#************************************************************************************
# Thesis title: "Napping: from sensory profiling technique to projective technique"
# Appendix 2: Bananas - Author: Tam Minh LE
# Supervisor : Francois Husson & Sebastien Le
# R version: 3.2.0 - Updated: 19.05.2015
#************************************************************************************
rm(list=ls())
options(max.print=5.5E5) # max print
# Require package ---------------------------------------------------------
library(FactoMineR)
library(SensoMineR)
library(ggplot2)
library(scales)
library(reshape2)
library(holos)
library(knitr)
library(conjoint) # traditional conjoint analysis with multiple linear reg.
library(AlgDesign) # fractional factorial design
library(clusterSim) # segmentation
## Loading required package: cluster
## Loading required package: MASS
library(agricolae) # design Graeco design
# Import sorting data ----------------------------------------------------
wd <- "C:/Users/LE Minh-Tam/Dropbox/00. Thesis/Data"
setwd(wd)
list.files(pattern=".rda")
## [1] "ba.qual.rda" "ba.sort.rda" "cards.rda" "emotion_fr.rda"
## [5] "emotion_vn.rda" "videotable.rda"
load("ba.sort.rda")
dim(ba.sort)
## [1] 9 96
# summary(ba.sort) # 32 subjects
# Example for Introduction ------------------------------------------------
exp.design <- matrix(c(1, 1, 1, 1,
1, 2, 3, 2,
1, 3, 2, 3,
2, 1, 3, 3,
2, 2, 2, 1,
2, 3, 1, 2,
3, 1, 2, 2,
3, 2, 1, 3,
3, 3, 3, 1), ncol=4, nrow=9, byrow=TRUE)
rownames(exp.design) <- paste("profile.", 1:9, sep="")
colnames(exp.design) <- c("color", "shape", "bruise", "cigar")
exp.design <- as.data.frame(exp.design)
rownames(exp.design) <- rownames(ba.sort)
res.pca <- PCA(exp.design)
barplot(res.pca$eig[,2],main="Eigenvalues",names.arg=1:nrow(res.pca$eig),
ylim=c(0,40), xlab="dimensions", ylab="Percentage variance")
# Factorial approach for sorting Task -------------------------------------
dat.sort <- ba.sort[,seq(3, ncol(ba.sort), by=3)]
res.fast <- fast(dat.sort, sep.words=";")
## Number of different words : 137
## using MCA
res.mca <- MCA(dat.sort)
barplot(res.mca$eig[,2], main="Eigenvalues", names.arg=1:nrow(res.mca$eig),
ylim=c(0,40), xlab="dimensions", ylab="Percentage variance")
# Conjoint analysis -------------------------------------------------------
## Create a data frame: levels of descriptors
color <- c("green", "green-yellow", "yellow")
shape <- c("high.heterogeneity", "heterogeneity", "homogeneity")
bruise <- c("high.bruise", "bruise", "non.bruise")
cigar.end.rot <- c("high.cigar", "med.cigar", "non.cigar")
lev.names <- data.frame(c(color, shape, bruise, cigar.end.rot))
colnames(lev.names) <- "levels"
lev.names
## levels
## 1 green
## 2 green-yellow
## 3 yellow
## 4 high.heterogeneity
## 5 heterogeneity
## 6 homogeneity
## 7 high.bruise
## 8 bruise
## 9 non.bruise
## 10 high.cigar
## 11 med.cigar
## 12 non.cigar
## Create a data frame: quality score
load("ba.qual.rda")
dim(ba.qual)
## [1] 9 96
#summary(ba.qual)
quality <- ba.qual[,seq(1, ncol(ba.qual), by=3)]
