| Title: | An Implementation of Conjoint Analysis Method |
|---|---|
| Description: | This is a simple R package that allows to measure the stated preferences using traditional conjoint analysis method. |
| Authors: | Andrzej Bak and Tomasz Bartlomowicz <[email protected]>, <[email protected]> |
| Maintainer: | Tomasz Bartlomowicz <[email protected]> |
| License: | GPL (>= 2) |
| Version: | 1.41 |
| Built: | 2024-12-11 07:14:34 UTC |
| Source: | CRAN |
Function caBTL estimates participation of simulation profiles using probabilistic model BTL (Bradley-Terry-Luce). Function returns vector of percentage participations. The sum of participation should be 100%.
caBTL(sym, y, x)caBTL(sym, y, x)
sym |
matrix of simulation profiles |
y |
matrix of preferences |
x |
matrix of profiles |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caLogit, caMaxUtility and ShowAllSimulations
#Example 1 library(conjoint) data(tea) simutil<-caBTL(tsimp,tpref,tprof) print("Percentage participation of profiles: ", quote=FALSE) print(simutil) #Example 2 library(conjoint) data(chocolate) simutil<-caBTL(csimp,cpref,cprof) print("Percentage participation of profiles:", quote=FALSE) print(simutil) #Example 3 library(conjoint) data(chocolate) ShowAllSimulations(csimp,cpref,cprof) #Example 4 #library(conjoint) #data(journey) #ShowAllSimulations(jsimp,jpref,jprof)#Example 1 library(conjoint) data(tea) simutil<-caBTL(tsimp,tpref,tprof) print("Percentage participation of profiles: ", quote=FALSE) print(simutil) #Example 2 library(conjoint) data(chocolate) simutil<-caBTL(csimp,cpref,cprof) print("Percentage participation of profiles:", quote=FALSE) print(simutil) #Example 3 library(conjoint) data(chocolate) ShowAllSimulations(csimp,cpref,cprof) #Example 4 #library(conjoint) #data(journey) #ShowAllSimulations(jsimp,jpref,jprof)
Function caEncodedDesign encodes full or fractional factorial design. Function converts design of experiment to matrix of profiles.
caEncodedDesign(design)caEncodedDesign(design)
design |
design of experiment returned by caFactorialDesign function |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caFactorialDesign and caRecreatedDesign
#Example 1 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="orthogonal") print(design) code=caEncodedDesign(design) print(code) print(cor(code)) write.csv2(design,file="orthogonal_factorial_design.csv",row.names=FALSE) write.csv2(code,file="encoded_orthogonal_factorial_design.csv",row.names=FALSE)#Example 1 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="orthogonal") print(design) code=caEncodedDesign(design) print(code) print(cor(code)) write.csv2(design,file="orthogonal_factorial_design.csv",row.names=FALSE) write.csv2(code,file="encoded_orthogonal_factorial_design.csv",row.names=FALSE)
Function caFactorialDesign creates full or fractional factorial design. Function can return orthogonal factorial design.
