Configural Frequencies Analysis Using Log-linear Modeling

Description

The package confreq offers some functions for Configural Frequencies Analysis (CFA) proposed by G.A. Lienert as an analysis of types and antitypes of persons or objects grouped according to their characteristic (response) pattern. The core principle in the package confreq is to use the function glm to compute the expected counts based on a model (design) matrix. The main functions are CFA and S2CFA (see details).

Details

The simplest entry to the package confreq is to use the main function CFA, which will compute several coefficients of Configural Frequencies Analysis at once.

More sophisticated control can be achieved by using the several single functions like expected_cfa, design_cfg_cfa, chi_local_test_cfa , stirling_cfa , etc. ...

Two-Sample-CFA, to detect discriminating pattern between two (sub-) samples, can be performed with the function S2CFA

For further description see description of the respective functions.

A good introduction into the theory and applications of Configural Frequencies Analysis is given in the Textbook 'Person-Centered Methods' by Mark Stemmler (see references).

Additional Information: Some users running R on 'Linux like' OS distributions (like e.g. Ubuntu – and in rare cases MAC OS) might report trouble during installation of confreq due to the package dependency gmp, which is used in confreq to perform the exact binomial test. This (miss-)behavior can usually traced back to a missing of 'the GNU Multiple Precision Arithmetic Library' in the respective OS installation. To fix this, users might consider to run the following Ubuntu Linux command in a terminal to install the latest GMP Library:

'sudo apt-get install libgmp3-dev'.

For additional information see also http://www.mathemagix.org/www/mmdoc/doc/html/external/gmp.en.html and https://gmplib.org/.

Annotation: The foundations for this R-Package were established and discussed in Rothenberge (2011) and (finally) in Klagenfurt at FGME 2013 with Rainer Alexandrowicz and Mark Stemmler ...

Author(s)

• Joerg-Henrik Heine <jhheine@googlemail.com>

• R.W. Alexandrowicz (function stirling_cfa())

References

von Eye, A. (2002). Configural Frequency Analysis. Methods, Models, and Applications. Mahwah, NJ, LEA.

Krauth, J., & Lienert, G. A. (1973). Die Konfigurationsfrequenzanalyse (KFA) und ihre Anwendung in Psychologie und Medizin: ein multivariates nichtparametrisches Verfahren zur Aufdeckung von Typen und Syndromen; mit 70 Tabellen. Freiburg; München: Alber Karl.

Lazarsfeld, P. F., & Henry, N. W. (1968). Latent structure analysis. Boston: Houghton Mifflin.

Lienert, G. A. (1978). Verteilungsfreie Methoden in der Biostatistik (Band II) [Non-parametrical methods in the field of biometrics (Vol. II)]. Meisenheim am Glan, Germany: Hain.

Lienert, G. A. (1971). Die Konfigurationsfrequenzanalyse: I. Ein neuer Weg zu Typen und Syndromen. Zeitschrift für Klinische Psychologie und Psychotherapie, 19(2), 99-115.

Stemmler, M. (2020). Person-Centered Methods – Configural Frequency Analysis (CFA) and Other Methods for the Analysis of Contingency Tables. Cham Heidelberg New York Dordrecht London: Springer.

Stemmler, M., & Hammond, S. (1997). Configural frequency analysis of dependent samples for intra-patient treatment comparisons. Studia Psychologica, 39, 167–175.

Examples

#######################################
######### some examples ########
data(LienertLSD)
LienertLSD
CFA(LienertLSD)
## testing with (full) interactions
CFA(LienertLSD,form="~ C + T + A + C:T + C:A + T:A + C:T:A")
## testing the null model
CFA(LienertLSD,form="null")
#######################
data(suicide)
suicide
# suicide data is in non tabulated data representation 
# so it must be tabulated !
CFA(dat2fre(suicide))

Configural Frequencies Analysis Main Function

Description

Calculates various coefficients for the Configural Frequencies Analysis (CFA) defining main- and (optional) interaction effects. The core principle is to use glm in package stats to calculate the expected counts considering a designmatrix, which is constructed based on an formula definition given in argument form.

