is.na(d) -> 0,  d > threshhold -> 1,  d <= threshhold -> 2,  na.rm=FALSE)d4 <- cases(  is.na(d) -> 0,  d > threshhold +2 -> 1,  d <= threshhold -> 2,  na.rm=FALSE)cbind(d,d1,d2,d3,d4)cases(  d > threshhold,  d <= threshhold)cases(  is.na(d),  d > threshhold,  d <= threshhold)cases(  d > threshhold,  d <= threshhold,  .complete=TRUE)cases(  d > threshhold + 2,  d <= threshhold,  .complete=TRUE)

Character Translation of Aspects of Objects

Description

This function uses the base package functionchartrto translate characters in variable descriptions (a.k.a variable labels) andvalue labels ofitem,data.set,andimporter objects.

It will be useful when the encoding of an important data set cannot be fully identified or if the encoding in the data fileis incorrect or unknown.

Usage

charTrans(x, old = "", new = "", ...)## S3 method for class 'annotation'charTrans(x, old = "", new = "", ...)## S3 method for class 'data.set'charTrans(x, old = "", new = "", ...)## S3 method for class 'importer'charTrans(x, old = "", new = "", ...)## S3 method for class 'item'charTrans(x, old = "", new = "", ...)## S3 method for class 'value.labels'charTrans(x, old = "", new = "", ...)

Arguments

Value

charTrans returns a copy of its first argument with character-translatedcharacter data or attributes.

See Also

chartr

Examples

## Not run: # Locate an SPSS 'portable' file and get info on variables, their labels S2601.POR <- spss.portable.file("POR-Files/S2601.POR", encoded = "cp850")# 'ß' appears to be correctly coded, but 'ä', 'ö', 'ü' are not, so we need to# to some fine-tuningS2601.POR <- charTrans(S2601.POR, old="{|}\r", new="äöü ")# Now labels etc. are correctly encoded.codebook(S2601.POR)## End(Not run)

Coarsen a vector into a factor with a lower number of levels

Description

coarsen can be used to obtain a factor from a vector, similartocut, but with less technical and more "aesthetic"labels of the factor levels.

Usage

coarsen(x,...)## S3 method for class 'numeric'coarsen(x,        n=5,        pretty=TRUE,        quantiles=!pretty,        breaks=NULL,        brackets=FALSE,        sep=if(brackets)";"else if(quantiles) "-" else " - ",        left="[",        right="]",        range=FALSE,        labels=NULL,        ...)

Arguments

Examples

x <- rnorm(200)table(coarsen(x))table(coarsen(x,quantiles=TRUE))table(coarsen(x,brackets=TRUE))table(coarsen(x,breaks=c(-1,0,1)))table(coarsen(x,breaks=c(-1,0,1),              range=TRUE,labels=letters[1:4]))

Generate a Codebook of a Data Set

Description

Functioncodebook collects documentation about an item,or the items in a data set or external data file. It returnsan object that, whenshown, print this documentationin a nicely formatted way.

Usage

codebook(x, weights = NULL, unweighted = TRUE, ...)## S4 method for signature 'item'codebook(x, weights = NULL, unweighted = TRUE, ...)## S4 method for signature 'atomic'codebook(x, weights = NULL, unweighted = TRUE, ...)## S4 method for signature 'factor'codebook(x, weights = NULL, unweighted = TRUE, ...)## S4 method for signature 'data.set'codebook(x, weights = NULL, unweighted = TRUE, ...)## S4 method for signature 'importer'codebook(x, weights = NULL, unweighted = TRUE, ...)## S4 method for signature 'data.frame'codebook(x, weights = NULL, unweighted = TRUE, ...)## S4 method for signature 'tbl_df'codebook(x, weights = NULL, unweighted = TRUE, ...)

Arguments

Value

An object of class "codebook", for which ashow method exists thatproduces a nicely formatted output.

Examples

Data <- data.set(          vote = sample(c(1,2,3,8,9,97,99),size=300,replace=TRUE),          region = sample(c(rep(1,3),rep(2,2),3,99),size=300,replace=TRUE),          income = exp(rnorm(300,sd=.7))*2000          )Data <- within(Data,{  description(vote) <- "Vote intention"  description(region) <- "Region of residence"  description(income) <- "Household income"  wording(vote) <- "If a general election would take place next tuesday,                    the candidate of which party would you vote for?"  wording(income) <- "All things taken into account, how much do all                    household members earn in sum?"  foreach(x=c(vote,region),{    measurement(x) <- "nominal"    })  measurement(income) <- "ratio"  labels(vote) <- c(                    Conservatives         =  1,                    Labour                =  2,                    "Liberal Democrats"   =  3,                    "Don't know"          =  8,                    "Answer refused"      =  9,                    "Not applicable"      = 97,                    "Not asked in survey" = 99)  labels(region) <- c(                    England               =  1,                    Scotland              =  2,                    Wales                 =  3,                    "Not applicable"      = 97,                    "Not asked in survey" = 99)  foreach(x=c(vote,region,income),{    annotation(x)["Remark"] <- "This is not a real survey item, of course ..."    })  missing.values(vote) <- c(8,9,97,99)  missing.values(region) <- c(97,99)})description(Data)codebook(Data)codebook(Data)$votecodebook(Data)[2]codebook(Data[2]) DataFr <- as.data.frame(Data)DataHv <- as_haven(Data,user_na=TRUE)codebook(DataFr)codebook(DataHv)   ## Not run: Write(description(Data),           file="Data-desc.txt")Write(codebook(Data),           file="Data-cdbk.txt")  ## End(Not run)

Describe structure of Data Sets and Importers

Description

The functioncodeplan() creates a data frame thatdescribes the structure of an item list (adata.set object oranimporter object), so that this structure can be stored andand recovered. The resulting data frame has a particular print methodthat delimits the output to one line per variable.

WithsetCodeplan an item list structure (as returned bycodeplan())can be applied to a data frame or data set. It is also possible to use anassignment likecodeplan(x) <- value to a similar effect.

Usage

codeplan(x)## S4 method for signature 'item.list'codeplan(x)## S4 method for signature 'item'codeplan(x)setCodeplan(x,value)## S4 method for signature 'data.frame,codeplan'setCodeplan(x,value)## S4 method for signature 'data.frame,NULL'setCodeplan(x,value)## S4 method for signature 'data.set,codeplan'setCodeplan(x,value)## S4 method for signature 'data.set,NULL'setCodeplan(x,value)## S4 method for signature 'item,codeplan'setCodeplan(x,value)## S4 method for signature 'item,NULL'setCodeplan(x,value)## S4 method for signature 'atomic,codeplan'setCodeplan(x,value)## S4 method for signature 'atomic,NULL'setCodeplan(x,value)codeplan(x) <- valueread_codeplan(filename,type)write_codeplan(x,filename,type,pretty)

Arguments

Value

If applicable,codeplan returns a list withadditional S3 class attribute"codeplan". For arguments forwhich the relevant information does not exist, the function returnsNULL.

The list has at least one element or several elements, named after thevariable in the "item.list" or "data.set"x. Each list elementis a list itself with the following elements:

Ifcodeplan(x)<-value orsetCodeplan(x,value) is usedandvalue isNULL, all the special information aboutannotation, labels, value filters, etc. is removed from the resultingobject, which then is usually a mere atomic vector or data frame.

Examples

Data1 <- data.set(          vote = sample(c(1,2,3,8,9,97,99),size=300,replace=TRUE),          region = sample(c(rep(1,3),rep(2,2),3,99),size=300,replace=TRUE),          income = exp(rnorm(300,sd=.7))*2000          )Data1 <- within(Data1,{  description(vote) <- "Vote intention"  description(region) <- "Region of residence"  description(income) <- "Household income"  foreach(x=c(vote,region),{    measurement(x) <- "nominal"    })  measurement(income) <- "ratio"  labels(vote) <- c(                    Conservatives         =  1,                    Labour                =  2,                    "Liberal Democrats"   =  3,                    "Don't know"          =  8,                    "Answer refused"      =  9,                    "Not applicable"      = 97,                    "Not asked in survey" = 99)  labels(region) <- c(                    England               =  1,                    Scotland              =  2,                    Wales                 =  3,                    "Not applicable"      = 97,                    "Not asked in survey" = 99)  foreach(x=c(vote,region,income),{    annotation(x)["Remark"] <- "This is not a real survey item, of course ..."    })  missing.values(vote) <- c(8,9,97,99)  missing.values(region) <- c(97,99)})cpData1 <- codeplan(Data1)Data2 <- data.frame(          vote = sample(c(1,2,3,8,9,97,99),size=300,replace=TRUE),          region = sample(c(rep(1,3),rep(2,2),3,99),size=300,replace=TRUE),          income = exp(rnorm(300,sd=.7))*2000          )codeplan(Data2) <- cpData1codeplan(Data2)codebook(Data2)# Note the difference between 'as.data.frame' and setting# the codeplan to NULL:Data2df <- as.data.frame(Data2)codeplan(Data2) <- NULLstr(Data2)str(Data2df)codeplan(Data2) <- NULL # Does not change anything# Codeplans of survey items can also be inquired and manipulated:vote <- Data1$votestr(vote)cp.vote <- codeplan(vote)codeplan(vote) <- NULLstr(vote)codeplan(vote) <- cp.votevotefn.json <- paste0(tempfile(),".json")write_codeplan(codeplan(Data1),filename=fn.json)codeplan(Data2) <- read_codeplan(fn.json)codeplan(Data2)

Collect Objects

Description

collect gathers several objects into one, matching theelements or subsets of the objects bynames ordimnames.

Usage

collect(...,names=NULL,inclusive=TRUE)## Default S3 method:collect(...,names=NULL,inclusive=TRUE)## S3 method for class 'array'collect(...,names=NULL,inclusive=TRUE)## S3 method for class 'matrix'collect(...,names=NULL,inclusive=TRUE)## S3 method for class 'table'collect(...,names=NULL,sourcename=".origin",fill=0)## S3 method for class 'data.frame'collect(...,names=NULL,inclusive=TRUE,                                  fussy=FALSE,warn=TRUE,                                  detailed.warnings=FALSE,use.last=FALSE,                                  sourcename=".origin")## S3 method for class 'data.set'collect(...,names=NULL,inclusive=TRUE,                                  fussy=FALSE,warn=TRUE,                                  detailed.warnings=FALSE,use.last=FALSE,                                  sourcename=".origin")

Arguments

Value

Ifx and all following ... arguments are vectors of the same mode (numeric,character, or logical)the result is a matrix with as many columns as vectors. If argumentinclusive is TRUE,then the number of rows equals the number of names that appear at least once in each of thevector names and the matrix is filled withNA where necessary,otherwise the number of rows equals the number of names that are present inallvector names.

Ifx and all ... arguments are matrices or arrays of the same mode (numeric,character, or logical)andn dimension the result will be an+1 dimensional array or table. The extend of then+1th dimension equals the number of matrix, array or table arguments,the extends of the lower dimension depends on theinclusive argument:either they equal to the number of dimnames that appear at least once for each givendimension and the array is filled withNA where necessary,or they equal to the number of dimnames that appear in all argumentsfor each given dimension.

Ifx and all ... arguments are data frames or data sets, theresult is a data frame or data set.The number of variables of the resulting data frame or data set depends ontheinclusive argument. If it is true, the number of variablesequals the number of variables that appear in each of the arguments at least onceand variables are filled withNA where necessary, otherwise thenumber of variables equals the number of variables that are present inall arguments.

Examples

x <- c(a=1,b=2)y <- c(a=10,c=30)xycollect(x,y)collect(x,y,inclusive=FALSE)X <- matrix(1,nrow=2,ncol=2,dimnames=list(letters[1:2],LETTERS[1:2]))Y <- matrix(2,nrow=3,ncol=2,dimnames=list(letters[1:3],LETTERS[1:2]))Z <- matrix(3,nrow=2,ncol=3,dimnames=list(letters[1:2],LETTERS[1:3]))XYZcollect(X,Y,Z)collect(X,Y,Z,inclusive=FALSE)X <- matrix(1,nrow=2,ncol=2,dimnames=list(a=letters[1:2],b=LETTERS[1:2]))Y <- matrix(2,nrow=3,ncol=2,dimnames=list(a=letters[1:3],c=LETTERS[1:2]))Z <- matrix(3,nrow=2,ncol=3,dimnames=list(a=letters[1:2],c=LETTERS[1:3]))collect(X,Y,Z)collect(X,Y,Z,inclusive=FALSE)df1 <- data.frame(a=rep(1,5),b=rep(1,5))df2 <- data.frame(a=rep(2,5),b=rep(2,5),c=rep(2,5))collect(df1,df2)collect(df1,df2,inclusive=FALSE)data(UCBAdmissions)Male <- as.table(UCBAdmissions[,1,])Female <- as.table(UCBAdmissions[,2,])collect(Male,Female,sourcename="Gender")collect(unclass(Male),unclass(Female))Male1 <- as.table(UCBAdmissions[,1,-1])Female2 <- as.table(UCBAdmissions[,2,-2])Female3 <- as.table(UCBAdmissions[,2,-3])collect(Male=Male1,Female=Female2,sourcename="Gender")collect(Male=Male1,Female=Female3,sourcename="Gender")collect(Male=Male1,Female=Female3,sourcename="Gender",fill=NA)f1 <- gl(3,5,labels=letters[1:3])f2 <- gl(3,6,labels=letters[1:3])collect(f1=table(f1),f2=table(f2))ds1 <- data.set(x = 1:3)ds2 <- data.set(x = 4:9,                y = 1:6)collect(ds1,ds2)

Convenience Methods for Setting Contrasts

Description

This package provides modified versions ofcontr.treatment andcontr.sum.contr.sumgains an optionalbase argument, analog to theone ofcontr.treatment, furthermore,thebase argument may be the name of afactor level.

contr returns a function that calls eithercontr.treatment,contr.sum, etc.,according to the value given to its first argument.