quality$id <- rownames(quality)
qual <- as.data.frame(melt(quality), quote=FALSE) # long.table
## Using id as id variables
qual.score <- qual[,3]
# Conjoint analysis -------------------------------------------------------
Conjoint(qual.score, exp.design, lev.names)
##
## Call:
## lm(formula = frml)
##
## Residuals:
## Min 1Q Median 3Q Max
## -723,31 -131,70 -2,73 127,70 657,75
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 451,594 12,267 36,813 < 2e-16 ***
## factor(x$color)1 -43,500 17,349 -2,507 0,0127 *
## factor(x$color)2 16,688 17,349 0,962 0,3369
## factor(x$shape)1 -2,656 17,349 -0,153 0,8784
## factor(x$shape)2 -106,385 17,349 -6,132 2,95e-09 ***
## factor(x$bruise)1 -173,125 17,349 -9,979 < 2e-16 ***
## factor(x$bruise)2 -71,240 17,349 -4,106 5,29e-05 ***
## factor(x$cigar)1 -16,500 17,349 -0,951 0,3424
## factor(x$cigar)2 28,146 17,349 1,622 0,1059
## ---
## Signif. codes: 0 '***' 0,001 '**' 0,01 '*' 0,05 '.' 0,1 ' ' 1
##
## Residual standard error: 208,2 on 279 degrees of freedom
## Multiple R-squared: 0,4922, Adjusted R-squared: 0,4776
## F-statistic: 33,8 on 8 and 279 DF, p-value: < 2,2e-16
## [1] "Part worths (utilities) of levels (model parameters for whole sample):"
## levnms utls
## 1 intercept 451,5938
## 2 green -43,5
## 3 green-yellow 16,6875
## 4 yellow 26,8125
## 5 high.heterogeneity -2,6562
## 6 heterogeneity -106,3854
## 7 homogeneity 109,0417
## 8 high.bruise -173,125
## 9 bruise -71,2396
## 10 non.bruise 244,3646
## 11 high.cigar -16,5
## 12 med.cigar 28,1458
## 13 non.cigar -11,6458
## [1] "Average importance of factors (attributes):"
## [1] 17,68 23,45 45,62 13,25
## [1] Sum of average importance: 100
## [1] "Chart of average factors importance"
caSegmentation(qual.score, exp.design, 2)
## K-means clustering with 2 clusters of sizes 3, 29
##
## Cluster means:
## [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
## 1 531.0000 65.0000 567.3333 258.3333 314.0000 524.3333 547.3333 738.3333
## 2 183.2069 626.8966 420.4828 743.8621 270.0345 422.8276 420.7931 130.2414
## [,9]
## 1 689.0000
## 2 828.3793
##
## Clustering vector:
## [1] 2 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2
##
## Within cluster sum of squares by cluster:
## [1] 1059861 8010402
## (between_SS / total_SS = 25.0 %)
##
## Available components:
##
## [1] "cluster" "centers" "totss" "withinss"
## [5] "tot.withinss" "betweenss" "size" "iter"
## [9] "ifault"
#X11(); par(mfrow = c(2,2))
ShowAllUtilities(qual.score, exp.design, lev.names) # estimating all models
##
## Call:
## lm(formula = frml)
##
## Residuals:
## Min 1Q Median 3Q Max
## -723,31 -131,70 -2,73 127,70 657,75
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 451,594 12,267 36,813 < 2e-16 ***