Usage

CFA(
  patternfreq,
  alpha = 0.05,
  form = NULL,
  ccor = FALSE,
  family = poisson(),
  intercept = FALSE,
  method = "log",
  blank = NULL,
  cova = NULL,
  bintest = TRUE,
  ...
)

Arguments

Details

This is the main function of the package. It internal calls several functions of the package confreq-package which are also available as single functions. For classification of the observed patterns into 'Types' and 'Antitypes' according to Lienert (1971), a S3 summary method for the resulting object of class "CFA" can be applied - see summary.CFA. An S3 plot method is useful for visualization of the contingency table and the 'Types' and 'Antitypes' – see plot.CFA. Since version 1.6.0-1 of confreq survey weights are supported when tabluating a data set with function dat2fre. In case that for the resulting tabulated data in the object of class c("data.frame","Pfreq") survey weights were used the function CFA will take into account those weigts for estimation of the expected counts – currently only when method="log".

Value

an object of class CFA with results.

References

Lienert, G. A. (1971). Die Konfigurationsfrequenzanalyse: I. Ein neuer Weg zu Typen und Syndromen. Zeitschrift für Klinische Psychologie und Psychotherapie, 19(2), 99-115.

Glück, J., & Von Eye, A. (2000). Including covariates in Configural Frequency Analysis. Psychologische Beitrage, 42, 405–417.

Victor, N. (1989). An Alternativ Approach to Configural Frequency Analysis. Methodika, 3, 61–73.

Stemmler, M. (2020). Person-Centered Methods. Cham: Springer International Publishing.

Examples

#######################################
######### some examples ########
data(LienertLSD)
LienertLSD
res1 <- CFA(LienertLSD)
summary(res1)
## testing with (full) interactions
res2 <- CFA(LienertLSD,form="~ C + T + A + C:T + C:A + T:A + C:T:A")
summary(res2)
#' ## testing the null model
res3 <- CFA(LienertLSD,form="null")
summary(res3)
#######################
data(suicide)
suicide
# suicide data is in non tabulated data representation - so it must be tabulated !
res4 <- CFA(dat2fre(suicide))  
summary(res4)

The Lienert (1978) Data

Description

Data used as an example for two-sample CFA in the textbook by Mark Stemmler (2020, Table 7.7, p.97) taken from Lienert (1978, p. 978). The data is in pattern frequencies table representation (object of class c("data.frame", "Pfreq" )).

Usage

  data(Lienert1978)

Format

A data frame (object of class c("data.frame", "Pfreq") ) with 4 columns and 12 rows. The last column gives the frequencies for the (response) pattern in column 1:2 of the respective 'Group' given in column 3.

Details

no details at the moment ...

References

Lienert, G. A. (1978). Verteilungsfreie Methoden in der Biostatistik (Band II) [Non-parametrical 168 methods in the field of biometrics (Vol. II)]. Meisenheim am Glan, Germany: Hain.

Stemmler, M. (2020). Person-Centered Methods – Configural Frequency Analysis (CFA) and Other Methods for the Analysis of Contingency Tables. Cham Heidelberg New York Dordrecht London: Springer.

Examples

data(Lienert1978)
dim(Lienert1978)
##############################################################
colnames(Lienert1978) # show all variable names of Lienert1978

The Lienert LSD Data

Description

Data from the classical Lienert LSD trial as an example for CFA (see Lienert, 1971, p. 103, ’Tabelle 1’). The data is in pattern frequencies table representation (object of class c("data.frame", "Pfreq" )).

Usage

  data(LienertLSD)

Format

A data frame (object of class c("data.frame", "Pfreq") ) with 4 columns and 8 rows. The last column gives the frequencies for the observed pattern of the psychotoxic basic syndrome (in column 1:3) due to the intake of lysergic acid diethylamide (LSD).

Details

The first three columns are named C, T and A which are abbreviations for the observed symptoms after taking LSD:

C = narrowed consciousness

T = thought disturbance

A = affective disturbance

The coding of the observations is for all symptoms: present='+' and absent='-'

References

Lienert, G. A. (1971). Die Konfigurationsfrequenzanalyse: I. Ein neuer Weg zu Typen und Syndromen. Zeitschrift für Klinische Psychologie und Psychotherapie, 19(2), 99-115.