Thecontrasts method for"item" objectsreturns a contrast matrix or a function to producea contrast matrix for the factor into whichthe item would be coerced viaas.factor oras.ordered.This matrix or function can be specified byusingcontrasts(x)<-value

Usage

contr(type,...)contr.treatment(n, base=1,contrasts=TRUE)contr.sum(n,base=NULL,contrasts=TRUE)## S4 method for signature 'item'contrasts(x,contrasts=TRUE,...)## S4 replacement method for signature 'item'contrasts(x,how.many) <- value# These methods are defined implicitely by making 'contrasts' generic.## S4 method for signature 'ANY'contrasts(x,contrasts=TRUE,...)## S4 replacement method for signature 'ANY'contrasts(x,how.many) <- value

Arguments

Value

contr returns a funtion that calls one ofcontr.treatment,contr.sum,....contr.treatment andcontr.sum return contrast matrices.contrasts(x) returns the "contrasts" attribute of anobject, which may be a function name, a function, a contrast matrix or NULL.

Examples

ctr.t <- contr("treatment",base="c")ctr.tctr.s <- contr("sum",base="c")ctr.h <- contr("helmert")ctr.t(letters[1:7])ctr.s(letters[1:7])ctr.h(letters[1:7])x <- factor(rep(letters[1:5],3))contrasts(x)x <- as.item(x)contrasts(x)contrasts(x) <- contr.sum(letters[1:5],base="c")contrasts(x)missing.values(x) <- 5contrasts(x)contrasts(as.factor(x))# Obviously setting missing values after specifying# contrast matrix breaks the contrasts.# Using the 'contr' function, however, prevents this:missing.values(x) <- NULLcontrasts(x) <- contr("sum",base="c")contrasts(x)missing.values(x) <- 5contrasts(x)contrasts(as.factor(x))

Contract data into pattern-frequency format

Description

contract() contracts data into pattern-frequency format, similarto a contatenation oftable() (orxtabs) andas.data.frame(). Yet it uses much less memory if patternsare sparse, because it does not create rows for patterns that do not occur.

Usage

contract(x,...)## S3 method for class 'data.frame'contract(x,by=NULL, weights=NULL,name="Freq",    force.name=FALSE,sort=FALSE,drop.na=TRUE,...)## S3 method for class 'data.set'contract(x,by=NULL, weights=NULL,name="Freq",    force.name=FALSE,sort=FALSE,drop.na=TRUE,...)

Arguments

Value

Ifx is a data fame, the value ofcontract() is also adata frame. If it is a"data.set" object, the result is also a"data.set" object.

Examples

iris_ <- sample(iris,size=nrow(iris),replace=TRUE)w <- rep(1,nrow(iris_))contract(iris[4:5])contract(iris[4:5],sort=TRUE)contract(iris[4:5],weights=w,sort=TRUE)contract(iris,by=c(Petal.Width,Species),sort=TRUE)contract(iris,by=~Petal.Width+Species)contract(iris,by=w~Species)library(MASS)contract(housing,         by=Sat~Infl+Type+Cont,         weights=Freq)contract(housing,         by=Sat~Infl+Type+Cont,         weights=Freq,         name="housing",force.name=TRUE         )

Data Set Objects

Description

"data.set" objects are collections of"item" objects,with similar semantics as data frames. They are distinguishedfrom data frames so that coercion byas.data.fameleads to a data frame that contains only vectors and factors.Nevertheless most methods for data frames are inherited bydata sets, except for the method for thewithin genericfunction. For thewithin method for data sets, see the details section.

Thus data preparation using data sets retains all informationsabout item annotations, labels, missing values etc.While (mostly automatic) conversion of data sets into dataframes makes the data amenable for the use of R's statisticalfunctions.

dsView is a function that displays data sets in a similarmanner asView displays data frames. (View workswith data sets as well, but changes them first into data frames.)

Usage

data.set(...,row.names = NULL, check.rows = FALSE, check.names = TRUE,    stringsAsFactors = FALSE, document = NULL)as.data.set(x, row.names=NULL, ...)## S4 method for signature 'list'as.data.set(x,row.names=NULL,...)is.data.set(x)## S3 method for class 'data.set'as.data.frame(x, row.names = NULL, optional = FALSE, ...)## S4 method for signature 'data.set'within(data, expr, ...)dsView(x)## S4 method for signature 'data.set'head(x,n=20,...)## S4 method for signature 'data.set'tail(x,n=20,...)

Arguments

Details

Theas.data.frame method for data sets is just a copyof the method for list. Consequently, all items in the data setare coerced in accordance to theirmeasurement setting,seeas.vector,item-method andmeasurement.

Thewithin method for data sets has the same effect asthewithin method for data frames, apart from two differences:all results of the computations are coerced into items ifthey have the appropriate length, otherwise, they are automaticallydropped.

Currently only one method for the generic functionas.data.setis defined: a method for "importer" objects.

Value

data.set and thewithin method fordata sets returns a "data.set" object,is.data.setreturns a logical value, andas.data.frame returnsa data frame.

Examples

Data <- data.set(          vote = sample(c(1,2,3,8,9,97,99),size=300,replace=TRUE),          region = sample(c(rep(1,3),rep(2,2),3,99),size=300,replace=TRUE),          income = exp(rnorm(300,sd=.7))*2000          )Data <- within(Data,{  description(vote) <- "Vote intention"  description(region) <- "Region of residence"  description(income) <- "Household income"  wording(vote) <- "If a general election would take place next tuesday,                    the candidate of which party would you vote for?"  wording(income) <- "All things taken into account, how much do all                    household members earn in sum?"  foreach(x=c(vote,region),{    measurement(x) <- "nominal"    })  measurement(income) <- "ratio"  labels(vote) <- c(                    Conservatives         =  1,                    Labour                =  2,                    "Liberal Democrats"   =  3,                    "Don't know"          =  8,                    "Answer refused"      =  9,                    "Not applicable"      = 97,                    "Not asked in survey" = 99)  labels(region) <- c(                    England               =  1,                    Scotland              =  2,                    Wales                 =  3,                    "Not applicable"      = 97,                    "Not asked in survey" = 99)  foreach(x=c(vote,region,income),{    annotation(x)["Remark"] <- "This is not a real survey item, of course ..."    })  missing.values(vote) <- c(8,9,97,99)  missing.values(region) <- c(97,99)  # These to variables do not appear in the  # the resulting data set, since they have the wrong length.  junk1 <- 1:5  junk2 <- matrix(5,4,4)  })# Since data sets may be huge, only a# part of them are 'show'nData## Not run: # If we insist on seeing all, we can use 'print' insteadprint(Data)## End(Not run)str(Data)summary(Data)## Not run: # If we want to 'View' a data set we can use 'dsView'dsView(Data)# Works also, but changes the data set into a data frame first:View(Data)## End(Not run)Data[[1]]Data[1,]head(as.data.frame(Data))EnglandData <- subset(Data,region == "England")EnglandDataxtabs(~vote+region,data=Data)xtabs(~vote+region,data=within(Data, vote <- include.missings(vote)))

Manipulation of Data Sets

Description

Like data frames,data.set objects havesubset,unique,cbind,rbind,merge methods defined for them.

The semantics are basically the same as the methods definedfor data frames in thebase package, with the only differencethat the return values aredata.set objects.In fact, the methods described here are front-ends to thecorresponding methods for data frames, which are constructedsuch that the "extra" information attached to variables withindata.set objects, that is, toitem objects.

Usage

## S3 method for class 'data.set'subset(x, subset, select, drop = FALSE, ...)## S4 method for signature 'data.set'unique(x, incomparables = FALSE, ...)## S3 method for class 'data.set'cbind(..., deparse.level = 1)## S3 method for class 'data.set'rbind(..., deparse.level = 1)## S4 method for signature 'data.set,data.set'merge(x,y, ...)## S4 method for signature 'data.set,data.frame'merge(x,y, ...)## S4 method for signature 'data.frame,data.set'merge(x,y, ...)

Arguments

Examples

ds1 <- data.set(      a = rep(1:3,5),      b = rep(1:5,each=3)  )ds2 <- data.set(      a = c(3:1,3,3),      b = 1:5  )ds1 <- within(ds1,{      description(a) <- "Example variable 'a'"      description(b) <- "Example variable 'b'"  })ds2 <- within(ds2,{      description(a) <- "Example variable 'a'"      description(b) <- "Example variable 'b'"  })str(ds3 <- rbind(ds1,ds2))description(ds3)ds3 <- within(ds1,{        c <- a        d <- b        description(c) <- "Copy of variable 'a'"        description(d) <- "Copy of variable 'b'"        rm(a,b)    })str(ds4 <- cbind(ds1,ds3))description(ds4)ds5 <- data.set(        c = 1:3,        d = c(1,1,2)        )ds5 <- within(ds5,{      description(c) <- "Example variable 'c'"      description(d) <- "Example variable 'd'"  })str(ds6 <- merge(ds1,ds5,by.x="a",by.y="c"))# Note that the attributes of the left-hand variables# have priority.description(ds6)

Handle duplicated labels

Description

The functiondeduplicate_labels can be used with "item" objects,"importer" objects or "data.set" objects to deal with duplicate labels,i.e. labels that are attached to more thanone code. There are several ways to de-duplicate labels: by combiningvalues that share their label or by making labels duplicate labels distinct.

Usage

deduplicate_labels(x,...)## S3 method for class 'item'deduplicate_labels(x,    method=c("combine codes",             "prefix values",             "postfix values"),...)# Applicable to 'importer' objects and 'data.set' objects## S3 method for class 'item.list'deduplicate_labels(x,...)

Arguments

Value

The functiondeduplicate_labels a copy ofxthat has unqiue value labels.

Examples

x1 <- as.item(rep(1:5,4),              labels=c(                  A = 1,                  A = 2,                  B = 3,                  B = 4,                  C = 5              ),              annotation = c(                  description="Yet another test"))              x2 <- as.item(rep(1:4,5),              labels=c(                  i   = 1,                  ii  = 2,                  iii = 3,                  iii = 4                  ),              annotation = c(                  description="Still another test"))x3 <- as.item(rep(1:2,10),              labels=c(                  a = 1,                  b = 2                  ),              annotation = c(                  description="Still another test"))                            codebook(deduplicate_labels(x1))codebook(deduplicate_labels(x1,method="prefix"))codebook(deduplicate_labels(x1,method="postfix"))ds <- data.set(x1,x2,x3)codebook(deduplicate_labels(ds))codebook(deduplicate_labels(ds,method="prefix"))codebook(deduplicate_labels(ds,method="postfix"))

Change dimnames, rownames, or colnames

Description

These functions provide an easy way to change thedimnames,rownames orcolnames ofan array.

Usage

dimrename(x, dim = 1, ..., gsub = FALSE, fixed = TRUE, warn = TRUE)rowrename(x, ..., gsub = FALSE, fixed = TRUE, warn = TRUE)colrename(x, ..., gsub = FALSE, fixed = TRUE, warn = TRUE)

Arguments

Details

dimrename changes the dimnames ofx along dimension(s)dim according to theremaining arguments. The argument names are theoldnames, the values are the new names.rowrename is a shorthand for changing the rownames,colrename is a shorthand for changing the colnames of a matrixor matrix-like object.

Ifgsub is FALSE, argument tags are theolddimnames, the values are the newdimnames.Ifgsub is TRUE, arguments are substrings of thedimnamesthat are substituted by the argument values.

Value

Objectx with changed dimnames.

Examples

m <- matrix(1,2,2)rownames(m) <- letters[1:2]colnames(m) <- LETTERS[1:2]mdimrename(m,1,a="first",b="second")dimrename(m,1,A="first",B="second")dimrename(m,2,"A"="first",B="second")rowrename(m,a="first",b="second")colrename(m,"A"="first",B="second")# Since version 0.99.22 - the following also works:dimrename(m,1,a=first,b=second)dimrename(m,1,A=first,B=second)dimrename(m,2,A=first,B=second)

Check for and report duplicated labels

Description

The functionduplicated_labels can be used with "item" objects,"importer" objects or "data.set" objects to check whether itemscontain duplicate labels, i.e. labels that are attached to more thanone code.

Usage

duplicated_labels(x)## S3 method for class 'item'duplicated_labels(x)# Applicable to 'importer' objects and 'data.set' objects## S3 method for class 'item.list'duplicated_labels(x)

Arguments

Value

The functionduplicate.labels returns a list with a classattribute, which allows pretty printing of duplicated value labels

Examples

x1 <- as.item(rep(1:5,4),              labels=c(                  A = 1,                  A = 2,                  B = 3,                  B = 4,                  C = 5              ),              annotation = c(                  description="Yet another test"))              x2 <- as.item(rep(1:4,5),              labels=c(                  i   = 1,                  ii  = 2,                  iii = 3,                  iii = 4                  ),              annotation = c(                  description="Still another test"))x3 <- as.item(rep(1:2,10),              labels=c(                  a = 1,                  b = 2                  ),              annotation = c(                  description="Still another test"))                            duplicated_labels(x1)ds <- data.set(x1,x2,x3)duplicated_labels(ds)codebook(ds)nes1948.por <- unzip(system.file("anes/NES1948.ZIP",package="memisc"),                     "NES1948.POR",exdir=tempfile())nes1948 <- spss.portable.file(nes1948.por)duplicated_labels(nes1948)

Loop over Variables in a Data Frame or Environment

Description

foreach evaluates an expression given as untagged argument by substitutingin variables. The expression may also contain assignments, which take effect inthe caller's environment.