caFactorialDesign(data, type="null", cards=NA, seed=123)caFactorialDesign(data, type="null", cards=NA, seed=123)
data |
experiment whose design consists of two or more factors, each with with 2 or more discrete levels |
type |
type of factorial design (possible values: "full", "fractional", "ca", "aca", "orthogonal"; default value: type="null") |
cards |
number of experimental runs |
seed |
seed settings (default value: seed=123) |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caEncodedDesign and caRecreatedDesign
#Example 1 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="full") print(design) print(cor(caEncodedDesign(design))) #Example 2 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment) print(design) print(cor(caEncodedDesign(design))) #Example 3 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="orthogonal") print(design) print(cor(caEncodedDesign(design))) #Example 4 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="fractional",cards=16) print(design) print(cor(caEncodedDesign(design))) #Example 5 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="fractional") print(design) print(cor(caEncodedDesign(design))) #Example 6 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="ca") print(design) print(cor(caEncodedDesign(design))) #Example 7 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="aca") print(design) print(cor(caEncodedDesign(design)))#Example 1 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="full") print(design) print(cor(caEncodedDesign(design))) #Example 2 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment) print(design) print(cor(caEncodedDesign(design))) #Example 3 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="orthogonal") print(design) print(cor(caEncodedDesign(design))) #Example 4 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="fractional",cards=16) print(design) print(cor(caEncodedDesign(design))) #Example 5 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="fractional") print(design) print(cor(caEncodedDesign(design))) #Example 6 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="ca") print(design) print(cor(caEncodedDesign(design))) #Example 7 library(conjoint) experiment<-expand.grid( price=c("low","medium","high"), variety=c("black","green","red"), kind=c("bags","granulated","leafy"), aroma=c("yes","no")) design=caFactorialDesign(data=experiment,type="aca") print(design) print(cor(caEncodedDesign(design)))
Function caImportance calculates importance of all attributes. Function returns vector of percentage attributes' importance and corresponding chart (barplot). The sum of importance should be 100%.
caImportance(y, x)caImportance(y, x)
y |
matrix of preferences |
x |
matrix of profiles |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
#Example 1 library(conjoint) data(tea) imp<-caImportance(tprefm,tprof) print("Importance summary: ", quote=FALSE) print(imp) print(paste("Sum: ", sum(imp)), quote=FALSE) #Example 2 library(conjoint) data(chocolate) imp<-caImportance(cprefm,cprof) print("Importance summary: ", quote=FALSE) print(imp) print(paste("Sum: ", sum(imp)), quote=FALSE) #Example 3 library(conjoint) data(journey) imp<-caImportance(jpref[1,],jprof) print("Importance summary of first respondent: ", quote=FALSE) print(imp) print(paste("Sum: ", sum(imp)), quote=FALSE) #Example 4 library(conjoint) data(journey) imp<-caImportance(jpref[1:5,],jprof) print("Importance summary of group of 5 respondents: ", quote=FALSE) print(imp) print(paste("Sum: ", sum(imp)), quote=FALSE)#Example 1 library(conjoint) data(tea) imp<-caImportance(tprefm,tprof) print("Importance summary: ", quote=FALSE) print(imp) print(paste("Sum: ", sum(imp)), quote=FALSE) #Example 2 library(conjoint) data(chocolate) imp<-caImportance(cprefm,cprof) print("Importance summary: ", quote=FALSE) print(imp) print(paste("Sum: ", sum(imp)), quote=FALSE) #Example 3 library(conjoint) data(journey) imp<-caImportance(jpref[1,],jprof) print("Importance summary of first respondent: ", quote=FALSE) print(imp) print(paste("Sum: ", sum(imp)), quote=FALSE) #Example 4 library(conjoint) data(journey) imp<-caImportance(jpref[1:5,],jprof) print("Importance summary of group of 5 respondents: ", quote=FALSE) print(imp) print(paste("Sum: ", sum(imp)), quote=FALSE)
Function caLogit estimates participation of simulation profiles using logit model. Function returns vector of percentage participations. The sum of participation should be 100%.
caLogit(sym, y, x)caLogit(sym, y, x)
sym |
matrix of simulation profiles |
y |
matrix of preferences |
x |
matrix of profiles |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caBTL, caMaxUtility and ShowAllSimulations
#Example 1 library(conjoint) data(tea) simutil<-caLogit(tsimp,tpref,tprof) print("Percentage participation of profiles: ", quote=FALSE) print(simutil) #Example 2 library(conjoint) data(chocolate) simutil<-caLogit(csimp,cpref,cprof) print("Percentage participation of profiles:", quote=FALSE) print(simutil) #Example 3 library(conjoint) data(chocolate) ShowAllSimulations(csimp,cpref,cprof) #Example 4 #library(conjoint) #data(journey) #ShowAllSimulations(jsimp,jpref,jprof)#Example 1 library(conjoint) data(tea) simutil<-caLogit(tsimp,tpref,tprof) print("Percentage participation of profiles: ", quote=FALSE) print(simutil) #Example 2 library(conjoint) data(chocolate) simutil<-caLogit(csimp,cpref,cprof) print("Percentage participation of profiles:", quote=FALSE) print(simutil) #Example 3 library(conjoint) data(chocolate) ShowAllSimulations(csimp,cpref,cprof) #Example 4 #library(conjoint) #data(journey) #ShowAllSimulations(jsimp,jpref,jprof)
Function caMaxUtility estimates participation of simulation profiles using model of maximum utility ("first position"). Function returns vector of percentage participations. The sum of participation should be 100%.