Examples

data(LienertLSD)
dim(LienertLSD)
##############################################################
colnames(LienertLSD) # show all variable names of matrix LienertLSD

Configural Frequencies Analysis for two Samples.

Description

Calculates coefficients for the two-sample CFA. Instead of differentiating between 'Types' and 'Antitypes', two-sample CFA looks for discrimination types, that is configurations with significant differences in frequencies between two sub samples.

Usage

S2CFA(patternfreq, alpha = 0.05, ccor = FALSE, ...)

Arguments

Details

no details at the moment ...

Value

an object of class S2CFA with results.

References

Stemmler, M. (2020). Person-Centered Methods – Configural Frequency Analysis (CFA) and Other Methods for the Analysis of Contingency Tables. Cham Heidelberg New York Dordrecht London: Springer.

Stemmler, M., & Hammond, S. (1997). Configural frequency analysis of dependent samples for intra-patient treatment comparisons. Studia Psychologica, 39, 167–175.

Examples

#######################################
############### some examples #########
######### example from Marks Textbook
data(Lienert1978)
res1 <- S2CFA(Lienert1978)
summary(res1)
res2 <- S2CFA(Lienert1978, ccor=TRUE) # with continuity correction
summary(res2)
######### example with biger numbers
data(suicide)
ftab(suicide) # 'Epoche' may divide the sample into 2 subsamples  
suicide_2s <- suicide[, c(1,3,2) ] # reorder data that 'Epoche' is the last column
ftab(suicide_2s) # check reordering
suicide_2s_fre <- dat2fre(suicide_2s)
res3 <- S2CFA(suicide_2s_fre)
summary(res3)
res4 <- S2CFA(suicide_2s_fre, ccor=TRUE) # with continuity correction
summary(res4)

Binomial Test

Description

Calculates the (exact) binomial test based on obseved, expected frequencies an the total number of observations.

Usage

binomial_test_cfa(observed, expected, ntotal = sum(observed))

Arguments

Details

No details

Value

a numeric giving the p-value.

References

No references in the moment

Examples

#######################################
# first calculate expected counts for LienertLSD data example.
designmatrix<-design_cfg_cfa(kat=c(2,2,2)) # generate an designmatrix (only main effects)
data(LienertLSD) # load example data
observed<-LienertLSD[,4] # extract observed counts
expected<-expected_cfa(des=designmatrix, observed=observed) # calculation of expected counts
 binomial_test_cfa(observed,expected)
####################################### 

Local Chi-Square Test

Description

Calculates the local chi-square test based on obseved and expected frequencies.

Usage

chi_local_test_cfa(observed, expected)

Arguments

Details

No details in the moment.

Value

a list with chi-square statistic and corresponding degrees of freedom an p-value.

References

No references in the moment

Examples

#######################################
# first calculate expected counts for LienertLSD data example.
designmatrix<-design_cfg_cfa(kat=c(2,2,2)) # generate an designmatrix (only main effects)
data(LienertLSD) # load example data
observed<-LienertLSD[,4] # extract observed counts
expected<-expected_cfa(des=designmatrix, observed=observed) # calculation of expected counts
 chi_local_test_cfa(observed,expected)
####################################### 

S3 coefficients for CFA

Description

S3 coefficients method for object of class"CFA".

Usage

## S3 method for class 'CFA'
coef(object, ...)

Arguments

Value

Coefficients extracted from the model object of class"CFA".

conversion of a covariate dataset into summary covariate values

Description

Given a dataset x, this function returns summary values for some (numeric) covariate variables in x for each pattern (configuration) defined by a set of factor variables in x.

Usage

dat2cov(
  x,
  FUN = "mean",
  ...,
  notobs = 0,
  katorder = FALSE,
  caseorder = TRUE,
  wgt = NULL
)

Arguments

Details

No further details

Value

An object of class c("data.frame", "Pcov") holding the summary statistics for the covariate variables corresponding to the pattern (configurations) of the given dataset in the argument x.

dataset to pattern frequency conversion

Description

Given a dataset this function returns a (response) pattern frequencies table representation of it.