## factor(x$color)1 -43,500 17,349 -2,507 0,0127 *
## factor(x$color)2 16,688 17,349 0,962 0,3369
## factor(x$shape)1 -2,656 17,349 -0,153 0,8784
## factor(x$shape)2 -106,385 17,349 -6,132 2,95e-09 ***
## factor(x$bruise)1 -173,125 17,349 -9,979 < 2e-16 ***
## factor(x$bruise)2 -71,240 17,349 -4,106 5,29e-05 ***
## factor(x$cigar)1 -16,500 17,349 -0,951 0,3424
## factor(x$cigar)2 28,146 17,349 1,622 0,1059
## ---
## Signif. codes: 0 '***' 0,001 '**' 0,01 '*' 0,05 '.' 0,1 ' ' 1
##
## Residual standard error: 208,2 on 279 degrees of freedom
## Multiple R-squared: 0,4922, Adjusted R-squared: 0,4776
## F-statistic: 33,8 on 8 and 279 DF, p-value: < 2,2e-16
## [1] "Part worths (utilities) of levels (model parameters for whole sample):"
## levnms utls
## 1 intercept 451,5938
## 2 green -43,5
## 3 green-yellow 16,6875
## 4 yellow 26,8125
## 5 high.heterogeneity -2,6562
## 6 heterogeneity -106,3854
## 7 homogeneity 109,0417
## 8 high.bruise -173,125
## 9 bruise -71,2396
## 10 non.bruise 244,3646
## 11 high.cigar -16,5
## 12 med.cigar 28,1458
## 13 non.cigar -11,6458
## [1] "Part worths (utilities) of all levels including intercept and reference levels a cross of respondents:"
## intercept green green-yellow yellow high.heterogeneity
## [1,] 464,000 37,000 31,333 -68,333 7,667
## [2,] 559,444 -106,778 0,889 105,889 -7,444
## [3,] 476,333 79,000 -26,667 -52,333 23,667
## [4,] 551,111 29,556 62,222 -91,778 26,222
## [5,] 396,778 92,222 -29,111 -63,111 69,222
## [6,] 396,889 -1,222 30,778 -29,556 -113,556
## [7,] 385,000 -345,000 205,333 139,667 107,333
## [8,] 393,667 22,333 172,667 -195,000 24,667
## [9,] 468,222 -97,556 5,444 92,111 -8,222
## [10,] 372,444 -30,778 -62,444 93,222 31,889
## [11,] 474,444 -37,111 34,889 2,222 -62,111
## [12,] 457,889 -192,889 153,444 39,444 -197,556
## [13,] 461,444 -36,778 48,889 -12,111 64,556
## [14,] 517,889 33,444 32,444 -65,889 27,778
## [15,] 434,222 -101,222 64,444 36,778 30,778
## [16,] 504,667 -39,000 75,333 -36,333 83,000
## [17,] 521,778 -108,111 -217,444 325,556 3,556
## [18,] 517,333 -109,333 -35,000 144,333 -110,333
## [19,] 411,222 135,444 -27,222 -108,222 -24,556
## [20,] 530,222 189,444 -134,556 -54,889 -150,556
## [21,] 378,000 -34,333 -10,333 44,667 2,000
## [22,] 357,778 25,556 5,889 -31,444 1,222
## [23,] 497,889 -167,556 136,778 30,778 -39,222
## [24,] 422,556 -16,556 -5,889 22,444 22,444
## [25,] 563,889 -35,556 4,111 31,444 29,111
## [26,] 430,000 -73,000 1,667 71,333 77,000
## [27,] 459,444 78,889 -69,444 -9,444 81,556
## [28,] 329,889 3,111 25,444 -28,556 4,444
## [29,] 452,556 -41,889 31,444 10,444 -45,556
## [30,] 369,667 -155,000 -2,667 157,667 -30,333
## [31,] 567,667 -162,667 8,333 154,333 -19,667
## [32,] 326,667 -225,667 23,000 202,667 6,000
## heterogeneity homogeneity high.bruise bruise non.bruise high.cigar
## [1,] -199,667 192,000 -154,333 -186,333 340,667 11,667
## [2,] -94,778 102,222 -208,111 -92,444 300,556 -128,111
## [3,] -130,333 106,667 -142,667 -106,000 248,667 28,667
## [4,] -125,111 98,889 -153,444 -124,778 278,222 -96,444
## [5,] -164,444 95,222 -131,444 -161,444 292,889 63,222
## [6,] -279,556 393,111 -74,889 -117,889 192,778 -30,222
## [7,] -189,333 82,000 -231,333 -18,000 249,333 84,000
## [8,] -37,000 12,333 -110,333 -101,000 211,333 -330,333
## [9,] -23,889 32,111 -327,889 -0,222 328,111 19,444
## [10,] 58,889 -90,778 179,222 100,889 -280,111 92,222
## [11,] -111,111 173,222 -248,444 -81,444 329,889 -67,778
## [12,] 38,444 159,111 -72,556 -198,889 271,444 5,111
## [13,] -107,778 43,222 -339,444 -26,778 366,222 -29,778
## [14,] -166,556 138,778 -106,889 -206,556 313,444 35,778
## [15,] -66,222 35,444 -257,889 -33,556 291,444 39,111
## [16,] -49,667 -33,333 -355,000 14,333 340,667 -23,667
## [17,] -136,111 132,556 130,556 26,556 -157,111 117,222
## [18,] -217,000 327,333 72,333 -110,333 38,000 -87,000
## [19,] -177,889 202,444 -249,222 -180,222 429,444 -25,889
## [20,] -52,889 203,444 -256,556 182,778 73,778 187,444
## [21,] -93,000 91,000 -233,000 -79,000 312,000 -24,667
## [22,] -151,444 150,222 -173,444 -85,111 258,556 -36,111
## [23,] -127,556 166,778 -176,222 -119,222 295,444 -50,889
## [24,] -97,222 74,778 -371,556 -105,556 477,111 -56,889
## [25,] -125,556 96,444 -249,222 -29,222 278,444 -31,222
## [26,] -162,667 85,667 -205,000 -34,000 239,000 -12,000
## [27,] -187,778 106,222 -252,778 -72,111 324,889 -1,111
## [28,] -140,889 136,444 -132,556 -99,556 232,111 -22,889
## [29,] -232,556 278,111 -163,889 -48,222 212,111 -109,222
## [30,] 62,333 -32,000 -149,000 -124,000 273,000 -35,333
## [31,] 70,333 -50,667 -283,333 -2,333 285,667 -17,000
## [32,] 13,667 -19,667 -111,667 -60,000 171,667 4,667
## med.cigar non.cigar
## [1,] 30,000 -41,667
## [2,] 119,556 8,556
## [3,] -28,667 0,000
## [4,] 99,222 -2,778
## [5,] -25,444 -37,778
## [6,] 43,111 -12,889
## [7,] 20,000 -104,000
## [8,] 192,667 137,667
## [9,] 8,111 -27,556
## [10,] -120,444 28,222
## [11,] 31,889 35,889
## [12,] 117,111 -122,222
## [13,] 31,889 -2,111
## [14,] 38,778 -74,556
## [15,] 57,778 -96,889
## [16,] 87,333 -63,667
## [17,] -120,444 3,222
## [18,] -34,000 121,000
## [19,] -98,222 124,111
## [20,] -21,556 -165,889
## [21,] 121,333 -96,667
## [22,] 37,556 -1,444
## [23,] 94,778 -43,889
## [24,] 33,111 23,778
## [25,] 63,778 -32,556
## [26,] 49,667 -37,667
## [27,] 10,556 -9,444
## [28,] 50,778 -27,889
## [29,] 83,778 25,444
## [30,] 4,000 31,333
## [31,] -7,000 24,000
## [32,] -70,333 65,667
## [1] "Total utilities of profiles a cross of respondents:"
## [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9]
## [1,] 366 672 465 802 121 563 247 0 940
## [2,] 109 778 471 862 245 574 685 371 940
## [3,] 465 645 556 722 242 385 313 151 808
## [4,] 357 833 552 915 267 658 460 178 740
## [5,] 490 592 385 692 105 306 216 0 785
## [6,] 177 352 658 494 0 789 179 0 923
## [7,] 0 120 0 843 467 461 634 0 940
## [8,] 0 783 465 940 98 661 315 189 92
## [9,] 54 683 375 766 469 186 560 181 940
## [10,] 645 0 380 90 562 278 478 732 187
## [11,] 59 688 565 813 249 466 365 153 912
## [12,] 0 692 103 563 456 815 218 341 933
## [13,] 120 715 439 939 346 246 519 0 829
## [14,] 508 737 409 817 213 621 312 104 940
## [15,] 145 616 238 724 438 334 526 50 837
## [16,] 170 844 383 940 521 279 653 0 752
## [17,] 665 0 576 154 312 447 757 845 940
## [18,] 283 195 746 531 68 848 