caMaxUtility(sym, y, x)caMaxUtility(sym, y, x)
sym |
matrix of simulation profiles |
y |
matrix of preferences |
x |
matrix of profiles |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caBTL, caLogit and ShowAllSimulations
#Example 1 library(conjoint) data(tea) simutil<-caMaxUtility(tsimp,tpref,tprof) print("Percentage participation of profiles: ", quote=FALSE) print(simutil) #Example 2 library(conjoint) data(chocolate) simutil<-caMaxUtility(csimp,cpref,cprof) print("Percentage participation of profiles:", quote=FALSE) print(simutil) #Example 3 library(conjoint) data(chocolate) ShowAllSimulations(csimp,cpref,cprof) #Example 4 #library(conjoint) #data(journey) #ShowAllSimulations(jsimp,jpref,jprof)#Example 1 library(conjoint) data(tea) simutil<-caMaxUtility(tsimp,tpref,tprof) print("Percentage participation of profiles: ", quote=FALSE) print(simutil) #Example 2 library(conjoint) data(chocolate) simutil<-caMaxUtility(csimp,cpref,cprof) print("Percentage participation of profiles:", quote=FALSE) print(simutil) #Example 3 library(conjoint) data(chocolate) ShowAllSimulations(csimp,cpref,cprof) #Example 4 #library(conjoint) #data(journey) #ShowAllSimulations(jsimp,jpref,jprof)
Function caModel estimates parameters of conjoint analysis model for one respondent. Function caModel returns vector of estimated parameters of traditional conjoint analysis model.
caModel(y, x)caModel(y, x)
y |
vector of preferences, vector should be like single profil of preferences |
x |
matrix of profiles |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
#Example 1 library(conjoint) data(tea) model=caModel(tprefm[1,], tprof) print(model) #Example 2 library(conjoint) data(chocolate) model=caModel(cprefm[1,], cprof) print(model) #Example 3 library(conjoint) data(journey) model=caModel(jpref[306,], jprof) print(model)#Example 1 library(conjoint) data(tea) model=caModel(tprefm[1,], tprof) print(model) #Example 2 library(conjoint) data(chocolate) model=caModel(cprefm[1,], cprof) print(model) #Example 3 library(conjoint) data(journey) model=caModel(jpref[306,], jprof) print(model)
Function caPartUtilities calculates matrix of individual utilities for respondents. Function returns matrix of partial utilities (parameters of conjoint model regresion) for all artificial variables including parameters for reference levels for respondents (with intercept on first place).
caPartUtilities(y, x, z)caPartUtilities(y, x, z)
y |
matrix of preferences |
x |
matrix of profiles |
z |
vector of levels names |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caUtilities, caTotalUtilities and ShowAllUtilities
#Example 1 library(conjoint) data(tea) uslall<-caPartUtilities(tprefm,tprof,tlevn) print(uslall) #Example 2 library(conjoint) data(chocolate) uslall<-caPartUtilities(cprefm,cprof,clevn) print(head(uslall)) #Example 3 library(conjoint) data(journey) usl<-caPartUtilities(jpref[1,],jprof,jlevn) print("Individual (partial) utilities for first respondent:") print(usl)#Example 1 library(conjoint) data(tea) uslall<-caPartUtilities(tprefm,tprof,tlevn) print(uslall) #Example 2 library(conjoint) data(chocolate) uslall<-caPartUtilities(cprefm,cprof,clevn) print(head(uslall)) #Example 3 library(conjoint) data(journey) usl<-caPartUtilities(jpref[1,],jprof,jlevn) print("Individual (partial) utilities for first respondent:") print(usl)
Function caRankToScore transforms ranking data into rating data design necessary for conjoint model.