Usage

dat2fre(
  x,
  katorder = FALSE,
  caseorder = TRUE,
  kat = NULL,
  codes = NULL,
  wgt = NULL,
  ...
)

Arguments

Details

To use survey weights a vector of positive numeric values with length matching the number of rows in x must be assigned to the argument wgt. The individual case weights are then aggregated (respective sum of weights) for each pattern observed in the data (assigned to argument x).

Value

An object of class c("data.frame","Pfreq") containing the (response) pattern frequencies table representation of the given dataset in the argument x.

References

No references in the moment

Examples

#######################################
data(suicide)# loading data in data frame (702 cases) representation 
dat2fre(suicide) # converting it into a pattern frequencies table

########### 
#######################################
data(LienertLSD)# loading example pattern frequencies table ..
test<-fre2dat(LienertLSD)# and coverting it into a simple (data) matrix
test<-test[sample(c(1:65),65),] # making a messy order 
############
dat2fre(test) # making a proper ordered pattern frequencies table again
##### try it with a data.frame too!
#######################################

Designmatrix for log linear CFA models

Description

Calculates the designmatrix corresponding to a dataset with length(kat) columns (variables).

Usage

design_cfg_cfa(
  kat,
  form = paste("~", paste(paste("V", 1:length(kat), sep = ""), collapse = " + ")),
  ...
)

Arguments

Details

This function internaly calls the function pos_cfg_cfa.

For further information on designmatrices see decription on function model.matrix in the package stats.

Value

A designmatrix - an object of class c("matrix","design_cfg_cfa") - for the formula therm given in argumentform.

References

No references in the moment

Examples

#######################################
# designmatrix with three main effects.
# three variables with two categories each.
design_cfg_cfa(kat=c(2,2,2))
# two variables with two categories each and one variable
# with 7 categories (Lienert LSD example).
design_cfg_cfa(kat=c(2,2,7))
###########
# designmatrix with three main effects an three interactions.
# three variables with two categories each.
design_cfg_cfa(kat=c(2,2,2),form="~ V1 + V2 + V3 + V1:V2 + V1:V3 + V2:V3")
# two variables with two categories each and one variable
# with 7 categories (e.g. The Krauth & Lienert suicide Data).
design_cfg_cfa(kat=c(2,2,7),form="~ V1 + V2 + V3 + V1:V2 + V1:V3 + V2:V3")
#######################################

Degrees of freedom

Description

Calculates the degrees of freedom based on an designmatrix for a (log liniear) CFA model.

Usage

df_des_cfa(des)

Arguments

Details

No details

Value

An object of class "integer" giving the degrees of freedom for the designmatrix defined in argument des.

References

No references in the moment

Examples

#######################################
# degrees of freedom for designmatrix with three main effects.
# three variables with two categories each.
df_des_cfa(design_cfg_cfa(kat=c(2,2,2)))
# two variables with two categories each and one variable
# with 7 categories (e.g. The Krauth & Lienert suicide Data).
df_des_cfa(design_cfg_cfa(kat=c(2,2,7)))
###########
# degrees of freedom for designmatrix with three main effects
# and three 'two by two'interactions.
# and tripple interaction --> saturated model --> df=0
# three variables with two categories each.
df_des_cfa(design_cfg_cfa(kat=c(2,2,2),form="~ V1 + V2 + V3 + V1:V2 + V1:V3 + V2:V3 + V1:V2:V3"))
####################################### 

Expected frequencies with glm

Description

Calculates the expected frequencies of counts using log liniear model.

Usage

expected_cfa(des, observed, family = poisson(), intercept = FALSE, ...)

Arguments

Details

No details

Value

An vector object giving the expected counts.

References

No references in the moment

Examples

#######################################
# expected counts for LienertLSD data example.
designmatrix<-design_cfg_cfa(kat=c(2,2,2)) # generate an designmatrix (only main effects)
data(LienertLSD) # load example data
observed<-LienertLSD[,4] # extract observed counts
expected_cfa(des=designmatrix, observed=observed) # calculation of expected counts
####################################### 

Expected frequencies using margins

Description

Calculates the expected frequencies of counts based on the margins of the k-dimensional contingency table.

Usage

expected_margin_cfa(Pfreq, blank = NULL)

Arguments

Details

only main effects are considered.