407 638 940
## [19,] 247 700 693 913 0 239 0 0 909
## [20,] 500 719 940 153 713 321 486 0 940
## [21,] 88 684 259 585 171 347 467 0 801
## [22,] 175 528 447 622 91 378 280 0 699
## [23,] 64 593 334 847 337 720 465 181 940
## [24,] 0 819 399 940 157 153 395 0 940
## [25,] 277 745 563 843 382 479 659 188 939
## [26,] 217 483 371 710 223 362 594 96 814
## [27,] 366 686 563 787 129 254 470 0 880
## [28,] 182 475 342 564 92 410 257 0 647
## [29,] 92 474 666 676 94 682 453 92 844
## [30,] 0 554 90 641 270 190 377 472 733
## [31,] 85 754 376 866 627 235 693 533 940
## [32,] 0 216 87 593 308 148 405 497 686
## [1] "Average importance of factors (attributes):"
## [1] 17,68 23,45 45,62 13,25
## [1] Sum of average importance: 100
#X11(); par(mfrow = c(2,2))
utility <- ShowAllUtilities(qual.score, exp.design, lev.names)
##
## Call:
## lm(formula = frml)
##
## Residuals:
## Min 1Q Median 3Q Max
## -723,31 -131,70 -2,73 127,70 657,75
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 451,594 12,267 36,813 < 2e-16 ***
## factor(x$color)1 -43,500 17,349 -2,507 0,0127 *
## factor(x$color)2 16,688 17,349 0,962 0,3369
## factor(x$shape)1 -2,656 17,349 -0,153 0,8784
## factor(x$shape)2 -106,385 17,349 -6,132 2,95e-09 ***
## factor(x$bruise)1 -173,125 17,349 -9,979 < 2e-16 ***
## factor(x$bruise)2 -71,240 17,349 -4,106 5,29e-05 ***
## factor(x$cigar)1 -16,500 17,349 -0,951 0,3424
## factor(x$cigar)2 28,146 17,349 1,622 0,1059
## ---
## Signif. codes: 0 '***' 0,001 '**' 0,01 '*' 0,05 '.' 0,1 ' ' 1
##
## Residual standard error: 208,2 on 279 degrees of freedom
## Multiple R-squared: 0,4922, Adjusted R-squared: 0,4776
## F-statistic: 33,8 on 8 and 279 DF, p-value: < 2,2e-16
## [1] "Part worths (utilities) of levels (model parameters for whole sample):"
## levnms utls
## 1 intercept 451,5938
## 2 green -43,5
## 3 green-yellow 16,6875
## 4 yellow 26,8125
## 5 high.heterogeneity -2,6562
## 6 heterogeneity -106,3854
## 7 homogeneity 109,0417
## 8 high.bruise -173,125
## 9 bruise -71,2396
## 10 non.bruise 244,3646
## 11 high.cigar -16,5
## 12 med.cigar 28,1458
## 13 non.cigar -11,6458
## [1] "Part worths (utilities) of all levels including intercept and reference levels a cross of respondents:"
## intercept green green-yellow yellow high.heterogeneity
## [1,] 464,000 37,000 31,333 -68,333 7,667
## [2,] 559,444 -106,778 0,889 105,889 -7,444
## [3,] 476,333 79,000 -26,667 -52,333 23,667
## [4,] 551,111 29,556 62,222 -91,778 26,222
## [5,] 396,778 92,222 -29,111 -63,111 69,222
## [6,] 396,889 -1,222 30,778 -29,556 -113,556
## [7,] 385,000 -345,000 205,333 139,667 107,333
## [8,] 393,667 22,333 172,667 -195,000 24,667
## [9,] 468,222 -97,556 5,444 92,111 -8,222
## [10,] 372,444 -30,778 -62,444 93,222 31,889
## [11,] 474,444 -37,111 34,889 2,222 -62,111
## [12,] 457,889 -192,889 153,444 39,444 -197,556
## [13,] 461,444 -36,778 48,889 -12,111 64,556
## [14,] 517,889 33,444 32,444 -65,889 27,778
## [15,] 434,222 -101,222 64,444 36,778 30,778
## [16,] 504,667 -39,000 75,333 -36,333 83,000
## [17,] 