caRankToScore(y.rank)caRankToScore(y.rank)
y.rank |
matrix of preferences in ranking format |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
#Example 1 library(conjoint) data(ice) print(ilevn) print(iprof) print(ipref) preferences<-caRankToScore(ipref) print(preferences) Conjoint(preferences, iprof, ilevn)#Example 1 library(conjoint) data(ice) print(ilevn) print(iprof) print(ipref) preferences<-caRankToScore(ipref) print(preferences) Conjoint(preferences, iprof, ilevn)
Function caRecreatedDesign reconstructs the factorial design on the basis of arguments in the form of: a vector of variables (attributes) names, a vector of the number of variables' levels, a vector of variable level names and the list of numbers of the reconstructed profiles.
caRecreatedDesign(attr.names,lev.numbers,z,prof.numbers)caRecreatedDesign(attr.names,lev.numbers,z,prof.numbers)
attr.names |
a vector of variables (attributes) names |
lev.numbers |
a vector of the number of variables' levels |
z |
a vector of variable level names |
prof.numbers |
list of numbers of the reconstructed profiles |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caFactorialDesign and caEncodedDesign
#Example 1 library(conjoint) attrNames<-c("price","variety","kind","aroma") levNumbers<-c(3,3,3,2) z<-c("low","medium","high","black","green","red","bags","granulated","leafy","yes","no") profNumbers<-c(3,4,14,20,27,29,33,35,39,43,46,50,51) design<-caRecreatedDesign(attrNames,levNumbers,z,profNumbers) print(design) write.csv2(design$dnumbers,file="design_numbers.csv",row.names=FALSE) write.csv2(design$dnames,file="design_names.csv",row.names=FALSE)#Example 1 library(conjoint) attrNames<-c("price","variety","kind","aroma") levNumbers<-c(3,3,3,2) z<-c("low","medium","high","black","green","red","bags","granulated","leafy","yes","no") profNumbers<-c(3,4,14,20,27,29,33,35,39,43,46,50,51) design<-caRecreatedDesign(attrNames,levNumbers,z,profNumbers) print(design) write.csv2(design$dnumbers,file="design_numbers.csv",row.names=FALSE) write.csv2(design$dnames,file="design_names.csv",row.names=FALSE)
Function caSegmentation divides respondents on n clusters (segments) using k-means method (function kmeans, package stats). There are two data sets used - matrix or vector of preferences and matrix of profiles.