Usage

  foreach(...,.sorted,.outer=FALSE)

Arguments

Examples

x <- 1:3y <- -(1:3)z <- c("Uri","Schwyz","Unterwalden")print(x)print(y)print(z)foreach(var=c(x,y,z),          # assigns names  names(var) <- letters[1:3]   # to the elements of x, y, and z  )print(x)print(y)print(z)ds <- data.set(        a = c(1,2,3,2,3,8,9),        b = c(2,8,3,2,1,8,9),        c = c(1,3,2,1,2,8,8)      )print(ds)ds <- within(ds,{       description(a) <- "First item in questionnaire"      description(b) <- "Second item in questionnaire"      description(c) <- "Third item in questionnaire"            wording(a) <- "What number do you like first?"      wording(b) <- "What number do you like second?"      wording(c) <- "What number do you like third?"      foreach(x=a:c,{ # Lazy data documentation:        labels(x) <- c(    # a,b,c get value labels in one statement                         one = 1,                         two = 2,                       three = 3,                "don't know" = 8,         "refused to answer" = 9)        missing.values(x) <- c(8,9)        })      })      codebook(ds)# The colon-operator respects the order of the variables# in the data set, if .sorted=FALSEwith(ds[c(3,1,2)],     foreach(x=a:c,             print(description(x))            ))# Since .sorted=TRUE, the colon operator creates a range # of alphabetically sorted variables.with(ds[c(3,1,2)],     foreach(x=a:c,             print(description(x)),             .sorted=TRUE            ))# The variables in reverse orderwith(ds,     foreach(x=c:a,             print(description(x))            ))# The colon operator can be combined with the # concatenation functionwith(ds,     foreach(x=c(a:b,c,c,b:a),             print(description(x))            ))# Variables can also be selected by regular expressions.with(ds,     foreach(x=rx("[a-b]"),             print(description(x))            ))# A demonstration for '.outer=TRUE'foreach(l=letters[1:2],        i=1:3,        cat(paste0(l,i,"\n")),        .outer=TRUE)

Format Objects in HTML, show the HTML Format or Write it to a File

Description

show_html is for showing objects in a convenient way in HTML format.write_html writes them in HTML format into a file.Both functions call the genericformat_html for the format conversion.

Usage

show_html(x, output = NULL, ...)write_html(x, file, ..., standalone = TRUE)format_html(x, ...)## S3 method for class 'data.frame'format_html(x,    toprule=2,midrule=1,bottomrule=2,    split.dec=TRUE,    row.names=TRUE,    digits=getOption("digits"),    format="f",    style=df_format_stdstyle,    margin="2ex auto",     ...)    ## S3 method for class 'matrix'format_html(x,    toprule=2,midrule=1,bottomrule=2,    split.dec=TRUE,    formatC=FALSE,    digits=getOption("digits"),    format="f",    style=mat_format_stdstyle,    margin="2ex auto",     ...)

Arguments

Value

format_html character string with code suitable for inclusion into a HTML-file.

Format Codebooks as HTML

Description

This is the method offormat_html for "codebook" objects as createdby the eponymous function (seecodebook)

Usage

## S3 method for class 'codebook'format_html(x,     toprule = 2,    midrule = 1,    indent = "3ex",    style = codebook_format_stdstyle,    var_tag = "code",     varid_prefix = "", title_tag = "p",...)

Arguments

See Also

See Also asformat_html,show_html,write_html.

Format "Flattened Tables" as HTML

Description

This is the method offormat_html for "ftable" objects (i.e. flattenedcontingency tables)

Usage

## S3 method for class 'ftable'format_html(x,                    show.titles = TRUE,                    digits = 0,                    format = "f",                    toprule = 2, midrule = 1, bottomrule = 2,                    split.dec = TRUE,                    style = ftable_format_stdstyle,                   margin="2ex auto",                    ...)## S3 method for class 'ftable_matrix'format_html(x,                   show.titles=TRUE,                   digits=0,                   format="f",                   toprule=2,midrule=1,bottomrule=2,                   split.dec=TRUE,                   style = ftable_format_stdstyle,                   margin="2ex auto",                    varontop,                   varinfront,                   grouprules=1,                   multi_digits=NULL,                   ...)

Arguments

See Also

See Also asformat_html,show_html,write_html.

Format Codebooks as Markdown

Description

format_md is for showing objects in a convenient way in Markdownformat. Can be included to Rmarkdown file with thecat() function and theresults='asis' code block option. The following example should be runnedin a Rmd file with different output formats.

Usage

## S3 method for class 'codebook'format_md(x, ...)## S3 method for class 'codebookEntry'format_md(x, name = "", add_rules = TRUE, ...)

Arguments

Value

format_md character string with code suitable for inclusion into a Markdown-file.

Examples

library(memisc)Data1 <- data.set(  vote = sample(c(1,2,3,8,9,97,99),size=300,replace=TRUE),  region = sample(c(rep(1,3),rep(2,2),3,99),size=300,replace=TRUE),  income = exp(rnorm(300,sd=.7))*2000)Data1 <- within(Data1,{  description(vote) <- "Vote intention"  description(region) <- "Region of residence"  description(income) <- "Household income"  foreach(x=c(vote,region),{    measurement(x) <- "nominal"  })  measurement(income) <- "ratio"  labels(vote) <- c(    Conservatives         =  1,    Labour                =  2,    "Liberal Democrats"   =  3,    "Don't know"          =  8,    "Answer refused"      =  9,    "Not applicable"      = 97,    "Not asked in survey" = 99)  labels(region) <- c(    England               =  1,    Scotland              =  2,    Wales                 =  3,    "Not applicable"      = 97,    "Not asked in survey" = 99)  foreach(x=c(vote,region,income),{    annotation(x)["Remark"] <- "This is not a real survey item, of course ..."  })  missing.values(vote) <- c(8,9,97,99)  missing.values(region) <- c(97,99)})codebook_data <- codebook(Data1)codebook_md <- format_md(codebook_data, digits = 2)writeLines(codebook_md)## Not run: writeLines(codebook_md,con="codebook-example.md")## End(Not run)

Combining flattened tables.

Description

With the method functions described here, flattened (contingency) tables can be combinedinto more complex objects, of class"ftable_matrix". For objects of these classformat andprint methods are provided

Usage

## S3 method for class 'ftable'cbind(..., deparse.level=1)## S3 method for class 'ftable'rbind(..., deparse.level=1)## S3 method for class 'ftable_matrix'cbind(..., deparse.level=1)## S3 method for class 'ftable_matrix'rbind(..., deparse.level=1)## S3 method for class 'ftable_matrix'format(x,quote=TRUE,digits=0,format="f",...)## S3 method for class 'ftable_matrix'Write(x,                            file = "",                            quote = TRUE,                            append = FALSE,                            digits = 0,                            ...)                            ## S3 method for class 'ftable_matrix'print(x,quote=FALSE,...)

Arguments

Value

cbind andrbind, when used with"ftable" or"ftable_matrix"objects, return objects of class"ftable_matrix".

Examples

ft1 <- ftable(Sex~Survived,Titanic)ft2 <- ftable(Age+Class~Survived,Titanic)ft3 <- ftable(Survived~Class,Titanic)ft4 <- ftable(Survived~Age,Titanic)ft5 <- ftable(Survived~Sex,Titanic)tab10 <- xtabs(Freq~Survived,Titanic)(c12.10 <- cbind(ft1,ft2,Total=tab10))(r345.10 <- rbind(ft3,ft4,ft5,Total=tab10))## Not run: tf <- tempfile()Write(c12.10,file=tf)file.show(tf)## End(Not run)

Generic Tables and Data Frames of Descriptive Statistics

Description

genTable creates a table of arbitrary summaries conditional ongiven values of independent variables given by a formula.

Aggregate does the same, but returns adata.frame instead.

fapply is a generic function that dispatches on itsdataargument. It is called internally byAggregate andgenTable.Methods for this function can be used to adaptAggregate andgenTable to data sources other than data frames.

Usage

Aggregate(formula, data=parent.frame(), subset=NULL,          names=NULL, addFreq=TRUE, drop = TRUE, as.vars=1,          ...)genTable(formula, data=parent.frame(), subset=NULL,         names=NULL, addFreq=TRUE,...)

Arguments

Details

If an expression is given as left hand side of the formula, itsvalue is computed for any combination of values of the values on theright hand side. If the right hand side is a dot, then allvariables indata are added to the right hand side of theformula.

If no expression is given as left hand side,then the frequency counts for the respectivevalue combinations of the right hand variables are computed.

If a single factor is on the left hand side, then the left hand side istranslated into an appropriatecall totable(). Note that also in this caseaddFreq takes effect.

If a single numeric variable is on the left hand side, frequencycounts weighted by this variable are computed. In these cases,genTable is equivalent toxtabs andAggregate is equivalent toas.data.frame(xtabs(...)).

Value

Aggregateresults in a data frame with conditional summaries and unique value combinationsof conditioning variables.

genTable returns atable, that is, an array with class"table".

See Also

aggregate.data.frame,xtabs

Examples

ex.data <- expand.grid(mu=c(0,100),sigma=c(1,10))[rep(1:4,rep(100,4)),]ex.data <- within(ex.data,                  x<-rnorm(                    n=nrow(ex.data),                    mean=mu,                    sd=sigma                    )                  )Aggregate(~mu+sigma,data=ex.data)Aggregate(mean(x)~mu+sigma,data=ex.data)Aggregate(mean(x)~mu+sigma,data=ex.data,name="Average")Aggregate(c(mean(x),sd(x))~mu+sigma,data=ex.data)Aggregate(c(Mean=mean(x),StDev=sd(x),N=length(x))~mu+sigma,data=ex.data)genTable(c(Mean=mean(x),StDev=sd(x),N=length(x))~mu+sigma,data=ex.data)Aggregate(table(Admit)~.,data=UCBAdmissions)Aggregate(Table(Admit,Freq)~.,data=UCBAdmissions)Aggregate(Admit~.,data=UCBAdmissions)Aggregate(percent(Admit)~.,data=UCBAdmissions)Aggregate(percent(Admit)~Gender,data=UCBAdmissions)Aggregate(percent(Admit)~Dept,data=UCBAdmissions)Aggregate(percent(Gender)~Dept,data=UCBAdmissions)Aggregate(percent(Admit)~Dept,data=UCBAdmissions,Gender=="Female")genTable(percent(Admit)~Dept,data=UCBAdmissions,Gender=="Female")

Get Model Summaries for Use with "mtable"

Description

A generic function and methods to collect coefficientsand summary statistics from a model object. It is used inmtable

Usage

    ## S3 method for class 'lm'getSummary(obj, alpha=.05,...)  ## S3 method for class 'glm'getSummary(obj, alpha=.05,...)  ## S3 method for class 'merMod'getSummary(obj, alpha=.05, ...)# These are contributed by Christopher N. Lawrence  ## S3 method for class 'clm'getSummary(obj, alpha=.05,...)  ## S3 method for class 'polr'getSummary(obj, alpha=.05,...)  ## S3 method for class 'simex'getSummary(obj, alpha=.05,...)# These are contributed by Jason W. Morgan  ## S3 method for class 'aftreg'getSummary(obj, alpha=.05,...)  ## S3 method for class 'coxph'getSummary(obj, alpha=.05,...)  ## S3 method for class 'phreg'getSummary(obj, alpha=.05,...)  ## S3 method for class 'survreg'getSummary(obj, alpha=.05,...)  ## S3 method for class 'weibreg'getSummary(obj, alpha=.05,...)# These are contributed by Achim Zeileis  ## S3 method for class 'ivreg'getSummary(obj, alpha=.05,...)  ## S3 method for class 'tobit'getSummary(obj, alpha=.05,...)  ## S3 method for class 'hurdle'getSummary(obj, alpha=.05,...)  ## S3 method for class 'zeroinfl'getSummary(obj, alpha=.05,...)  ## S3 method for class 'betareg'getSummary(obj, alpha=.05,...)  ## S3 method for class 'multinom'getSummary(obj, alpha=.05,...)  # A variant that reports exponentiated coefficients.# The default method calls 'getSummary()' internally and should# be applicable to all classes for which 'getSummary()' methods exist.getSummary_expcoef(obj, alpha=.05,...)  ## Default S3 method:getSummary_expcoef(obj, alpha=.05,...)

Arguments

Details

The generic functiongetSummary is called bymtablein order to obtain the coefficients and summaries of model objects.In order to adaptmtable to models of classes otherthanlm orglm one needs todefinegetSummary methods for these classes andto set a summary template viasetSummaryTemplate

Value

Any method ofgetSummary must return a list with the followingcomponents:

Building Blocks for HTML Code

Description

The functions described here form building blocks fortheformat_html methods functions forcodebook,ftable,ftable_matrix, andmtable objects, etc.

The most basic of these functions ishtml, which constructs anobject that represents a minimal piece of HTML code and is member of theclass"html_elem". Unlike a character string containing HTMLcode, the resulting code element can relatively easily modified usingother functions presented here. The actual code is created when thefunctionas.character is applied to these objects.

