The Recovery of Structure in Linear and Ordinal Data: Indscal versus ALSCAL

Sharon L. Weinberg; Violeta C. Menil

doi:10.1207/s15327906mbr2802_4

The Recovery of Structure in Linear and Ordinal Data: Indscal versus ALSCAL

Sharon L. Weinberg, Violeta C. Menil

Applied Statistics, Social Science, and Humanities

Research output: Contribution to journal › Article › peer-review

Abstract

The ability of three-way INDSCAL and ALSCAL to recover true structure in proximity data is examined in a Monte Carlo study using data based on two-dimensional configurations. Extending earlier work in this area, data were simulated to vary on four factors: number of subjects (15 and 30), number of stimuli (12 and 20), amount of error (0.10 and 0.70), and type of monotonic transformation (linear, square, logarithmic, general cubic, and rank order). Recovery was measured using two indicators, corresponding to the two sets of parameters estimated by these techniques: recovery of true stimulus dimensions, and recovery of true subject weights. In this study, INDSCAL outperformed metric and nonmetric versions of ALSCAL under all conditions. Implications for practice and future research are discussed.

Original language	English (US)
Pages (from-to)	215-233
Number of pages	19
Journal	Multivariate Behavioral Research
Volume	28
Issue number	2
DOIs	https://doi.org/10.1207/s15327906mbr2802_4
State	Published - Apr 1 1993

ASJC Scopus subject areas

Statistics and Probability
Experimental and Cognitive Psychology
Arts and Humanities (miscellaneous)

Access to Document

10.1207/s15327906mbr2802_4

Cite this

@article{3df74a7e58d6407e99ecd50c13bb31f5,

title = "The Recovery of Structure in Linear and Ordinal Data: Indscal versus ALSCAL",

abstract = "The ability of three-way INDSCAL and ALSCAL to recover true structure in proximity data is examined in a Monte Carlo study using data based on two-dimensional configurations. Extending earlier work in this area, data were simulated to vary on four factors: number of subjects (15 and 30), number of stimuli (12 and 20), amount of error (0.10 and 0.70), and type of monotonic transformation (linear, square, logarithmic, general cubic, and rank order). Recovery was measured using two indicators, corresponding to the two sets of parameters estimated by these techniques: recovery of true stimulus dimensions, and recovery of true subject weights. In this study, INDSCAL outperformed metric and nonmetric versions of ALSCAL under all conditions. Implications for practice and future research are discussed.",

author = "Weinberg, {Sharon L.} and Menil, {Violeta C.}",

year = "1993",

month = apr,

day = "1",

doi = "10.1207/s15327906mbr2802_4",

language = "English (US)",

volume = "28",

pages = "215--233",

journal = "Multivariate Behavioral Research",

issn = "0027-3171",

publisher = "Psychology Press Ltd",

number = "2",

}

TY - JOUR

T1 - The Recovery of Structure in Linear and Ordinal Data

T2 - Indscal versus ALSCAL

AU - Weinberg, Sharon L.

AU - Menil, Violeta C.

PY - 1993/4/1

Y1 - 1993/4/1

N2 - The ability of three-way INDSCAL and ALSCAL to recover true structure in proximity data is examined in a Monte Carlo study using data based on two-dimensional configurations. Extending earlier work in this area, data were simulated to vary on four factors: number of subjects (15 and 30), number of stimuli (12 and 20), amount of error (0.10 and 0.70), and type of monotonic transformation (linear, square, logarithmic, general cubic, and rank order). Recovery was measured using two indicators, corresponding to the two sets of parameters estimated by these techniques: recovery of true stimulus dimensions, and recovery of true subject weights. In this study, INDSCAL outperformed metric and nonmetric versions of ALSCAL under all conditions. Implications for practice and future research are discussed.

AB - The ability of three-way INDSCAL and ALSCAL to recover true structure in proximity data is examined in a Monte Carlo study using data based on two-dimensional configurations. Extending earlier work in this area, data were simulated to vary on four factors: number of subjects (15 and 30), number of stimuli (12 and 20), amount of error (0.10 and 0.70), and type of monotonic transformation (linear, square, logarithmic, general cubic, and rank order). Recovery was measured using two indicators, corresponding to the two sets of parameters estimated by these techniques: recovery of true stimulus dimensions, and recovery of true subject weights. In this study, INDSCAL outperformed metric and nonmetric versions of ALSCAL under all conditions. Implications for practice and future research are discussed.

UR - http://www.scopus.com/inward/record.url?scp=21144460221&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21144460221&partnerID=8YFLogxK

U2 - 10.1207/s15327906mbr2802_4

DO - 10.1207/s15327906mbr2802_4

M3 - Article

AN - SCOPUS:21144460221

SN - 0027-3171

VL - 28

SP - 215

EP - 233

JO - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

IS - 2

ER -

The Recovery of Structure in Linear and Ordinal Data: Indscal versus ALSCAL

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this