caSegmentation(y, x, c)caSegmentation(y, x, c)
y |
matrix of preferences |
x |
matrix of profiles |
c |
number of clusters (optional), default value c=2 |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
#Example 1 library(conjoint) require(fpc) data(tea) segments<-caSegmentation(tprefm,tprof) print(segments$seg) plotcluster(segments$util,segments$sclu) #Example 2 library(conjoint) require(fpc) data(tea) segments<-caSegmentation(tpref,tprof,3) print(segments$seg) plotcluster(segments$util,segments$sclu) #example 3 library(conjoint) require(fpc) require(broom) require(ggplot2) data(tea) segments<-caSegmentation(tprefm,tprof,3) dcf<-discrcoord(segments$util,segments$sclu) assignments<-augment(segments$segm,dcf$proj[,1:2]) ggplot(assignments)+geom_point(aes(x=X1,y=X2,color= .cluster))+labs(color="Cluster Assignment", title="K-Means Clustering Results") #Example 4 library(conjoint) require(ggfortify) data(tea) segments<-caSegmentation(tpref,tprof,3) print(segments$seg) util<-as.data.frame(segments$util) set.seed(123) ggplot2::autoplot(kmeans(util,3),data=util,label=TRUE,label.size=4,frame=TRUE) #Example 5 #library(conjoint) #require(ggfortify) #require(cluster) #data(tea) #segments<-caSegmentation(tpref,tprof,3) #print(segments$seg) #util<-as.data.frame(segments$util) #ggplot2::autoplot(pam(util,3),label=TRUE,label.size=4,frame=TRUE,frame.type='norm')#Example 1 library(conjoint) require(fpc) data(tea) segments<-caSegmentation(tprefm,tprof) print(segments$seg) plotcluster(segments$util,segments$sclu) #Example 2 library(conjoint) require(fpc) data(tea) segments<-caSegmentation(tpref,tprof,3) print(segments$seg) plotcluster(segments$util,segments$sclu) #example 3 library(conjoint) require(fpc) require(broom) require(ggplot2) data(tea) segments<-caSegmentation(tprefm,tprof,3) dcf<-discrcoord(segments$util,segments$sclu) assignments<-augment(segments$segm,dcf$proj[,1:2]) ggplot(assignments)+geom_point(aes(x=X1,y=X2,color= .cluster))+labs(color="Cluster Assignment", title="K-Means Clustering Results") #Example 4 library(conjoint) require(ggfortify) data(tea) segments<-caSegmentation(tpref,tprof,3) print(segments$seg) util<-as.data.frame(segments$util) set.seed(123) ggplot2::autoplot(kmeans(util,3),data=util,label=TRUE,label.size=4,frame=TRUE) #Example 5 #library(conjoint) #require(ggfortify) #require(cluster) #data(tea) #segments<-caSegmentation(tpref,tprof,3) #print(segments$seg) #util<-as.data.frame(segments$util) #ggplot2::autoplot(pam(util,3),label=TRUE,label.size=4,frame=TRUE,frame.type='norm')
Function caTotalUtilities calculates matrix of theoreticall total utilities for respondents. Function returns matrix of total utilities for all profiles.
caTotalUtilities(y, x)caTotalUtilities(y, x)
y |
matrix of preferences |
x |
matrix of profiles |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caUtilities, caPartUtilities and ShowAllUtilities
#Example 1 library(conjoint) data(tea) uslall<-caTotalUtilities(tprefm,tprof) print(uslall) #Example 2 library(conjoint) data(chocolate) uslall<-caTotalUtilities(cprefm,cprof) print(uslall) #Example 3 library(conjoint) data(journey) usl<-caTotalUtilities(jpref[1,],jprof) print("Individual (total) utilities for first respondent:") print(usl)#Example 1 library(conjoint) data(tea) uslall<-caTotalUtilities(tprefm,tprof) print(uslall) #Example 2 library(conjoint) data(chocolate) uslall<-caTotalUtilities(cprefm,cprof) print(uslall) #Example 3 library(conjoint) data(journey) usl<-caTotalUtilities(jpref[1,],jprof) print("Individual (total) utilities for first respondent:") print(usl)
Function caUtilities calculates utilities of attribute's levels. Function returns vector of utilities.
caUtilities(y,x,z)caUtilities(y,x,z)
y |
matrix of preferences |
x |
matrix of profiles |
z |
matrix of levels names |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caPartUtilities and caTotalUtilities
#Example 1 library(conjoint) data(tea) uslall<-caUtilities(tprefm,tprof,tlevn) print(uslall) #Example 2 library(conjoint) data(chocolate) uslall<-caUtilities(cprefm,cprof,clevn) print(uslall) #Example 3 library(conjoint) data(journey) usl<-caUtilities(jpref[1,],jprof,jlevn) print("Individual utilities for first respondent:") print(usl)#Example 1 library(conjoint) data(tea) uslall<-caUtilities(tprefm,tprof,tlevn) print(uslall) #Example 2 library(conjoint) data(chocolate) uslall<-caUtilities(cprefm,cprof,clevn) print(uslall) #Example 3 library(conjoint) data(journey) usl<-caUtilities(jpref[1,],jprof,jlevn) print("Individual utilities for first respondent:") print(usl)
Sample data in score mode. Rating (score) data does not need any conversion. Data collected in the survey conducted by W. Nowak in 2000.