521,778 -108,111 -217,444 325,556 3,556
## [18,] 517,333 -109,333 -35,000 144,333 -110,333
## [19,] 411,222 135,444 -27,222 -108,222 -24,556
## [20,] 530,222 189,444 -134,556 -54,889 -150,556
## [21,] 378,000 -34,333 -10,333 44,667 2,000
## [22,] 357,778 25,556 5,889 -31,444 1,222
## [23,] 497,889 -167,556 136,778 30,778 -39,222
## [24,] 422,556 -16,556 -5,889 22,444 22,444
## [25,] 563,889 -35,556 4,111 31,444 29,111
## [26,] 430,000 -73,000 1,667 71,333 77,000
## [27,] 459,444 78,889 -69,444 -9,444 81,556
## [28,] 329,889 3,111 25,444 -28,556 4,444
## [29,] 452,556 -41,889 31,444 10,444 -45,556
## [30,] 369,667 -155,000 -2,667 157,667 -30,333
## [31,] 567,667 -162,667 8,333 154,333 -19,667
## [32,] 326,667 -225,667 23,000 202,667 6,000
## heterogeneity homogeneity high.bruise bruise non.bruise high.cigar
## [1,] -199,667 192,000 -154,333 -186,333 340,667 11,667
## [2,] -94,778 102,222 -208,111 -92,444 300,556 -128,111
## [3,] -130,333 106,667 -142,667 -106,000 248,667 28,667
## [4,] -125,111 98,889 -153,444 -124,778 278,222 -96,444
## [5,] -164,444 95,222 -131,444 -161,444 292,889 63,222
## [6,] -279,556 393,111 -74,889 -117,889 192,778 -30,222
## [7,] -189,333 82,000 -231,333 -18,000 249,333 84,000
## [8,] -37,000 12,333 -110,333 -101,000 211,333 -330,333
## [9,] -23,889 32,111 -327,889 -0,222 328,111 19,444
## [10,] 58,889 -90,778 179,222 100,889 -280,111 92,222
## [11,] -111,111 173,222 -248,444 -81,444 329,889 -67,778
## [12,] 38,444 159,111 -72,556 -198,889 271,444 5,111
## [13,] -107,778 43,222 -339,444 -26,778 366,222 -29,778
## [14,] -166,556 138,778 -106,889 -206,556 313,444 35,778
## [15,] -66,222 35,444 -257,889 -33,556 291,444 39,111
## [16,] -49,667 -33,333 -355,000 14,333 340,667 -23,667
## [17,] -136,111 132,556 130,556 26,556 -157,111 117,222
## [18,] -217,000 327,333 72,333 -110,333 38,000 -87,000
## [19,] -177,889 202,444 -249,222 -180,222 429,444 -25,889
## [20,] -52,889 203,444 -256,556 182,778 73,778 187,444
## [21,] -93,000 91,000 -233,000 -79,000 312,000 -24,667
## [22,] -151,444 150,222 -173,444 -85,111 258,556 -36,111
## [23,] -127,556 166,778 -176,222 -119,222 295,444 -50,889
## [24,] -97,222 74,778 -371,556 -105,556 477,111 -56,889
## [25,] -125,556 96,444 -249,222 -29,222 278,444 -31,222
## [26,] -162,667 85,667 -205,000 -34,000 239,000 -12,000
## [27,] -187,778 106,222 -252,778 -72,111 324,889 -1,111
## [28,] -140,889 136,444 -132,556 -99,556 232,111 -22,889
## [29,] -232,556 278,111 -163,889 -48,222 212,111 -109,222
## [30,] 62,333 -32,000 -149,000 -124,000 273,000 -35,333
## [31,] 70,333 -50,667 -283,333 -2,333 285,667 -17,000
## [32,] 13,667 -19,667 -111,667 -60,000 171,667 4,667
## med.cigar non.cigar
## [1,] 30,000 -41,667
## [2,] 119,556 8,556
## [3,] -28,667 0,000
## [4,] 99,222 -2,778
## [5,] -25,444 -37,778
## [6,] 43,111 -12,889
## [7,] 20,000 -104,000
## [8,] 192,667 137,667
## [9,] 8,111 -27,556
## [10,] -120,444 28,222
## [11,] 31,889 35,889
## [12,] 117,111 -122,222
## [13,] 31,889 -2,111
## [14,] 38,778 -74,556