Value

An vector object giving the expected counts.

References

No references in the moment

Examples

#######################################
# expected counts for LienertLSD data example.
data(LienertLSD) # load example data
expected_margin_cfa(Pfreq = LienertLSD) # calculation of expected counts (only main effects).
####################################### 

pattern frequency to dataset conversion

Description

Given a (response) pattern frequencies table this function returns a dataset representation of it.

Usage

fre2dat(x, fact = FALSE, ...)

Arguments

Details

No details

Value

An object of class "matrix" or "data.frame" (depending on the argument fact) containing the dataset representation of the (response) pattern frequencies table give in the argument x.

References

No references in the moment

Examples

#######################################
data(LienertLSD)# loading example pattern frequencies table
fre2dat(LienertLSD)# coverting it into a (data) matrix
# for a matrix without colnames
colnames(LienertLSD)<-NULL # first removing the colnames
fre2dat(LienertLSD) # conversion with automatic new colnames
# requesting a data.frame using factor levels 
fre2dat(LienertLSD,fact=TRUE,labels=c("yes","no"))

pattern frequency to table conversion

Description

Given data as pattern frequencies (object of class class c("data.frame","Pfreq", see function dat2fre) this function returns a typical array representation (class "table" , see table) of it.

Usage

fre2tab(patternfreq, form = NULL)

Arguments

Details

This function was introduced in order to connect the typical confreq data representation in the objects of the class c("data.frame","Pfreq"), see function dat2fre, to the R-typical array representation as it exists in objects of the "table" class, see table. This array representation of multi-dimensional contingency tables is used more universally in R – e.g. also in the R package vcd, see the examples section below.

It is assumed, that the last column of the object patternfreq represents the frequencies of the (response) pattern represented by the other columns in patternfreq.

Value

An object of class "table" see table.

References

No references at the moment

Examples

#######################################
data(LienertLSD)# loading example pattern frequencies table
fre2tab(LienertLSD)# coverting it into a table

### examples using functions from package vcd
data(Lienert1978)# loading example pattern frequencies table
fre2tab(Lienert1978)# coverting it into a table
strucplot(fre2tab(Lienert1978))# plotting data with 'vcd'
structable(fre2tab(Lienert1978),direction = "v")# flatten table (vertical) with 'vcd'

# changing the vertical grouping when flattening the table by unsing a 'formula':
structable(fre2tab(Lienert1978, form=~Group + Student + Teacher),direction = "v")# flatten table

Tabulating Answer Categories in Data

Description

Function tabulating (answer) categories in X.

Usage

ftab(X, catgories = NULL, na.omit = FALSE)

Arguments

Details

X can either be a ("numeric" or "character") "matrix" containing response vectors of persons (rows) or a "data.frame" containing "numeric", "character" or "factor" variables (columns).

Value

a "matrix" with category frequencies

Examples

########
data(suicide)
ftab(suicide)

The Data Example from Lazarsfeld and Henry

Description

data example by Lazarsfeld and Henry (1968) where n = 1000 subjects need to solve questions or problems (i.e., A,B, and C). They either ‘1’ = solved or ‘2’ = did not solve the problems. The data is in pattern frequencies table representation (object of class c("data.frame", "Pfreq" )). This data example is used in the textbook by Mark Stemmler (2020, Table 6.6, p. 81).

Usage

  data(lazar)

Format

A matrix with 4 columns and 8 rows. The last column gives the frequencies for the (response) pattern in column 1:3.

Details

No detail in the moment

References

Lazarsfeld, P. F., & Henry, N. W. (1968). Latent structure analysis. Boston: Houghton Mifflin.

Stemmler, M. (2020). Person-Centered Methods – Configural Frequency Analysis (CFA) and Other Methods for the Analysis of Contingency Tables. Cham Heidelberg New York Dordrecht London: Springer.

Examples

##############################################################
data(lazar)
dim(lazar)
##############################################################

Likelihood Ratio Chi-square (LR)

Description

Calculates the likelihod ratio chi-square statistic based on observed and expected counts.

Usage

lr(observed, expected)

Arguments

Details

No details in the moment.

Value

numeric giving the likelihood ratio chi-square statistic.