data(chocolate) cpref cprefm cprof clevn csimpdata(chocolate) cpref cprefm cprof clevn csimp
cprefVector of preferences (length 1392).
cprefmMatrix of preferences (87 respondents and 16 profiles).
cprofMatrix of profiles (5 attributes and 16 profiles).
clevnCharacter vector of names for the attributes' levels.
csimpMatrix of simulation profiles.
library(conjoint) data(chocolate) print(cprefm) print(cprof) print(clevn) print(csimp)library(conjoint) data(chocolate) print(cprefm) print(cprof) print(clevn) print(csimp)
Function Conjoint is a combination of following conjoint pakage's functions: caPartUtilities, caUtilities and caImportance. Therefore it sums up the main results of conjoint analysis. Function Conjoint returns matrix of partial utilities for levels of variables for respondents, vector of utilities for attribute's levels and vector of percentage attributes' importance with corresponding chart (barplot). The sum of importance should be 100
Conjoint(y, x, z, y.type)Conjoint(y, x, z, y.type)
y |
matrix of preferences |
x |
matrix of profiles |
z |
matrix of levels names |
y.type |
type of data preferences (possible values: "score" for preferences as rating data, "rank" for preferences as ranking data; default value: y.type="score") |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caImportance, caPartUtilities and caUtilities
#Example 1 library(conjoint) data(ice) print("Preferences of all respondents (preferences as ranking data):") Conjoint(ipref,iprof,ilevn,y.type="rank") #Example 2 library(conjoint) data(ice) ipref=caRankToScore(ipref) print("Preferences of all respondents (preferences converted into rating data):") Conjoint(ipref,iprof,ilevn,y.type="score") #Example 3 library(conjoint) data(journey) print("Preferences of all respondents (preferences as default - rating data):") Conjoint(jpref,jprof,jlevn) #Example 4 library(conjoint) data(tea) print("Preferences of all respondents (preferences as rating data):") Conjoint(tprefm,tprof,tlevn,y.type="score") #Example 5 library(conjoint) data(tea) print("Preferences of first respondent (preferences as default - rating data):") Conjoint(tprefm[1,],tprof,tlevn) #Example 6 library(conjoint) data(tea) print("Preferences of group of 5 respondents (preferences as rating data):") Conjoint(tprefm[11:15,],tprof,tlevn,y.type="score")#Example 1 library(conjoint) data(ice) print("Preferences of all respondents (preferences as ranking data):") Conjoint(ipref,iprof,ilevn,y.type="rank") #Example 2 library(conjoint) data(ice) ipref=caRankToScore(ipref) print("Preferences of all respondents (preferences converted into rating data):") Conjoint(ipref,iprof,ilevn,y.type="score") #Example 3 library(conjoint) data(journey) print("Preferences of all respondents (preferences as default - rating data):") Conjoint(jpref,jprof,jlevn) #Example 4 library(conjoint) data(tea) print("Preferences of all respondents (preferences as rating data):") Conjoint(tprefm,tprof,tlevn,y.type="score") #Example 5 library(conjoint) data(tea) print("Preferences of first respondent (preferences as default - rating data):") Conjoint(tprefm[1,],tprof,tlevn) #Example 6 library(conjoint) data(tea) print("Preferences of group of 5 respondents (preferences as rating data):") Conjoint(tprefm[11:15,],tprof,tlevn,y.type="score")
Sample data in score mode. Rating (score) data does not need any conversion. Data collected in the survey conducted by W. Nowak in 2000.
data(czekolada) czpref czprefm czprof czlevn czsimpdata(czekolada) czpref czprefm czprof czlevn czsimp
czprefVector of preferences (length 1392).
