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

Sharon L. Weinberg, Violeta C. Menil

Research output: Contribution to journalArticle

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 languageEnglish (US)
Pages (from-to)215-233
Number of pages19
JournalMultivariate Behavioral Research
Volume28
Issue number2
DOIs
StatePublished - Apr 1 1993

Fingerprint

Ordinal Data
INDSCAL
Recovery
Weights and Measures
Rank order
Monte Carlo Study
Linear transformation
Monotonic
Proximity
Logarithmic
Vary
Metric
Configuration
Stimulus

ASJC Scopus subject areas

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

Cite this

The Recovery of Structure in Linear and Ordinal Data : Indscal versus ALSCAL. / Weinberg, Sharon L.; Menil, Violeta C.

In: Multivariate Behavioral Research, Vol. 28, No. 2, 01.04.1993, p. 215-233.

Research output: Contribution to journalArticle

@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 = "4",
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

VL - 28

SP - 215

EP - 233

JO - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

SN - 0027-3171

IS - 2

ER -