References

Stemmler, M. (2014). Person-Centered Methods – Configural Frequency Analysis (CFA) and Other Methods for the Analysis of Contingency Tables. Cham Heidelberg New York Dordrecht London: Springer.

Examples

#######################################
######### some examples ########
data(newborns)
newborns
designmatrix <- design_cfg_cfa(kat=c(2,2)) # generate an designmatrix (only main effects)
observed <- newborns[,3] # extract observed counts
expected <- expected_cfa(des=designmatrix, observed=observed) # calculation of expected counts
lr(observed,expected) # calculation of the likelihood ratio chi-square statistic

The Data Example from Stemmler 2020

Description

data example by Stemmler (2020, table 4.1, p. 33) where n = 56 newborns 'with seizures' = 1 or 'without seizures' = 2 (coded in the in first column named 'A') were tested with an intelligence test while they attended kindergarten. Children’s intelligence was divided into 'average or above' = 1 and 'below average' = 2 (coded in the in second column named 'B'). The third column gives the frequencies of the respective pattern.

Usage

  data(newborns)

Format

A data.frame with 3 columns and 4 rows. The last column gives the frequencies for the observed pattern in column 1:2. The data is in pattern frequencies table representation (object of class c("data.frame", "Pfreq"))

Details

No detail in the moment

References

Stemmler, M. (2020). Person-Centered Methods – Configural Frequency Analysis (CFA) and Other Methods for the Analysis of Contingency Tables. Cham Heidelberg New York Dordrecht London: Springer.

Examples

##############################################################
data(newborns)
dim(newborns)
newborns
##############################################################

S3 plot for CFA

Description

S3 plot method for object of class"CFA"

Usage

## S3 method for class 'CFA'
plot(
  x,
  type = "z.pChi",
  fill = c("red", "blue", "grey"),
  adjalpha = "bonferroni",
  ...
)

Arguments

Value

a plot visualizing the results.

References

Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6(2), 65–70.

Bonferroni, C. E. (1935). Il calcolo delle assicurazioni su gruppi di teste. In S.O. Carboni (Ed.), Studi in Onore del Professore Salvatore Ortu Carboni (S. 13–60). Roma, Tipografia del Senato: Bardi.

S3 plot for S2CFA

Description

S3 plot method for object of class"S2CFA"

Usage

## S3 method for class 'S2CFA'
plot(
  x,
  type = "ex.fisher.test",
  fill = c("red", "grey"),
  adjalpha = "bonferroni",
  ...
)

Arguments

Value

a plot visualizing the results.

References

Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6(2), 65–70.

Bonferroni, C. E. (1935). Il calcolo delle assicurazioni su gruppi di teste. In S.O. Carboni (Ed.), Studi in Onore del Professore Salvatore Ortu Carboni (S. 13–60). Roma, Tipografia del Senato: Bardi.

Possible configurations

Description

Calculates all possible configuartions for some variables with different numbers of categories.

Usage

pos_cfg_cfa(kat, fact = FALSE)

Arguments

Details

No details

Value

An object of class "matrix" or "data.frame" (depending on the argument fact) containing all possible configurations for lenght(kat) variables with the respective number of categories given as kardinal numbers in the vector kat.

References

No references in the moment

Examples

#######################################
# possible configurations for ...
# three variables with two categories each (Lienert LSD example).
pos_cfg_cfa(kat=c(2,2,2))
#######################################

S3 print for Pfreq

Description

S3 print method for object of class"Pfreq"

Usage

## S3 method for class 'Pfreq'
print(
  x,
  ...,
  digits = NULL,
  quote = FALSE,
  right = TRUE,
  row.names = TRUE,
  max = NULL
)

Arguments

Value

output printed to the console

Approximation to the binomial using Stirling's Formula

Description

Calculates the binomial aproximation using stirling's formula (Version of function: V 1.0 - November 2013)

Usage

stirling_cfa(
  observed,
  expected = NULL,
  n = sum(observed),
  p = NULL,
  cum = T,
  verb = T
)

Arguments

Details

• Vector p must be of same length as observed _or_ p may be a scalar (e.g. in case of the zero-order CFA).