czprefmMatrix of preferences (87 respondents and 16 profiles).
czprofMatrix of profiles (5 attributes and 16 profiles).
czlevnCharacter vector of names for the attributes' levels.
czsimpMatrix of simulation profiles.
library(conjoint) data(czekolada) print(czprefm) print(czprof) print(czlevn) print(czsimp)library(conjoint) data(czekolada) print(czprefm) print(czprof) print(czlevn) print(czsimp)
Sample data in score mode. Rating (score) data does not need any conversion. Data collected in the survey conducted by M. Baran in 2007.
data(herbata) hpref hprefm hprof hlevn hsimpdata(herbata) hpref hprefm hprof hlevn hsimp
hprefVector of preferences (length 1300).
hprefmMatrix of preferences (100 respondents and 13 profiles).
hprofMatrix of profiles (4 attributes and 13 profiles).
hlevnCharacter vector of names for the attributes' levels.
hsimpMatrix of simulation profiles.
library(conjoint) data(herbata) print(hprefm) print(hprof) print(hlevn) print(hsimp)library(conjoint) data(herbata) print(hprefm) print(hprof) print(hlevn) print(hsimp)
Sample artificial data in rank mode. Ranking (rank) data needs conversion into rating (score) data.
data(ice) ipref iprof ilevndata(ice) ipref iprof ilevn
iprefMatrix of preferences (6 respondents and 9 profiles).
iprofMatrix of profiles (4 attributes and 9 profiles).
ilevnCharacter vector of names for the attributes' levels.
library(conjoint) data(ice) print(iprof) print(ipref) print(ilevn)library(conjoint) data(ice) print(iprof) print(ipref) print(ilevn)
Sample data in score mode. Rating (score) data does not need any conversion. Data collected in the survey conducted by M. Gordzicz in 2015/2016.
data(journey) jpref jprof jlevn jsimpdata(journey) jpref jprof jlevn jsimp
jprefMatrix of preferences (306 respondents and 14 profiles).
jprofMatrix of profiles (4 attributes and 14 profiles).
jlevnCharacter vector of names for the attributes' levels.
jsimpMatrix of simulation profiles.
library(conjoint) data(journey) print(jpref) print(jprof) print(jlevn) print(jsimp)library(conjoint) data(journey) print(jpref) print(jprof) print(jlevn) print(jsimp)
Sample artificial data in rank mode. Ranking (rank) data needs conversion into rating (score) data.
data(lody) lpref lprof llevndata(lody) lpref lprof llevn
lprefMatrix of preferences (6 respondents and 9 profiles).
lprofMatrix of profiles (4 attributes and 9 profiles).
llevnCharacter vector of names for the attributes' levels.
library(conjoint) data(lody) print(lprof) print(lpref) print(llevn)library(conjoint) data(lody) print(lprof) print(lpref) print(llevn)
Sample artificial data in score mode. Rating (score) data does not need any conversion.
data(plyty) ppref pprof plevndata(plyty) ppref pprof plevn
pprefMatrix of preferences (6 respondents and 8 profiles).
pprofMatrix of profiles (3 attributes and 8 profiles).
plevnCharacter vector of names for the attributes' levels.
library(conjoint) data(plyty) print(pprof) print(ppref) print(plevn)library(conjoint) data(plyty) print(pprof) print(ppref) print(plevn)
Function ShowAllSimulations sums up the results of all simulation functions. It's a combination of following conjoint pakage's functions: caMaxUtility, caBTL and caLogit. Therefore it sums up the main results of simulation using conjoint analysis method. Function ShowAllSimulations returns three vectors of percentage participations using maximum utility, BTL and logit models. The sum of importance for every vector should be 100%.
ShowAllSimulations(sym, y, x)ShowAllSimulations(sym, y, x)
sym |
matrix of simulation profiles |
y |
matrix of preferences |
x |
matrix of profiles |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A., Bartlomowicz T. (2012), Conjoint analysis method and its implementation in conjoint R package, [In:] Pociecha J., Decker R. (Eds.), Data analysis methods and its applications, C.H.Beck, Warszawa, p.239-248.