Longer sequences of HTML code can be prepared by concatenating them withc, or byhtml_group,or by applyingas.html_group to a list of"html_elem" objects. All these result in objectsof class"html_group".

Attributes (such as class, id etc.) of HTML elements can be added to thecall tohtml, but can also later recalled or modified withattribs orsetAttribs. An important attributeis the style attribute, which can contain CSS styling. It canbe recalled or modified withstyle orsetStyle. Stylingstrings can also be created withhmtl_style oras.css

Usage

html(tag, ..., .content = NULL, linebreak = FALSE)html_group(...)as.html_group(x)content(x)content(x)<-valuesetContent(x,value)attribs(x)attribs(x)<-valuesetAttribs(x,...)## S3 method for class 'character'setAttribs(x,...)## S3 method for class 'html_elem'setAttribs(x,...)## S3 method for class 'html_group'setAttribs(x,...)css(...)as.css(x)style(x)style(x) <- valuesetStyle(x,...)## S3 method for class 'character'setStyle(x,...)## S3 method for class 'html_elem'setStyle(x,...)## S3 method for class 'html_group'setStyle(x,...)

Arguments

Details

Objects created withhtml are lists with class attribute"html_elem" and components

tag

a character string

attributes

a named character vector

content

a character vector, an"html_elem" or"html_group"object, or a list of such.

linebreak

a logical value or vector of length 2.

Objects created withhtml_group or by concatenationof"html_elem" or"html_group"objectare lists of such objects, with class attribute"html_group".

Examples

html("img")html("img",src="test.png")html("div",class="element",id="first","Sisyphus")html("div",class="element",id="first",.content="Sisyphus")div <- html("div",class="element",id="first",linebreak=c(TRUE,TRUE))content(div) <- "Sisyphus"divtag <- html("tag",linebreak=TRUE)attribs(tag)["class"] <- "something"attribs(tag)["class"]tagstyle(tag) <- c(color="#342334")style(tag)tagstyle(tag)["bg"] <- "white"tagsetStyle(tag,bg="black")setStyle(tag,c(bg="black"))c(div,tag,tag)c(  c(div,tag),  c(div,tag,tag))c(  c(div,tag),  div,tag,tag)c(  div,tag,  c(div,tag,tag))content(div) <- c(tag,tag,tag)divcss("background-color"="black",                  color="white")as.css(c("background-color"="black",                  color="white"))Hello <- "Hello World!"Hello <- html("p",Hello,linebreak=c(TRUE,TRUE))style(Hello) <- c(color="white",                  "font-size"="40px",                  "text-align"="center")     Link <- html("a","More examples here ...",             href="http://elff.eu/software/memisc",             title="More examples here ...",             style=css(color="white"),             linebreak=c(TRUE,FALSE))Link <- html("p"," (",Link,")",linebreak=c(TRUE,TRUE))style(Link) <- c(color="white",                 "font-size"="15px",                 "text-align"="center")Hello <- html("div",c(Hello,Link),linebreak=c(TRUE,TRUE))style(Hello) <- c("background-color"="#160666",                  padding="20px")Helloshow_html(Hello)

Object Oriented Interface to Foreign Files

Description

Importer objects are objects that refer to an externaldata file. Currently only Stata files,SPSS system, portable, and fixed-column files are supported.

Data are actually imported by ‘translating’ animporter file into adata.set usingas.data.set orsubset.

Theimporter mechanism is more flexible and extensiblethanread.spss andread.dtaof package "foreign", as most of the parsing of the file headers is done in R.It is also adapted to efficiently load large data sets.Most importantly, importer objects support thelabels,missing.values,anddescriptions, provided by this package.

Usage

spss.file(file,...)spss.fixed.file(file,  columns.file,  varlab.file=NULL,  codes.file=NULL,  missval.file=NULL,  count.cases=TRUE,  to.lower=getOption("spss.fixed.to.lower",FALSE),  iconv=TRUE,  encoded=getOption("spss.fixed.encoding","cp1252"),  negative2missing = FALSE)spss.portable.file(file,  varlab.file=NULL,  codes.file=NULL,  missval.file=NULL,  count.cases=TRUE,  to.lower=getOption("spss.por.to.lower",FALSE),  iconv=TRUE,  encoded=getOption("spss.por.encoding","cp1252"),  negative2missing = FALSE)spss.system.file(file,  varlab.file=NULL,  codes.file=NULL,  missval.file=NULL,  count.cases=TRUE,  to.lower=getOption("spss.sav.to.lower",FALSE),  iconv=TRUE,  encoded=getOption("spss.sav.encoding","cp1252"),  ignore.scale.info = FALSE,  negative2missing = FALSE)Stata.file(file,           iconv=TRUE,           encoded=if(new_format)                        getOption("Stata.new.encoding","utf-8")                   else getOption("Stata.old.encoding","cp1252"),           negative2missing = FALSE)## The most important methods for "importer" objects are:## S3 method for class 'spss.system.importer'subset(x, subset, select, drop = FALSE, ...)## S3 method for class 'spss.portable.importer'subset(x, subset, select, drop = FALSE, ...)## S3 method for class 'spss.fixed.importer'subset(x, subset, select, drop = FALSE, ...)## S3 method for class 'Stata.importer'subset(x, subset, select, drop = FALSE, ...)## S3 method for class 'Stata_new.importer'subset(x, subset, select, drop = FALSE, ...)## S4 method for signature 'importer'as.data.set(x,row.names=NULL,optional=NULL,                    compress.storage.modes=FALSE,...)## S4 method for signature 'importer'head(x,n=20,...)## S4 method for signature 'importer'tail(x,n=20,...)

Arguments

Details

A call to a ‘constructor’ for an importer object, that is,spss.fixed.file,spss.portable.file,spss.sysntax.file,orStata.file,causes R to read in the header of the data file and/orthe syntax files that contain information aboutthe variables, such as the columns that they occupy(in case ofspss.fixed.file), variable labels,value labels and missing values.

The information in the file header and/or the accompagnyingfiles is then processed to prepare the file for importing.Thus the inner structure of animporter object maywell vary according to what type of file is to imported andwhat additional information is given.

Theas.data.set andsubset methodsfor"importer" objects internally use thegeneric functionsseekData,readData,readSlice,andreadChunk, which have methods for thesubclasses of"importer".These functions are not callablefrom outside the package, however.

Thesubset method for"importer" objects reads inthe data ‘chunk-wise’ to create the subset of observations ifthe option"subset.chunk.size" is set to a non-NULLvalue, e.g. byoptions(subset.chunk.size=1000). This may beuseful in case of very large data sets from which only a tiny subsetof observations is needed for analysis.

Since the functions described here are more or less complete rewritebased on the description of the file structure providedby the documenation for PSPP, they are perhaps not as thorougly tested as the functions in theforeign package, apart from the frequent useby the author of this package.

Value

spss.fixed.file,spss.portable.file,spss.system.file, andStata.filereturn, respectively, objects of class"spss.fixed.importer","spss.portable.importer","spss.system.importer","Stata.importer", or"Stata_new.importer",which, by inheritance, are also objects of class"importer"."Stata.importer" is for files in the format of Stata versions upto 12, while"Stata_new.importer" is for files in the newerformat of Stata versions from 13.

Objects of class"importer" have at least the following two slots:

Theas.data.frame forimporter objects doesthe actual data import and returns a data frame. Note that in contrasttoread.spss, the variable names of theresulting data frame will be lower case, unless the importer functionis called withto.lower=FALSE. If long variable namesare defined (in case of a PSPP/SPSS system file), they takeprecedence and arenot coerced to lower case.

See Also

codebook,description,read.spss

Examples

# Extract American National Election Study of 1948nes1948.por <- unzip(system.file("anes/NES1948.ZIP",package="memisc"),                     "NES1948.POR",exdir=tempfile())# Get information about the variables contained.nes1948 <- spss.portable.file(nes1948.por)# The data are not yet loaded:show(nes1948)# ... but one can see what variables are present:description(nes1948)# Now a subset of the data is loaded:vote.socdem.48 <- subset(nes1948,              select=c(                  V480018,                  V480029,                  V480030,                  V480045,                  V480046,                  V480047,                  V480048,                  V480049,                  V480050                  ))# Let's make the names more descriptive:vote.socdem.48 <- rename(vote.socdem.48,                  V480018 = "vote",                  V480029 = "occupation.hh",                  V480030 = "unionized.hh",                  V480045 = "gender",                  V480046 = "race",                  V480047 = "age",                  V480048 = "education",                  V480049 = "total.income",                  V480050 = "religious.pref"        )# It is also possible to do both# in one step:# vote.socdem.48 <- subset(nes1948,#              select=c(#                  vote           = V480018,#                  occupation.hh  = V480029,#                  unionized.hh   = V480030,#                  gender         = V480045,#                  race           = V480046,#                  age            = V480047,#                  education      = V480048,#                  total.income   = V480049,#                  religious.pref = V480050#                  ))# We examine the data more closely:codebook(vote.socdem.48)# ... and conduct some analyses.#t(genTable(percent(vote)~occupation.hh,data=vote.socdem.48))# We consider only the two main candidates.vote.socdem.48 <- within(vote.socdem.48,{  truman.dewey <- vote  valid.values(truman.dewey) <- 1:2  truman.dewey <- relabel(truman.dewey,              "VOTED - FOR TRUMAN" = "Truman",              "VOTED - FOR DEWEY"  = "Dewey")  })summary(truman.relig.glm <- glm((truman.dewey=="Truman")~religious.pref,    data=vote.socdem.48,    family="binomial",))

Survey Items

Description

Objects of classitem are data vectors with additional informationattached to them like “value labels” and “user-defined missing values”known from software packages like SPSS or Stata.

The classitem is intended to facilitate data management ofsurvey data. Objects in this class shouldnot directly usedin data analysis. Instead they should changed into "ordinary" vectorsor factors before. For this see the documentation foras.vector,item-method.

Usage

## The constructor for objects of class "item"## more convenient than new("item",...)## S4 method for signature 'numeric'as.item(x,  labels=NULL,  missing.values=NULL,  valid.values=NULL,  valid.range=NULL,  value.filter=NULL,  measurement=NULL,  annotation=attr(x,"annotation"), ...  )## S4 method for signature 'character'as.item(x,  labels=NULL,  missing.values=NULL,  valid.values=NULL,  valid.range=NULL,  value.filter=NULL,  measurement=NULL,  annotation=attr(x,"annotation"), ...  )## S4 method for signature 'logical'as.item(x,...)# x is first coerced to integer,# arguments in ... are then passed to the "numeric"# method.## S4 method for signature 'factor'as.item(x,...)## S4 method for signature 'ordered'as.item(x,...)## S4 method for signature 'POSIXct'as.item(x,...)## S4 method for signature 'double.item'as.item(x,  labels=NULL,  missing.values=NULL,  valid.values=NULL,  valid.range=NULL,  value.filter=NULL,  measurement=NULL,  annotation=attr(x,"annotation"), ...  )## S4 method for signature 'integer.item'as.item(x,  labels=NULL,  missing.values=NULL,  valid.values=NULL,  valid.range=NULL,  value.filter=NULL,  measurement=NULL,  annotation=attr(x,"annotation"), ...  )## S4 method for signature 'character.item'as.item(x,  labels=NULL,  missing.values=NULL,  valid.values=NULL,  valid.range=NULL,  value.filter=NULL,  measurement=NULL,  annotation=attr(x,"annotation"), ...  )## S4 method for signature 'datetime.item'as.item(x,  labels=NULL,  missing.values=NULL,  valid.values=NULL,  valid.range=NULL,  value.filter=NULL,  measurement=NULL,  annotation=attr(x,"annotation"), ...  )

Arguments

See Also

annotationlabelsvalue.filter

Examples

  x <- as.item(rep(1:5,4),      labels=c(          "First"      = 1,          "Second"     = 2,          "Third"      = 3,          "Fourth"     = 4,          "Don't know" = 5        ),      missing.values=5,      annotation = c(        description="test"      ))  str(x)  summary(x)  as.numeric(x)  test <- as.item(rep(1:6,2),labels=structure(1:6,                                      names=letters[1:6]))  test  test == 1  test != 1  test == "a"  test != "a"  test == c("a","z")  test != c("a","z")  test   test   codebook(test)  Test <- as.item(rep(letters[1:6],2),                    labels=structure(letters[1:6],                                     names=LETTERS[1:6]))  Test  Test == "a"  Test != "a"  Test == "A"  Test != "A"  Test == c("a","z")  Test != c("a","z")  Test   Test   as.factor(test)  as.factor(Test)  as.numeric(test)  as.character(test)  as.character(Test)  as.data.frame(test)[[1]]

How Survey Items Are Converted into "Ordinary" Data Vectors

Description

Survey item objects in are numeric or character vectors with some extra informationthat may helpful for for managing and documenting survey data, but they are not suitablefor statistical data analysis. To run regressions etc. one should convertitem objects into "ordinary" numeric vectors or factors.This means that codes or values declared as "missing" (if present) are translated intothe generial missing valueNA, while value labels (if defined) are translated intofactor levels.