• The routine autoselects the upper or lower tail:

  • if obs > exp then sum obs:n

  • else sum 0:obs

• The stirling approximation cannot be evaluated if the observed frequency is 0 or n. Therefore, the proposal of A. von Eye (20xx) is adopted, taking the sum up to 1 or n-1, respectively.

Author(s)

R.W. Alexandrowicz

References

von Eye, A. (2002). Configural Frequency Analysis. Methods, Models, and Applications. Mahwah, NJ, LEA.

The Krauth & Lienert suicide Data

Description

Data from the Krauth & Lienert suicide example for CFA (see Tables 39a and 39b; Krauth & Lienert, 1973). The data describe suicide patterns in pre- and post-WWII Germany – see von Eye, A., (2002); p. 385.

Usage

  data(suicide)

Format

A data.frame with 3 columns (as factors). The data is in data list representation – each row represents one case.

Details

The three columns are named 'Geschlecht', 'Epoche' and 'Suizidart' which is 'gender', 'epoch' and 'type od suicide'. each of the variables are factors with the following levels:

Geschlecht: 'm' = 1 (male); 'w' = 2 (female)

Epoche: '44' = 1 (the epoch 1944); '52' = 2 (the epoch 1952)

Suizidart: 'Eh' = 1(hang); 'Es' = 2 (shoot); 'Et' = 3(drown); 'G' = 4(gas); 'H' = 5(crashing down); 'P' = 6(open vein); 'S' = 7(barbiturate);

References

Krauth, J., & Lienert, G. A. (1973). Die Konfigurationsfrequenzanalyse (KFA) und ihre Anwendung in Psychologie und Medizin: ein multivariates nichtparametrisches Verfahren zur Aufdeckung von Typen und Syndromen; mit 70 Tabellen. Freiburg; München: Alber Karl.

von Eye, A. (2002). Configural Frequency Analysis: Methods, models, and applications. Mahwah, N.J.: Lawrence Erlbaum Associates.

Examples

##############################################################
data(suicide) # to load the data.frame included in the package
class(suicide)
dim(suicide)
str(suicide)

S3 Summary for CFA

Description

S3 summary method for object of class"CFA"

Usage

## S3 method for class 'CFA'
summary(
  object,
  digits = 3,
  type = "z.pChi",
  sorton = NULL,
  decreasing = FALSE,
  showall = TRUE,
  holm = FALSE,
  wide = FALSE,
  adjalpha = "bonferroni",
  ...
)

Arguments

Value

a summary of the results printed on the console.

References

Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6(2), 65–70.

Bonferroni, C. E. (1935). Il calcolo delle assicurazioni su gruppi di teste. In S.O. Carboni (Ed.), Studi in Onore del Professore Salvatore Ortu Carboni (S. 13–60). Roma, Tipografia del Senato: Bardi.

S3 Summary for S2CFA

Description

S3 summary method for object of class"S2CFA"

Usage

## S3 method for class 'S2CFA'
summary(
  object,
  digits = 3,
  type = "ex.fisher.test",
  sorton = NULL,
  decreasing = FALSE,
  showall = TRUE,
  adjalpha = "bonferroni",
  ...
)

Arguments

Value

a summary of the results printed on the console.

References

Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6(2), 65–70.

Bonferroni, C. E. (1935). Il calcolo delle assicurazioni su gruppi di teste. In S.O. Carboni (Ed.), Studi in Onore del Professore Salvatore Ortu Carboni (S. 13–60). Roma, Tipografia del Senato: Bardi.

Two z-Approximation Tests

Description

Calculates the Chi-square approximation to the z-test and the binomial approximation to the z-test.

Usage

z_tests_cfa(observed, expected, ccor = FALSE, ntotal = sum(observed))

Arguments

Details

An continuity correction can be applied to the binomial approximation – see argument ccor.

Value

a list with z and p-values.

References

No references in the moment

Examples

#######################################
# expected counts for LienertLSD data example.
designmatrix<-design_cfg_cfa(kat=c(2,2,2)) # generate an designmatrix (only main effects)
data(LienertLSD) # load example data
observed<-LienertLSD[,4] # extract observed counts
expected<-expected_cfa(des=designmatrix, observed=observed) # calculation of expected counts
z_tests_cfa(observed,expected)
#######################################