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa, p. 283-317.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, p. 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caBTL, caLogit and caMaxUtility
#Example 1 library(conjoint) data(tea) ShowAllSimulations(tsimp,tpref,tprof) #Example 2 library(conjoint) data(chocolate) ShowAllSimulations(csimp,cpref,cprof) #Example 3 #library(conjoint) #data(journey) #ShowAllSimulations(jsimp,jpref,jprof)#Example 1 library(conjoint) data(tea) ShowAllSimulations(tsimp,tpref,tprof) #Example 2 library(conjoint) data(chocolate) ShowAllSimulations(csimp,cpref,cprof) #Example 3 #library(conjoint) #data(journey) #ShowAllSimulations(jsimp,jpref,jprof)
Function ShowAllUtilities is a combination of following conjoint pakage's functions: caPartUtilities, caTotalUtilities, caUtilities and caImportance. Function ShowAllUtilities returns: matrix of partial utilities (basic matrix of utilities with the intercept), matrix of total utilities for n profiles and all respondents, vector of utilities for attribute's levels and vector of percentage attributes' importance, with sum of importance. The sum of importance should be 100%.
ShowAllUtilities(y, x, z)ShowAllUtilities(y, x, z)
y |
matrix of preferences |
x |
matrix of profiles |
z |
matrix of levles names |
Andrzej Bak [email protected],
Tomasz Bartlomowicz [email protected]
Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/conjoint
Bak A. (2009), Analiza Conjoint [Conjoint Analysis], [In:] Walesiak M., Gatnar E. (Eds.), Statystyczna analiza danych z wykorzystaniem programu R [Statistical Data Analysis using R], Wydawnictwo Naukowe PWN, Warszawa.
Green P.E., Srinivasan V. (1978), Conjoint Analysis in Consumer Research: Issues and Outlook, "Journal of Consumer Research", September, 5, 103-123.
SPSS 6.1 Categories (1994), SPSS Inc., Chicago.
caImportance, caPartUtilities, caTotalUtilities and caUtilities
#Example 1 library(conjoint) data(tea) ShowAllUtilities(tprefm,tprof,tlevn) #Example 2 library(conjoint) data(chocolate) ShowAllUtilities(cprefm,cprof,clevn)#Example 1 library(conjoint) data(tea) ShowAllUtilities(tprefm,tprof,tlevn) #Example 2 library(conjoint) data(chocolate) ShowAllUtilities(cprefm,cprof,clevn)
Sample data in score mode. Rating (score) data does not need any conversion. Data collected in the survey conducted by M. Baran in 2007.
data(tea) tpref tprefm tprof tlevn tsimpdata(tea) tpref tprefm tprof tlevn tsimp
tprefVector of preferences (length 1300).
tprefmMatrix of preferences (100 respondents and 13 profiles).
tprofMatrix of profiles (4 attributes and 13 profiles).
tlevnCharacter vector of names for the attributes' levels.
tsimpMatrix of simulation profiles.
library(conjoint) data(tea) print(tprefm) print(tprof) print(tlevn) print(tsimp)library(conjoint) data(tea) print(tprefm) print(tprof) print(tlevn) print(tsimp)
Sample data in score mode. Rating (score) data does not need any conversion. Data collected in the survey conducted by M. Gordzicz in 2015/2016.
data(wycieczka) wpref wprof wlevn wsimpdata(wycieczka) wpref wprof wlevn wsimp
wprefMatrix of preferences (306 respondents and 14 profiles).
wprofMatrix of profiles (4 attributes and 14 profiles).
wlevnCharacter vector of names for the attributes' levels.
wsimpMatrix of simulation profiles.
library(conjoint) data(wycieczka) print(wpref) print(wprof) print(wlevn) print(wsimp)library(conjoint) data(wycieczka) print(wpref) print(wprof) print(wlevn) print(wsimp)