Usage

# The following methods can be used to covert items into# vectors with a given mode or into factors. ## S4 method for signature 'item'as.vector(x, mode = "any")## S4 method for signature 'item'as.numeric(x, ...)## S4 method for signature 'item'as.integer(x, ...)## S4 method for signature 'item.vector'as.factor(x)## S4 method for signature 'item.vector'as.ordered(x)## S4 method for signature 'item.vector'as.character(x, use.labels = TRUE, include.missings = FALSE, ...)## S4 method for signature 'datetime.item.vector'as.character()## S4 method for signature 'Date.item.vector'as.character()# The following methods are unlikely to be useful in practice, other than# that they are called internally by the 'as.data.frame()' method for "data.set"# objects.## S3 method for class 'character.item'as.data.frame(x, row.names = NULL, optional = FALSE, ...)## S3 method for class 'double.item'as.data.frame(x, row.names = NULL, optional = FALSE, ...)## S3 method for class 'integer.item'as.data.frame(x, row.names = NULL, optional = FALSE, ...)## S3 method for class 'Date.item'as.data.frame(x, row.names = NULL, optional = FALSE, ...)## S3 method for class 'datetime.item'as.data.frame(x, row.names = NULL, optional = FALSE, ...)

Arguments

Value

The functionas.vector() returns a logical, numeric, or characterdepending on themode= argument. Ifmode="any", the vectorhas the mode that corresponds to the (internal) mode of the itemvector, that is, an item in class "integer.item" will become an integervector, an item in class "double.item" will become a double-precisionnumeric vector, an item in class "character.item" will become acharacter vector; since the internal mode of a "dateitem.item" or a"Date.item" vector is numeric, a numeric vector will be returned.

The functionsas.integer(),as.numeric(),as.character(),as.factor(), andas.ordered() return an integer, numeric,or character vector, or an ordered or unordered factor, respectively.

Whenas.data.frame() is applied to an survey item object, theresult is a single-column data frame, where the single column is anumeric vector or character vector or factor depending on themeasurement attribute of the item. In particular, if themeasurement attribute equals"ratio" or"interval" this column will be the result ofas.vector(),if themeasurement attribute equals"ordinal" thiscolumn will be an ordered factor (seeordered), and ifthemeasurement attribute equals"nominal" thiscolumn will be an unordered factor (seefactor).

All these functions have in common that values declared as "missing" byvirtue of thevalue.filter attribute will be turned intoNA.

See Also

itemsannotationlabelsvalue.filter

Examples

  x <- as.item(rep(1:5,4),      labels=c(          "First"      = 1,          "Second"     = 2,          "Third"      = 3,          "Fourth"     = 4,          "Don't know" = 5        ),      missing.values=5,      annotation = c(        description="test"      ))  str(x)  summary(x)  as.numeric(x)  test <- as.item(rep(1:6,2),labels=structure(1:6,                                      names=letters[1:6]))  as.factor(test)  as.numeric(test)  as.character(test)  as.character(test,include.missings=TRUE)  as.data.frame(test)[[1]]

Value Labels

Description

Value labels associate character labels to possible valuesof an encoded survey item. Value labels are representedas objects of class "value.labels".

Value labels of an item can be obtainedusinglabels(x) andcan be associated to items and to vectorsusing labels(x) <- value

Value labels also can be updated using the+and- operators.

Usage

labels(object,...)labels(x) <- value

Arguments

Examples

  x <- as.item(rep(1:5,4),      labels=c(          "First"      = 1,          "Second"     = 2,          "Third"      = 3,          "Fourth"     = 4,          "Don't know" = 5        ),      missing.values=5,      annotation = c(        description="test"      ))  labels(x)  labels(x) <- labels(x) - c("Second"=2)  labels(x)  labels(x) <- labels(x) + c("Second"=2)  labels(x)  puvl <- getOption("print.use.value.labels")  options(print.use.value.labels=FALSE)  x  options(print.use.value.labels=TRUE)  x  options(print.use.value.labels=puvl)

Levels of Measurement of Survey Items

Description

The measurement level of a"item" object, which is one of "nominal", "ordinal", "interval", "ratio",determines what happens to it, if it or thedata.setcontaining it is coerced into adata.frame.If the level of measurement level is "nominal", the it will beconverted into an (unordered)factor, if the level of measurement is "ordinal",the item will be converted into anordered vector. If the measurementis "interval" or "ratio", the item will be converted into a numerical vector.

Usage

## S4 method for signature 'item'measurement(x)## S4 replacement method for signature 'item'measurement(x) <- value## S4 method for signature 'data.set'measurement(x)## S4 replacement method for signature 'data.set'measurement(x) <- valueis.nominal(x)is.ordinal(x)is.interval(x)is.ratio(x)as.nominal(x)as.ordinal(x)as.interval(x)as.ratio(x)set_measurement(x,...)

Arguments

Value

Theitem method ofmeasurement(x) returns a characterstring, thedata.set method returns a named character vector,where the name of each element is a variable name and each.

as.nominal,as.ordinal,as.interval,as.ratioreturn an item with the requested level of measurement setting.

is.nominal,is.ordinal,is.interval,is.ratioreturn a logical value.

References

Stevens, Stanley S. 1946. "On the theory of scales of measurement."Science 103: 677-680.

See Also

data.set,item

Examples

vote <- sample(c(1,2,3,8,9),size=30,replace=TRUE)labels(vote) <- c(Conservatives         =  1,                  Labour                =  2,                  "Liberal Democrats"   =  3,                  "Don't know"          =  8,                  "Answer refused"      =  9                  )missing.values(vote) <- c(8,9)as.data.frame(vote)[[1]]measurement(vote) <- "interval"as.data.frame(vote)[[1]]vote <- as.nominal(vote)as.data.frame(vote)[[1]]group <- sample(c(1,2),size=30,replace=TRUE)labels(group) <- c(A=1,B=2)DataS <- data.set(group,vote)measurement(DataS)measurement(DataS) <- list(interval=c("group","vote"))head(as.data.frame(DataS))DataS <- set_measurement(DataS,                         nominal=c(group,vote))head(as.data.frame(DataS))

Automatically Adapt Measurement Levels

Description

The generic functionmeasurement_autolevel changes the measurementlevels of "item" objects to "nominal" or "ordinal", ifthe proportion of its values that have labels is above a certainthreshold.

Usage

measurement_autolevel(x, ...)## S4 method for signature 'ANY'measurement_autolevel(x, ...) # Returns its argument as is## S4 method for signature 'item.vector'measurement_autolevel(x,                 to=getOption("measurement.adapt.default","nominal"),                threshold=getOption("measurement.adapt.threshold",.75),                ...)## S4 method for signature 'data.set'measurement_autolevel(x,                 to=getOption("measurement.adapt.default","nominal"),                threshold=getOption("measurement.adapt.threshold",.75),                except=NULL,                only=NULL,                ...)

Arguments

Examples

 exvect <- as.item(rep(1:2,5)) labels(exvect) <- c(a=1,b=2) codebook(exvect) codebook(measurement_autolevel(exvect)) avect <- as.item(sample(1:3,16,replace=TRUE)) labels(avect) <- c(a=1,b=2,c=3) bvect <- as.item(sample(1:4,16,replace=TRUE)) labels(bvect) <- c(A=1,B=2,C=3,D=4) ds <- data.set(a=avect,b=bvect) codebook(ds) codebook(measurement_autolevel(ds)) codebook(measurement_autolevel(ds,except=c(a,b))) codebook(measurement_autolevel(ds,only=a))

Deprecated Functions in Packagememisc

Description

These functions are provided for compatibility with older versions ofmemisc only, and may be defunct as soon as the next release.

Usage

fapply(formula,data,...) # calls UseMethod("fapply",data)## Default S3 method:fapply(formula, data, subset=NULL,      names=NULL, addFreq=TRUE,...)

Arguments

Comparative Table of Model Estimates

Description

mtable produces a table of estimates for several models.

Usage

mtable(...,coef.style=getOption("coef.style"),    summary.stats=TRUE,    signif.symbols=getOption("signif.symbols"),    factor.style=getOption("factor.style"),    show.baselevel=getOption("show.baselevel"),    baselevel.sep=getOption("baselevel.sep"),    getSummary=eval.parent(quote(getSummary)),    float.style=getOption("float.style"),    digits=min(3,getOption("digits")),    sdigits=digits,    show.eqnames=getOption("mtable.show.eqnames",NA),    gs.options=NULL,    controls=NULL,    collapse.controls=FALSE,    control.var.indicator=getOption("control.var.indicator",c("Yes","No"))  )## S3 method for class 'memisc_mtable'relabel(x, ..., gsub = FALSE, fixed = !gsub, warn = FALSE)## S3 method for class 'memisc_mtable'format(x,target=c("print","LaTeX","HTML","delim"),    ...    )## S3 method for class 'memisc_mtable'print(x,    center.at=getOption("OutDec"),    topsep="=",bottomsep="=",sectionsep="-",...)write.mtable(object,file="",             format=c("delim","LaTeX","HTML"),...)## S3 method for class 'memisc_mtable'toLatex(object,...)

Arguments

.

Details

mtable constructs a table of estimates for regression-type models.format.memisc_mtable formats suitable for use with output or conversion functionssuch asprint.memisc_mtable,toLatex.memisc_mtable, orwrite.memisc_mtable.

Value

A call tomtable results in an object of class"mtable"with the following components:

Examples

#### Basic workflowlm0 <- lm(sr ~ pop15 + pop75,              data = LifeCycleSavings)lm1 <- lm(sr ~                 dpi + ddpi, data = LifeCycleSavings)lm2 <- lm(sr ~ pop15 + pop75 + dpi + ddpi, data = LifeCycleSavings)options(summary.stats.lm=c("R-squared","N"))mtable("Model 1"=lm0,"Model 2"=lm1,"Model 3"=lm2)options(summary.stats.lm=c("sigma","R-squared","N"))mtable("Model 1"=lm0,"Model 2"=lm1,"Model 3"=lm2)options(summary.stats.lm=NULL)mtable123 <- mtable("Model 1"=lm0,"Model 2"=lm1,"Model 3"=lm2,    summary.stats=c("sigma","R-squared","F","p","N"))(mtable123 <- relabel(mtable123,  "(Intercept)" = "Constant",          pop15 = "Percentage of population under 15",          pop75 = "Percentage of population over 75",            dpi = "Real per-capita disposable income",           ddpi = "Growth rate of real per-capita disp. income"  ))# This produces output in tab-delimited format:write.mtable(mtable123)## Not run: # This produces output in tab-delimited format:file123 <- "mtable123.txt"write.mtable(mtable123,file=file123)file.show(file123)# The contents of this file can be pasted into Word# and converted into a Word table.## End(Not run)## Not run: texfile123 <- "mtable123.tex"write.mtable(mtable123,format="LaTeX",file=texfile123)file.show(texfile123)## End(Not run)#### Examples with UC Berkeley databerkeley <- Aggregate(Table(Admit,Freq)~.,data=UCBAdmissions)berk0 <- glm(cbind(Admitted,Rejected)~1,data=berkeley,family="binomial")berk1 <- glm(cbind(Admitted,Rejected)~Gender,data=berkeley,family="binomial")berk2 <- glm(cbind(Admitted,Rejected)~Gender+Dept,data=berkeley,family="binomial")mtable(berk0,summary.stats=c("Deviance","N"))mtable(berk1,summary.stats=c("Deviance","N"))mtable(berk0,berk1,berk2,summary.stats=c("Deviance","N"))mtable(berk0,berk1,berk2,          coef.style="horizontal",          summary.stats=c("Deviance","AIC","N"))mtable(berk0,berk1,berk2,          coef.style="stat",          summary.stats=c("Deviance","AIC","N"))mtable(berk0,berk1,berk2,          coef.style="ci",          summary.stats=c("Deviance","AIC","N"))mtable(berk0,berk1,berk2,          coef.style="ci.se",          summary.stats=c("Deviance","AIC","N"))mtable(berk0,berk1,berk2,          coef.style="ci.se.horizontal",          summary.stats=c("Deviance","AIC","N"))mtable(berk0,berk1,berk2,          coef.style="ci.p.horizontal",          summary.stats=c("Deviance","AIC","N"))mtable(berk0,berk1,berk2,          coef.style="ci.horizontal",          summary.stats=c("Deviance","AIC","N"))mtable(berk0,berk1,berk2,          coef.style="all",          summary.stats=c("Deviance","AIC","N"))mtable(berk0,berk1,berk2,          coef.style="all.nostar",          summary.stats=c("Deviance","AIC","N"))mtable(by(berkeley,berkeley$Dept,  function(x)glm(cbind(Admitted,Rejected)~Gender,        data=x,family="binomial")),      summary.stats=c("Likelihood-ratio","N"))mtable(By(~Gender,  glm(cbind(Admitted,Rejected)~Dept,        family="binomial"),        data=berkeley),      summary.stats=c("Likelihood-ratio","N"))berkfull <- glm(cbind(Admitted,Rejected)~Dept/Gender - 1,                      data=berkeley,family="binomial")relabel(mtable(berkfull),Dept="Department",gsub=TRUE)#### Array-like semanticsmtable123 <- mtable("Model 1"=lm0,"Model 2"=lm1,"Model 3"=lm2,    summary.stats=c("sigma","R-squared","F","p","N"))dim(mtable123)dimnames(mtable123)mtable123[c("dpi","ddpi"),          c("Model 2","Model 3")]#### Concatentionmt01 <- mtable(lm0,lm1,summary.stats=c("R-squared","N"))mt12 <- mtable(lm1,lm2,summary.stats=c("R-squared","F","N"))c(mt01,mt12) # not that this makes sense, but ...c("Group 1"=mt01,  "Group 2"=mt12)

Format for 'mtable' Objects for Writing into File

Description

mtable_mtable_print formats 'mtable' in a way suitable for output into a filewithwrite.table

Usage

mtable_format_delim(x,          colsep="\t",          rowsep="\n",          interaction.sep = " x ",          ...          )

Arguments

Value

A character string.