## [15,] 57,778 -96,889
## [16,] 87,333 -63,667
## [17,] -120,444 3,222
## [18,] -34,000 121,000
## [19,] -98,222 124,111
## [20,] -21,556 -165,889
## [21,] 121,333 -96,667
## [22,] 37,556 -1,444
## [23,] 94,778 -43,889
## [24,] 33,111 23,778
## [25,] 63,778 -32,556
## [26,] 49,667 -37,667
## [27,] 10,556 -9,444
## [28,] 50,778 -27,889
## [29,] 83,778 25,444
## [30,] 4,000 31,333
## [31,] -7,000 24,000
## [32,] -70,333 65,667
## [1] "Total utilities of profiles a cross of respondents:"
## [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9]
## [1,] 366 672 465 802 121 563 247 0 940
## [2,] 109 778 471 862 245 574 685 371 940
## [3,] 465 645 556 722 242 385 313 151 808
## [4,] 357 833 552 915 267 658 460 178 740
## [5,] 490 592 385 692 105 306 216 0 785
## [6,] 177 352 658 494 0 789 179 0 923
## [7,] 0 120 0 843 467 461 634 0 940
## [8,] 0 783 465 940 98 661 315 189 92
## [9,] 54 683 375 766 469 186 560 181 940
## [10,] 645 0 380 90 562 278 478 732 187
## [11,] 59 688 565 813 249 466 365 153 912
## [12,] 0 692 103 563 456 815 218 341 933
## [13,] 120 715 439 939 346 246 519 0 829
## [14,] 508 737 409 817 213 621 312 104 940
## [15,] 145 616 238 724 438 334 526 50 837
## [16,] 170 844 383 940 521 279 653 0 752
## [17,] 665 0 576 154 312 447 757 845 940
## [18,] 283 195 746 531 68 848 407 638 940
## [19,] 247 700 693 913 0 239 0 0 909
## [20,] 500 719 940 153 713 321 486 0 940
## [21,] 88 684 259 585 171 347 467 0 801
## [22,] 175 528 447 622 91 378 280 0 699
## [23,] 64 593 334 847 337 720 465 181 940
## [24,] 0 819 399 940 157 153 395 0 940
## [25,] 277 745 563 843 382 479 659 188 939
## [26,] 217 483 371 710 223 362 594 96 814
## [27,] 366 686 563 787 129 254 470 0 880
## [28,] 182 475 342 564 92 410 257 0 647
## [29,] 92 474 666 676 94 682 453 92 844
## [30,] 0 554 90 641 270 190 377 472 733
## [31,] 85 754 376 866 627 235 693 533 940
## [32,] 0 216 87 593 308 148 405 497 686
## [1] "Average importance of factors (attributes):"
## [1] 17,68 23,45 45,62 13,25
## [1] Sum of average importance: 100
#**************************** END *****************************************
# # Conjoint analysis step by step ----------------------------------------
#
# ## A experimental design for 32 subjects
# design <- exp.design
# design$id <- rownames(design)
# df <- design[rep(1:nrow(design),each=32),]
# rownames(df) <- NULL
# # df
#
# ## A vector of quality score for 32 subjects
# t.hedo <- t(ba.qual[,seq(1, ncol(ba.qual), by=3)])
# hedo.anova <- as.matrix(t.hedo)
# hedo <- melt(hedo.anova, id=1)
# colnames(hedo) = c("subject","product","score")
# hedo.plan <- cbind(df, hedo)
#
#
# ## model1: variables are considered as continuous variables
# ## For understand, dont use
# # model1 <- aov(score ~ color + shape + bruise + cigar, data=hedo.plan)
# # summary(model1) # factors: bruise > shape > color
#
# ## model2: varibales are considered as dummy variables
# model2 <- lm(score ~ factor(color) + factor(shape) + factor(bruise) + factor(cigar),
# data=hedo.plan)
# summary(model2)