HTML Formatting for 'mtable' Results

Description

These functions formats 'mtable' objects into HTML format.

Usage

mtable_format_html(x,                    interaction.sep = NULL,                    toprule=2,midrule=1,bottomrule=2,                    split.dec=TRUE,                    style=mtable_format_stdstyle,                    margin="2ex auto",                     sig.notes.style=c(width="inherit"),                    ...  )## S3 method for class 'memisc_mtable'format_html(x,                    interaction.sep = NULL,                    toprule=2,midrule=1,bottomrule=2,                    split.dec=TRUE,                    style=mtable_format_stdstyle,                    margin="2ex auto",                     sig.notes.style=c(width="inherit"),                    ...  )

Arguments

Value

A character string with code suitable for inclusion into a HTML-file.

Examples

lm0 <- lm(sr ~ pop15 + pop75,              data = LifeCycleSavings)lm1 <- lm(sr ~                 dpi + ddpi, data = LifeCycleSavings)lm2 <- lm(sr ~ pop15 + pop75 + dpi + ddpi, data = LifeCycleSavings)mtable123 <- mtable("Model 1"=lm0,"Model 2"=lm1,"Model 3"=lm2,                    summary.stats=c("sigma","R-squared","F","p","N"))(mtable123 <- relabel(mtable123,                      "(Intercept)" = "Constant",                      pop15 = "Percentage of population under 15",                      pop75 = "Percentage of population over 75",                      dpi = "Real per-capita disposable income",                      ddpi = "Growth rate of real per-capita disp. income"))# Use HTML entity '&minus;' for minus signoptions(html.use.ampersand=TRUE)show_html(mtable123)show_html(mtable123[1:2],          sig.notes.style=c(width="30ex"))# Use unicode for minus sign (default)options(html.use.ampersand=FALSE)show_html(mtable123)

Format 'mtable' Results for LaTeX

Description

This function formats objects created bymtable for inclusioninto LaTeX files.

Usage

  mtable_format_latex(x,            useDcolumn=getOption("useDcolumn",TRUE),            colspec=if(useDcolumn)                       paste("D{.}{",LaTeXdec,"}{",ddigits,"}",sep="")                     else "l",            LaTeXdec=".",            ddigits=min(3,getOption("digits")),            useBooktabs=getOption("useBooktabs",TRUE),            toprule=if(useBooktabs) "\\toprule" else "\\hline\\hline",            midrule=if(useBooktabs) "\\midrule" else "\\hline",            cmidrule=if(useBooktabs) "\\cmidrule" else "\\cline",            bottomrule=if(useBooktabs) "\\bottomrule" else "\\hline\\hline",            interaction.sep = " $\\times$ ",            sdigits=min(1,ddigits),            compact=FALSE,            sumry.multicol=FALSE,            escape.tex=getOption("toLatex.escape.tex",FALSE),            signif.notes.type=getOption("toLatex.signif.notes.type","include"),            signif.notes.spec=getOption("toLatex.signif.notes.spec","p{.5\\linewidth}"),            ...  )

Arguments

Value

A character string with code suitable for inclusion into a LaTeX-file.

Print Format for 'mtable' Objects

Description

mtable_format_print formats 'mtable' in a way suitable for screen outputwith 'print'.

Usage

mtable_format_print(x,  topsep="=",  bottomsep="=",  sectionsep="-",  interaction.sep = " x ",  center.at=getOption("OutDec"),  align.integers=c("dot","right","left"),  padding = "  ",  ...  )

Arguments

Value

A character string.

Mark Negative Values as Missing

Description

In many newer survey data sets available from socialscience data archives non-valid responses (such as "don't know" or"answer refused") are given negative codes. The functionneg2miss allows to mark them as missing values.)

Usage

neg2mis(x,all=FALSE,exclude=NULL,select=NULL,zero=FALSE)

Arguments

Examples

ds <- data.set(          var1 = c(0,1,-1,2,3),          var2 = c(-1,-1,1,1,1),          var3 = c(1,2,3,4,5)          )neg2mis(ds,all=TRUE)neg2mis(ds,all=TRUE,zero=TRUE)neg2mis(ds,exclude=var1)neg2mis(ds,select=var1)

Negative Match

Description

%nin%is a convenience operator:x %nin% tableis equivalent to!(x %in% table).

Usage

  x %nin% table

Arguments

Value

A logical vector

Examples

x <- sample(1:6,12,replace=TRUE)x %in% 1:3x %nin% 1:3

Table of Percentages with Percentage Base

Description

percent returns a table of percentages along withthe percentage base. It will be usefulin conjunction withAggregate orgenTable.

Usage

  percent(x,...)  ## Default S3 method:percent(x,weights=NULL,total=!(se || ci),      se=FALSE,ci=FALSE,ci.level=.95,      total.name="N",perc.label="Percentage",...)  ## S3 method for class 'logical'percent(x,weights=NULL,total=!(se || ci),      se=FALSE,ci=FALSE,ci.level=.95,      total.name="N",perc.label="Percentage",...)

Arguments

Value

A table of percentages.

Examples

x <- rnorm(100)y <- rnorm(100)z <- rnorm(100)f <- sample(1:3,100,replace=TRUE)f <- factor(f,labels=c("a","b","c"))percent(x>0)percent(f)genTable(  cbind(percent(x>0),        percent(y>0),        percent(z>0)) ~ f  )gt <- genTable(  cbind("x > 0" = percent(x>0,ci=TRUE),        "y > 0" = percent(y>0,ci=TRUE),        "z > 0" = percent(z>0,ci=TRUE)) ~ f  )ftable(gt,row.vars=3:2,col.vars=1)ex.data <- expand.grid(mean=c(0,25,50),sd=c(1,10,100))[rep(1:9,rep(250,9)),]ex.data <- within(ex.data,x <- rnorm(n=nrow(ex.data),mean=ex.data$mean,sd=ex.data$sd))ex.data <- within(ex.data,x.grp <- cases( x < 0,                                            x >= 0 & x < 50,                                            x >= 50 & x < 100,                                            x >= 100                                          ))genTable(percent(x.grp)~mean+sd,data=ex.data)Aggregate(percent(Admit,weight=Freq)~Gender+Dept,data=UCBAdmissions)

Easy Creation of Tables of Percentages

Description

The generic functionpercentages and its methodscreate one- or multidimensional tables of percentages. As such,the functionpercentages can be viewed as a convenienceinterface toprop.table. However, it alsoallows to obtain standard errors and confidence intervals.

Usage

percentages(obj, ...)## S3 method for class 'table'percentages(obj,      by=NULL, which=NULL, se=FALSE, ci=FALSE, ci.level=.95, ...)## S3 method for class 'formula'percentages(obj,      data=parent.frame(), weights=NULL, ...)## Default S3 method:percentages(obj,      weights=NULL, ...)## S3 method for class 'data.frame'percentages(obj,      weights=NULL, ...)## S3 method for class 'list'percentages(obj,      weights=NULL, ...)## S3 method for class 'percentage.table'as.data.frame(x, ...)## S3 method for class 'xpercentage.table'as.data.frame(x, ...)

Arguments

Value

An array that inherits classes "percentage.table" and "table". Ifpercentages was called withse=TRUE orci=TRUEthen the result additionally inherits class "xpercentage.table".

Examples

percentages(UCBAdmissions)# Three equivalent ways to create the same table of conditional# percentagespercentages(Admit~Gender+Dept,data=UCBAdmissions)percentages(UCBAdmissions,by=c("Gender","Dept"))percentages(UCBAdmissions,which="Admit")# Percentage table as data frameas.data.frame(percentages(Admit~Gender+Dept,data=UCBAdmissions))# Standard errors and confidence intervalspercentages(Admit~Dept,data=UCBAdmissions,se=TRUE)percentages(Admit~Dept,data=UCBAdmissions,ci=TRUE)(p<- percentages(Admit~Dept,data=UCBAdmissions,ci=TRUE,se=TRUE))# An extended table of percentages as data frameas.data.frame(p)# A table of percentages of a factorpercentages(iris$Species)UCBA <- as.data.frame(UCBAdmissions)percentages(UCBA$Admit,weights=UCBA$Freq)percentages(UCBA,weights=UCBA$Freq)

Query an Object for Information

Description

The functionquery can be used to search an objectfor a keyword.

Thedata.set andimporter methods perform such a searchthrough the annotations and value labels ofthe items in the data set.

Usage

query(x,pattern,...)## S4 method for signature 'data.set'query(x,pattern,...)## S4 method for signature 'importer'query(x,pattern,...)## S4 method for signature 'item'query(x,pattern,...)# (Called by the methods above.)

Arguments

Value

If both the annotation and the value labels of an item match the patternthequery method for 'item' objects returns a list containing the annotationand the value labels, otherwise if only the annotation or the value labelsmatch the pattern, either the annotation or the value labels are returned,otherwise if neither matches the pattern,query returnsNULL.

The methods ofquery for 'data.set' and 'importer' objects returna list of all non-NULL query results of all items contained by theseobjects, orNULL.

Examples

nes1948.por <- unzip(system.file("anes/NES1948.ZIP",package="memisc"),                     "NES1948.POR",exdir=tempfile())nes1948 <- spss.portable.file(nes1948.por)query(nes1948,"TRUMAN")

Recode Items, Factors and Numeric Vectors

Description

recode substitutes old values of a factor or a numericvector by new ones, just like the recoding facilities in somecommercial statistical packages.

Usage

recode(x,...,       copy=getOption("recode_copy",identical(otherwise,"copy")),       otherwise=NA)## S4 method for signature 'vector'recode(x,...,    copy=getOption("recode_copy",identical(otherwise,"copy")),    otherwise=NA)## S4 method for signature 'factor'recode(x,...,    copy=getOption("recode_copy",identical(otherwise,"copy")),    otherwise=NA)## S4 method for signature 'item'recode(x,...,    copy=getOption("recode_copy",identical(otherwise,"copy")),    otherwise=NA)

Arguments

Details

recode relies on the lazy evaluation mechanism ofR:Arguments are not evaluated until required by the function they are given to.recode does not cause arguments that appear in... to be evaluated.Instead,recode parses the... arguments. Therefore, althoughexpressions like1 <- 1:4 would cause an error action, if evaluatedat any place elsewhere inR, they will not cause an error action,if given torecode as an argument. However, a call of theformrecode(x,1=1:4), would be a syntax error.

If John Fox' package "car" is installed,recode will also be callablewith the syntax of therecode function of that package.

Value

A numerical vector, factor or anitem object.

See Also

recode of package "car".

Examples

x <- as.item(sample(1:6,20,replace=TRUE),        labels=c( a=1,                  b=2,                  c=3,                  d=4,                  e=5,                  f=6))print(x)codebook(    recode(x,           a = 1 <- 1:2,           b = 2 <- 4:6))codebook(    recode(x,           a = 1 <- 1:2,           b = 2 <- 4:6,           copy = TRUE))# Note the handling of labels if the recoding rules are bijectivecodebook(    recode(x,           1 <- 2,           2 <- 1,           copy=TRUE))codebook(    recode(x,           a = 1 <- 2,           b = 2 <- 1,           copy=TRUE))# A recoded version of x is returned# containing the values 1, 2, 3, which are# labelled as "A", "B", "C".recode(x,  A = 1 <- range(min,2),  B = 2 <- 3:4,  C = 3 <- range(5,max), # this last comma is ignored  )# This causes an error action: the sets# of original values overlap.try(recode(x,  A = 1 <- range(min,2),  B = 2 <- 2:4,  C = 3 <- range(5,max)  ))recode(x,  A = 1 <- range(min,2),  B = 2 <- 3:4,  C = 3 <- range(5,6),  D = 4 <- 7  )  # This results in an all-missing vector:recode(x,  D = 4 <- 7,  E = 5 <- 8  )f <- as.factor(x)x <- as.integer(x)recode(x,  1 <- range(min,2),  2 <- 3:4,  3 <- range(5,max)  )# This causes another error action:# the third argument is an invalid# expression for a recoding.try(recode(x,  1 <- range(min,2),  3:4,  3 <- range(5,max)  ))# The new values are character strings,# therefore a factor is returned.recode(x,  "a" <- range(min,2),  "b" <- 3:4,  "c" <- range(5,6)  )  recode(x,  1 <- 1:3,  2 <- 4:6  )  recode(x,  4 <- 7,  5 <- 8,  otherwise = "copy"  )recode(f,  "A" <- c("a","b"),  "B" <- c("c","d"),  otherwise="copy"  )recode(f,  "A" <- c("a","b"),  "B" <- c("c","d"),  otherwise="C"  ) recode(f,  "A" <- c("a","b"),  "B" <- c("c","d")  )DS <- data.set(x=as.item(sample(1:6,20,replace=TRUE),        labels=c( a=1,                  b=2,                  c=3,                  d=4,                  e=5,                  f=6)))print(DS)DS <- within(DS,{    xf <- recode(x,                 "a" <- range(min,2),                 "b" <- 3:4,                 "c" <- range(5,6)                 )    xn <- x@.Data    xc <- recode(xn,                 "a" <- range(min,2),                 "b" <- 3:4,                 "c" <- range(5,6)                 )    xc <- as.character(x)    xcc <- recode(xc,                  1 <- letters[1:2],                  2 <- letters[3:4],                  3 <- letters[5:6]                  )})DSDS <- within(DS,{    xf <- recode(x,                 "a" <- range(min,2),                 "b" <- 3:4,                 "c" <- range(5,6)                 )    x1 <- recode(x,                 1 <- range(1,2),                 2 <- range(3,4),                 copy=TRUE                 )    xf1 <- recode(x,                 "A" <- range(1,2),                 "B" <- range(3,4),                 copy=TRUE                 )})DScodebook(DS)DF <- data.frame(x=rep(1:6,4,replace=TRUE))DF <- within(DF,{    xf <- recode(x,                 "a" <- range(min,2),                 "b" <- 3:4,                 "c" <- range(5,6)                 )    x1 <- recode(x,                 1 <- range(1,2),                 2 <- range(3,4),                 copy=TRUE                 )    xf1 <- recode(x,                 "A" <- range(1,2),                 "B" <- range(3,4),                 copy=TRUE                 )    xf2 <- recode(x,                 "B" <- range(3,4),                 "A" <- range(1,2),                 copy=TRUE                 )})DFcodebook(DF)

Change labels of factors or labelled objects

Description

Functionrelabel changes the labels of a factor or any objectthat has anames,labels,value.labels, orvariable.labels attribute.Functionrelabel4 is an (internal) generic which is called byrelabelto handle S4 objects.

Usage

## Default S3 method:relabel(x, ..., gsub = FALSE, fixed = TRUE, warn = TRUE)## S3 method for class 'factor'relabel(x, ..., gsub = FALSE, fixed = TRUE, warn = TRUE)## S4 method for signature 'item'relabel4(x, ...)# This is an internal method, see details.# Use relabel(x, \dots) for 'item' objects

Arguments

Details

This function changes the names or labels ofx according to theremaining arguments.Ifgsub is FALSE, argument tags are theoldlabels, the values are the new labels.Ifgsub is TRUE, arguments are substrings of the labelsthat are substituted by the argument values.

Functionrelabel is S3 generic. If its first argument is an S4 object,it calls the (internal)relabel4 generic function.

Value

The objectx with new labels defined by the ... arguments.

Examples

  f <- as.factor(rep(letters[1:4],5))  levels(f)  F <- relabel(f,    a="A",    b="B",    c="C",    d="D"    )  levels(F)    f <- as.item(f)  labels(f)  F <- relabel(f,    a="A",    b="B",    c="C",    d="D"    )  labels(F)  # Since version 0.99.22 - the following also works:  f <- as.factor(rep(letters[1:4],5))  levels(f)  F <- relabel(f,    a=A,    b=B,    c=C,    d=D    )  levels(F)    f <- as.item(f)  labels(f)  F <- relabel(f,    a=A,    b=B,    c=C,    d=D    )  labels(F)

Change Names of a Named Object

Description

rename changes the names of a named object.

Usage

rename(x, ..., gsub = FALSE, fixed = TRUE, warn = TRUE)

Arguments

Details

This function changes the names ofx according to theremaining arguments.Ifgsub is FALSE, argument tags are theoldnames, the values are the new names.Ifgsub is TRUE, arguments are substrings of the namesthat are substituted by the argument values.

Value

The objectx with new names defined by the ... arguments.

Examples

  x <- c(a=1, b=2)  rename(x,a="A",b="B")  # Since version 0.99.22 - the following also works:  rename(x,a=A,b=B)    str(rename(iris,                  Sepal.Length="Sepal_Length",                  Sepal.Width ="Sepal_Width",                  Petal.Length="Petal_Length",                  Petal.Width ="Petal_Width"                  ))  str(rename(iris,                  .="_"                  ,gsub=TRUE))  # Since version 0.99.22 - the following also works:  str(rename(iris,                  Sepal.Length=Sepal_Length,                  Sepal.Width =Sepal_Width,                  Petal.Length=Petal_Length,                  Petal.Width =Petal_Width                  ))

Reorder an Array or Matrix

Description

reorder.array reorders an array along a specifieddimension according given names, indices or results ofa function applied.

Usage

## S3 method for class 'array'reorder(x,dim=1,names=NULL,indices=NULL,FUN=mean,...)## S3 method for class 'matrix'reorder(x,dim=1,names=NULL,indices=NULL,FUN=mean,...)

Arguments

Details

Typical usages are

  reorder(x,dim,names)  reorder(x,dim,indices)  reorder(x,dim,FUN)

The result ofrename(x,dim,names) isxreordered such thatdimnames(x)[[dim]] is equal tothe concatenation of those elements ofnamesthat are indimnames(x)[[dim]] and the remaining elementsofdimnames(x)[[dim]].

The result ofrename(x,dim,indices) isxreordered alongdim according toindices.

The result ofrename(x,dim,FUN) isxreordered alongdim according toorder(apply(x,dim,FUN)).

Value

The reordered objectx.

See Also

The default method ofreorder in packagestats.

Examples

  (M <- matrix(rnorm(n=25),5,5,dimnames=list(LETTERS[1:5],letters[1:5])))  reorder(M,dim=1,names=c("E","A"))  reorder(M,dim=2,indices=3:1)  reorder(M,dim=1)  reorder(M,dim=2)

Retain Objects in an Environment

Description

retain removes all objects from the environmentexcept those mentioned as argument.

Usage

retain(..., list = character(0), envir = parent.frame(),force=FALSE)

Arguments

Examples

local({  foreach(x=c(a,b,c,d,e,f,g,h),x<-1)  cat("Objects before call to 'retain':\n")  print(ls())  retain(a)  cat("Objects after call to 'retain':\n")  print(ls())})x <- 1y <- 2retain(x)

Reverse the codes of a survey item or the levels of a factor

Description

The functionreversed() returns a copy of its argument with codesor levels in reverse order.

Usage

reversed(x)## S4 method for signature 'item.vector'reversed(x)## S4 method for signature 'factor'reversed(x)

Arguments

Value

If the argument of the functionreversed() than either theunique valid values or the labelled valid values recoded into thereverse order.

If th argument is a factor than the function returns the factor withlevels in reverse order.

Examples

ds <- data.set(    x = as.item(sample(c(1:3,9),100,replace=TRUE),                labels=c("One"=1,                         "Two"=2,                         "Three"=3,                         "Missing"=9)))df <- as.data.frame(ds)ds <- within(ds,{    xr <- reversed(x)})codebook(ds)df <- within(df,{    xr <- reversed(x)})codebook(df)

Take a Sample from a Data Frame-like Object

Description

The methods below are convenience short-cuts totake samples from data frames and data sets.They result in a data frame or data set, respectively,the rows of which are a sample of the completedata frame/data set.

Usage

## S4 method for signature 'data.frame'sample(x, size, replace = FALSE, prob = NULL)## S4 method for signature 'data.set'sample(x, size, replace = FALSE, prob = NULL)## S4 method for signature 'importer'sample(x, size, replace = FALSE, prob = NULL)

Arguments

Value

A data frame or data set.

Examples

for(.i in 1:4)  print(sample(iris,5))

Convenience Methods to Sort Data Frames and Data Sets

Description

The methods below return a sorted version ofthe data frame or data set, given as first argument.

Usage

## S3 method for class 'data.frame'sort(x,decreasing=FALSE,by=NULL,na.last=NA,...)## S3 method for class 'data.set'sort(x,decreasing=FALSE,by=NULL,na.last=NA,...)

Arguments

Value

A sorted copy ofx.

Examples

DF <- data.frame(        a = sample(1:2,size=20,replace=TRUE),        b = sample(1:4,size=20,replace=TRUE))sort(DF)sort(DF,by=~a+b)sort(DF,by=~b+a)sort(DF,by=c("b","a"))sort(DF,by=c("a","b"))

Formatting Styles for Coefficients, Factor Contrasts, and Summary Statistics

Description

Methods for setting and getting templates for formattingmodel coefficients and summaries for use inmtable.

Usage

setCoefTemplate(...)getCoefTemplate(style)getSummaryTemplate(x)setSummaryTemplate(...)summaryTemplate(x)

Arguments

Details

The style in which model coefficients are formatted bymtableis by default selected from thecoef.style setting ofoptions,"factory-fresh" setting beingoptions(coef.style="default").

The appearance of factor levels in anmtablecan be influenced by thefactor.style setting ofoptions.The "factory-fresh" setting isoptions(factor.style="($f): ($l)"),where($f) stands for the factor name and($l) standsfor the factor level. In case of treatment contrasts, the baseline levelwill also appear in anmtable separated from the currentfactor level by thebaselevel.sep setting ofoptions.The "factory-fresh" setting isoptions(baselevel.sep="-"),

Users may specify additional coefficient styles by a call tosetCoefTemplate.

In order to adapt the display of summary statistics of other model classes, users need toset a template for model summaries via a call tosetSummaryTemplateor to define a method of the generic functionsummaryTemplate.

Interface to Packages 'tibble' and 'haven'

Description

Aas_tibble method (as_table.data.set) allows to transform"data.set" objectsinto objects of class"tbl_df" as defined by the package"tibble".

as.item methods for objects of classes"haven_labelled"and"have_labelled_spss" allow to transform a "tibble" importedusingread_dta,read_spss, etc. from the package "haven"into an object of class"data.set".

as_haven can be used to transform"data.set" objectsinto objects of class"tbl_df" with that additional informationthat objects imported using the "haven" package usually have, i.e.variable labels and value labels (as the"label" and"labels" attributes of the columns).

Usage

as_tibble.data.set(x,...)## S4 method for signature 'haven_labelled'as.item(x,...)## S4 method for signature 'haven_labelled_spss'as.item(x,...)as_haven(x,...)## S4 method for signature 'data.set'as_haven(x,user_na=FALSE,...)## S4 method for signature 'item.vector'as_haven(x,user_na=FALSE,...)## S4 method for signature 'tbl_df'as.data.set(x,row.names=NULL,...)

Arguments

Value

as_tibble.data.set and the"data.set"-method ofas_haven return a "tibble". The"item.vector"-method(which is for internal use only) returns a vector with S3 class either"haven_labelled" or"haven_labelled_spss".

Convert an Array into a Data Frame

Description

to.data.frame converts an array into a data frame, in such a waythat a chosen dimensional extent forms variables in the data frame.The elements of the array must be either atomic, data frameswith matching variables, or coercable into such data frames.

Usage

to.data.frame(X,as.vars=1,name="Freq")

Arguments

Value

A data frame.

Examples

berkeley <- Aggregate(Table(Admit,Freq)~.,data=UCBAdmissions)berktest1 <- By(~Dept+Gender,                glm(cbind(Admitted,Rejected)~1,family="binomial"),                data=berkeley)berktest2 <- By(~Dept,                glm(cbind(Admitted,Rejected)~Gender,family="binomial"),                data=berkeley)Stest1 <- Lapply(berktest2,function(x)predict(x,,se.fit=TRUE)[c("fit","se.fit")])Stest2 <- Sapply(berktest2,function(x)coef(summary(x)))Stest2.1 <- Lapply(berktest1,function(x)predict(x,,se.fit=TRUE)[c("fit","se.fit")])to.data.frame(Stest1)to.data.frame(Stest2,as.vars=2)to.data.frame(Stest2.1)# Recasting a contingency tableto.data.frame(UCBAdmissions,as.vars="Admit")

Additional Methods for LaTeX Representations for R objects

Description

Methods for the generic functiontoLatex of package “utils”are provided for generating LaTeX representationsof matrices and flat contingency tables (seeftable). Also a default method is definedthat coerces its first argument into a matrix and appliesthe matrix method.

Usage

## Default S3 method:toLatex(object,...)## S3 method for class 'matrix'toLatex(object,    show.titles=TRUE,    show.vars=FALSE,    show.xvar=show.vars,    show.yvar=show.vars,    digits=if(is.table(object)) 0 else getOption("digits"),    format="f",    useDcolumn=getOption("useDcolumn",TRUE),    colspec=if(useDcolumn)                paste("D{.}{",LaTeXdec,"}{",ddigits,"}",sep="")             else "r",    LaTeXdec=".",    ddigits=digits,    useBooktabs=getOption("useBooktabs",TRUE),    toprule=if(useBooktabs) "\\toprule" else "\\hline\\hline",    midrule=if(useBooktabs) "\\midrule" else "\\hline",    cmidrule=if(useBooktabs) "\\cmidrule" else "\\cline",    bottomrule=if(useBooktabs) "\\bottomrule" else "\\hline\\hline",    toLatex.escape.tex=getOption("toLatex.escape.tex",FALSE),    ...)## S3 method for class 'data.frame'toLatex(object,    digits=getOption("digits"),    format="f",    useDcolumn=getOption("useDcolumn",TRUE),    numeric.colspec=if(useDcolumn)                       paste("D{.}{",LaTeXdec,"}{",ddigits,"}",sep="")                    else "r",    factor.colspec="l",    LaTeXdec=".",    ddigits=digits,    useBooktabs=getOption("useBooktabs",TRUE),    toprule=if(useBooktabs) "\\toprule" else "\\hline\\hline",    midrule=if(useBooktabs) "\\midrule" else "\\hline",    cmidrule=if(useBooktabs) "\\cmidrule" else "\\cline",    bottomrule=if(useBooktabs) "\\bottomrule" else "\\hline\\hline",    row.names=is.character(attr(object,"row.names")),    NAas="",    toLatex.escape.tex=getOption("toLatex.escape.tex",FALSE),    ...)## S3 method for class 'ftable'toLatex(object,    show.titles=TRUE,    digits=if(is.integer(object)) 0 else getOption("digits"),    format=if(is.integer(object)) "d" else "f",    useDcolumn=getOption("useDcolumn",TRUE),    colspec=if(useDcolumn)                paste("D{.}{",LaTeXdec,"}{",ddigits,"}",sep="")             else "r",    LaTeXdec=".",    ddigits=digits,    useBooktabs=getOption("useBooktabs",TRUE),    toprule=if(useBooktabs) "\\toprule" else "\\hline\\hline",    midrule=if(useBooktabs) "\\midrule" else "\\hline\n",    cmidrule=if(useBooktabs) "\\cmidrule" else "\\cline",    bottomrule=if(useBooktabs) "\\bottomrule" else "\\hline\\hline",    extrarowsep = NULL,    toLatex.escape.tex=getOption("toLatex.escape.tex",FALSE),    fold.leaders=FALSE,    ...)## S3 method for class 'ftable_matrix'toLatex(object,    show.titles=TRUE,    digits=getOption("digits"),    format="f",    useDcolumn=getOption("useDcolumn",TRUE),    colspec=if(useDcolumn)                paste("D{.}{",LaTeXdec,"}{",ddigits,"}",sep="")             else "r",    LaTeXdec=".",    ddigits=digits,    useBooktabs=getOption("useBooktabs",TRUE),    toprule=if(useBooktabs) "\\toprule" else "\\hline\\hline",    midrule=if(useBooktabs) "\\midrule" else "\\hline",    cmidrule=if(useBooktabs) "\\cmidrule" else "\\cline",    bottomrule=if(useBooktabs) "\\bottomrule" else "\\hline\\hline",    compact=FALSE,    varontop,varinfront,    groupsep="3pt",    grouprule=midrule,    toLatex.escape.tex=getOption("toLatex.escape.tex",FALSE),    multi_digits=NULL,    ...)

Arguments

Examples

toLatex(diag(5))toLatex(ftable(UCBAdmissions))toLatex(rbind(  ftable(margin.table(UCBAdmissions,c(2,1))),  ftable(margin.table(UCBAdmissions,c(3,1)))))

Trim Codes from the Labels of an Item

Description

Occasionally, labels of codes in a survey data sets (e.g. from the2016 American National Election Study) include acharacter representation of the codes being labelled. While there maybe technical reasons for this, it is often inconvenient (e.g. if onewants to reorder the labelled codes). The functiontrim_labelstrims the code representations (if they are present.)

Usage

trim_labels(x,...)## S4 method for signature 'item.vector'trim_labels(x,...)## S4 method for signature 'data.set'trim_labels(x,...)

Arguments

Details

The "data.set" method applies the "item.vector" method to all thelabelled items in the data set.

The "item.vector" returns a copy of its argument with modified labels,where a label such as "1. First alternative" is changed into "First alternative".

Examples

x <- as.item(sample(1:3,10,replace=TRUE),             labels=c("1. One"=1,                      "2. Two"=2,                      "2. Three"=3))y <- as.item(sample(1:2,10,replace=TRUE),             labels=c("1. First category"=1,                      "2. Second category"=2))ds <- data.set(x,y)x <- trim_labels(x)codebook(x)ds <- trim_labels(ds)codebook(ds)

Value Filters

Description

Value filters, that is objects that inherit from class "value.filter",are a mechanism to distinguish between valid codes of a surveyitem and codes that are considered to be missing, such asthe codes for answers like "don't know" or "answer refused".

Value filters are optional slot values of "item" objects.They determine which codes of "item" objects arereplaced byNA when they are coerced intoa vector or a factor.

There are three (sub)classes of value filters:"missing.values", which specify individualmissing values and/or a range of missing values;"valid.values", which specify individualvalid values (that is, all other values of theitem are considered as missing);"valid.range", which specify a range ofvalid values (that is, all values outside the rangeare considered as missing).Value filters of class "missing.values" correspondto missing-values declarations in SPSS files,imported byspss.fixed.file,spss.portable.file, orspss.system.file.

Value filters also can be updated using the+and- operators.

Usage

value.filter(x)missing.values(x)missing.values(x)<-valuevalid.values(x)valid.values(x)<-valuevalid.range(x)valid.range(x)<-valueis.valid(x)nvalid(x)is.missing(x)include.missings(x,mark="*")

Arguments

Value

value.filter(x),missing.values(x),valid.values(x), andvalid.range(x),return the value filter associated withx, anobject of class "value.filter", that is, of class"missing.values", "valid.values", or "valid.range", respectively.

is.missing(x) returns a logical vector indicating foreach element ofx whether it is a missing value or not.is.valid(x) returns a logical vector indicating foreach element ofx whether it is a valid value or not.nvalid(x) returns the number of elements ofxthat are valid.

For convenience,is.missing(x) andis.valid(x) also workfor atomic vectors and factors, where they are equivalent tois.na(x) and!is.na(x). For atomic vectors and factors,nvalid(x) returns the number of elements ofx forwhich!is.na(x) is TRUE.

include.missings(x,...) returns a copy ofxthat has all values declared as valid.

Examples

x <- rep(c(1:4,8,9),2,length=60)labels(x) <- c(    a=1,    b=2,    c=3,    d=4,    dk=8,    refused=9    )missing.values(x) <- 9missing.values(x)missing.values(x) <- missing.values(x) + 8missing.values(x)missing.values(x) <- NULLmissing.values(x)missing.values(x) <- list(range=c(8,Inf))missing.values(x)valid.values(x)print(x)is.missing(x)is.valid(x)as.factor(x)as.factor(include.missings(x))as.integer(x)as.integer(include.missings(x))

A Generic Viewing Function

Description

The functionview provides generic interface to the non-genericfunctionView.

In contrast to the implementation ofView provided by eitherbasicR orRStudio, this function can be extended tohandle new kinds of objects by definingviewPrep methods forthem. Further,view can be adapted to other GUIs by specifyingthe"vfunc" option or thevfunc= optional argument.

Internally,view usues the generic functionviewPrepto prepare data so it can be passed on to the (non-generic) functionView or (optionally) a different graphical user interfacefunction that can be used to display matrix- or data frame-likeobjects.

Thevfunc argument determines how the result ofviewPrepis displayed. Its default is the functionView, but analternative isview_html which creates and displays an HTML grid.

Usage

view(x,     title=deparse(substitute(x)),     vfunc=getOption("vfunc","View"),     ...)# The internal generic, not intended to be used by the end-user.viewPrep(x,title,...)## S3 method for class 'data.set'viewPrep(x,title,...)## S3 method for class 'data.frame'viewPrep(x,title,...)## S3 method for class 'descriptions'viewPrep(x,title,...)## S3 method for class 'codeplan'viewPrep(x,title,compact=FALSE,...)## S3 method for class 'importer'viewPrep(x,title,compact=TRUE,...)

Arguments

Examples

## Not run:     example(data.set)    view(Data)    view(description(Data))    view(codeplan(Data))    # Note that this file is *not* included in the package    # and has to be obtained from GESIS in order to run the     # following    ZA7500sav <- spss.file("ZA7500_v2-0-0.sav")    view(ZA7500sav)## End(Not run)

HTML Output for 'view.

Description

An alternative to 'View' for use with 'view'.

Usage

view_html(x,title=deparse(substitute(x)),output,...)

Arguments

Examples

## Not run:     example(data.set)    view(Data,vfunc=view_html)## End(Not run)

Table of frequencies for unlabelled codes

Description

The functionwild.codes creates a table of frequencies of those codesof an item that do not have labelled attached to them. This way, it helps to identifycoding errors.

Usage

wild.codes(x)## S4 method for signature 'item'wild.codes(x)

Arguments

Value

A table of frequencies (i.e. an array of class "table")

Add Alternative Variance Estimates to Models Estimates

Description

A simple object-orientation infrastructure to add alternative standarderrors, e.g. sandwich estimates or New-West standard errors to fitted regression-type models, such as fitted bylm() orglm().

Usage

withSE(object, vcov, ...) withVCov(object, vcov, ...)## S3 method for class 'lm'withVCov(object, vcov, ...)## S3 method for class 'withVCov'summary(object, ...)## S3 method for class 'withVCov.lm'summary(object, ...)

Arguments

Details

UsingwithVCov() an alternative variance-covariance matrix isattributed to a fitted model object. Such a matrix may be produced byany of the variance estimators provided by the "sandwich" package orany package that extends it.

withVCov() has no consequences on how a fitted model itself isprinted or represented, but it does have consequences what standarderrors are reported, when the functionsummary() or the functionmtable() is applied.

withSE() is a convenience front-end towithVCov(). It canbe called in the same way aswithVCov, but also allows to specifythe type of variance estimate by a character string that identifiesthe function that gives the covariance matrix (e.g."OPG" forvcovOPG).

Value

withVCov returns a slightly modified model object: It adds anattribute named ".VCov" that contains the alternate covaraince matrixand modifies the class attribute. If e.g. the original model object has class"lm" then the model object modified bywithVCov has the classattributec("withVCov.lm", "withVCov", "lm").

Examples

## Generate poisson regression relationshipx <- sin(1:100)y <- rpois(100, exp(1 + x))## compute usual covariance matrix of coefficient estimatesfm <- glm(y ~ x, family = poisson)library(sandwich)fmo <- withVCov(fm,vcovOPG)vcov(fm)vcov(fmo)summary(fm)summary(fmo)mtable(Default=fm,       OPG=withSE(fm,"OPG"),       summary.stats=c("Deviance","N")       )vo <- vcovOPG(fm)mtable(Default=fm,       OPG=withSE(fm,vo),       summary.stats=c("Deviance","N")       )

Operators to abbreviate use of "with" and "within"

Description

The operators%$% and%$$% provideabbrevitions for calls towith() andwithin()respectively.The functionWithin() is a variant ofwith() werethe resulting data frame contains any newly created variables in theorder in which they are created (and not in the reverse order).

Usage

data %$% exprdata %$$% exprWithin(data,expr,...)## S3 method for class 'data.frame'Within(data,expr,...)

Arguments

See Also

with andwithin in package "base".

Examples

df <- data.frame(a = 1:7,                 b = 7:1)dfdf <- within(df,{  ab <- a + b  a2b2 <- a^2 + b^2})dfdf <- data.frame(a = 1:7,                 b = 7:1)df <- Within(df,{  ab <- a + b  a2b2 <- a^2 + b^2})dfdf <- data.frame(a = 1:7,                 b = 7:1)dfds <- as.data.set(df)dsdf %$$% {  ab <- a + b  a2b2 <- a^2 + b^2}dfds %$$% {  ab <- a + b  a2b2 <- a^2 + b^2}dsdf %$% c(a.ssq = sum(a^2),         b.ssq = sum(b^2))

Apply a function to ranges of variables

Description

xapply evaluates an expression given as second argument by substitutingin variables. The results are collected in a list or array in asimilar way as done bySapply orlapply.

Usage

  xapply(...,.sorted,simplify=TRUE,USE.NAMES=TRUE,.outer=FALSE)

Arguments

Examples

x <- 1:3y <- -(1:3)z <- c("Uri","Schwyz","Unterwalden")print(x)print(y)print(z)foreach(var=c(x,y,z),          # assigns names  names(var) <- letters[1:3]   # to the elements of x, y, and z  )print(x)print(y)print(z)ds <- data.set(        a = c(1,2,3,2,3,8,9),        b = c(2,8,3,2,1,8,9),        c = c(1,3,2,1,2,8,8)      )print(ds)ds <- within(ds,{       description(a) <- "First item in questionnaire"      description(b) <- "Second item in questionnaire"      description(c) <- "Third item in questionnaire"            wording(a) <- "What number do you like first?"      wording(b) <- "What number do you like second?"      wording(c) <- "What number do you like third?"      foreach(x=a:c,{ # Lazy data documentation:        labels(x) <- c(    # a,b,c get value labels in one statement                         one = 1,                         two = 2,                       three = 3,                "don't know" = 8,         "refused to answer" = 9)        missing.values(x) <- c(8,9)        })      })      codebook(ds)# The colon-operator respects the order of the variables# in the data set, if .sorted=FALSEwith(ds[c(3,1,2)],     xapply(x=a:c,            description(x)            ))# Since .sorted=TRUE, the colon operator creates a range # of alphabetically sorted variables.with(ds[c(3,1,2)],     xapply(x=a:c,            description(x),            .sorted=TRUE            ))# The variables in reverse orderwith(ds,     xapply(x=c:a,             description(x)            ))# The colon operator can be combined with the # concatenation functionwith(ds,     xapply(x=c(a:b,c,c,b:a),             description(x)            ))# Variables can also be selected by regular expressions.with(ds,     xapply(x=rx("[a-b]"),             description(x)            ))# Demonstrating the effects of the 'USE.NAMES' argument.with(ds,     xapply(x=a:c,mean(x)))with(ds,     xapply(x=a:c,mean(x),     USE.NAMES=FALSE))t(with(ds,      xapply(i=1:3,             x=a:c,             c(Index=i,               Mean=mean(x)),      USE.NAMES=2)))# Result with 'simplify=FALSE'with(ds,     xapply(x=a:c,mean(x),     simplify=FALSE))# It is also possible to loop over functions:xapply(fun=c(exp,log),       fun(1))# Two demonstrations for '.outer=TRUE'with(ds,      xapply(x=a:c,             y=a:c,             cov(x,y),             .outer=TRUE))with(ds,      xapply(x=a:c,             y=a:c,             fun=c(cov,cor),             fun(x,y),             .outer=TRUE))