Stability and Similarity of Clusters under Reduced Response Data

Publikation: Bidrag til konferencePaperForskningpeer review

Resumé

This study presents a validated recommendation on how to shorten the surveys while still obtaining segmentation-based insights that are consistent with the analysis of the full length version of the same survey. We use latent class analysis to cluster respondents based on their responses to a survey on human values. We first define the clustering performance based on stability and similarity measures for ten random subsamples relative to the complete set. We find foremost that the use of true binary scale can potentially reduce survey completion time while still providing sufficient response information to derive clusters with characteristics that resemble those obtained with the full Likert scale version. The main motivation for this study is to provide a baseline performance of a standard clustering tool for cases when it is preferable or necessary to limit survey scope, in consideration of issues like respondent fatigue or resource constraints.
This study presents a validated recommendation on how to shorten the surveys while still obtaining segmentation-based insights that are consistent with the analysis of the full length version of the same survey. We use latent class analysis to cluster respondents based on their responses to a survey on human values. We first define the clustering performance based on stability and similarity measures for ten random subsamples relative to the complete set. We find foremost that the use of true binary scale can potentially reduce survey completion time while still providing sufficient response information to derive clusters with characteristics that resemble those obtained with the full Likert scale version. The main motivation for this study is to provide a baseline performance of a standard clustering tool for cases when it is preferable or necessary to limit survey scope, in consideration of issues like respondent fatigue or resource constraints.

Konference

KonferenceThe 32nd Annual Conference of the Japanese Society for Artificial Intelligence
Nummer32
LandJapan
ByKagoshima
Periode05/06/201808/06/2018
Internetadresse

Citer dette

Litong-Palima, M., Albers, K. J., & Kano Glückstad, F. (2018). Stability and Similarity of Clusters under Reduced Response Data. Afhandling præsenteret på The 32nd Annual Conference of the Japanese Society for Artificial Intelligence, Kagoshima, Japan.
Litong-Palima, Marisciel ; Albers, Kristoffer Jon ; Kano Glückstad, Fumiko . / Stability and Similarity of Clusters under Reduced Response Data. Afhandling præsenteret på The 32nd Annual Conference of the Japanese Society for Artificial Intelligence, Kagoshima, Japan.4 s.
@conference{3f5710db14e747aca5857a06280044d5,
title = "Stability and Similarity of Clusters under Reduced Response Data",
abstract = "This study presents a validated recommendation on how to shorten the surveys while still obtaining segmentation-based insights that are consistent with the analysis of the full length version of the same survey. We use latent class analysis to cluster respondents based on their responses to a survey on human values. We first define the clustering performance based on stability and similarity measures for ten random subsamples relative to the complete set. We find foremost that the use of true binary scale can potentially reduce survey completion time while still providing sufficient response information to derive clusters with characteristics that resemble those obtained with the full Likert scale version. The main motivation for this study is to provide a baseline performance of a standard clustering tool for cases when it is preferable or necessary to limit survey scope, in consideration of issues like respondent fatigue or resource constraints.",
author = "Marisciel Litong-Palima and Albers, {Kristoffer Jon} and {Kano Gl{\"u}ckstad}, Fumiko",
year = "2018",
language = "English",
note = "null ; Conference date: 05-06-2018 Through 08-06-2018",
url = "https://confit.atlas.jp/guide/event/jsai2018/top",

}

Litong-Palima, M, Albers, KJ & Kano Glückstad, F 2018, 'Stability and Similarity of Clusters under Reduced Response Data' Paper fremlagt ved The 32nd Annual Conference of the Japanese Society for Artificial Intelligence, Kagoshima, Japan, 05/06/2018 - 08/06/2018, .

Stability and Similarity of Clusters under Reduced Response Data. / Litong-Palima, Marisciel; Albers, Kristoffer Jon; Kano Glückstad, Fumiko .

2018. Afhandling præsenteret på The 32nd Annual Conference of the Japanese Society for Artificial Intelligence, Kagoshima, Japan.

Publikation: Bidrag til konferencePaperForskningpeer review

TY - CONF

T1 - Stability and Similarity of Clusters under Reduced Response Data

AU - Litong-Palima,Marisciel

AU - Albers,Kristoffer Jon

AU - Kano Glückstad,Fumiko

PY - 2018

Y1 - 2018

N2 - This study presents a validated recommendation on how to shorten the surveys while still obtaining segmentation-based insights that are consistent with the analysis of the full length version of the same survey. We use latent class analysis to cluster respondents based on their responses to a survey on human values. We first define the clustering performance based on stability and similarity measures for ten random subsamples relative to the complete set. We find foremost that the use of true binary scale can potentially reduce survey completion time while still providing sufficient response information to derive clusters with characteristics that resemble those obtained with the full Likert scale version. The main motivation for this study is to provide a baseline performance of a standard clustering tool for cases when it is preferable or necessary to limit survey scope, in consideration of issues like respondent fatigue or resource constraints.

AB - This study presents a validated recommendation on how to shorten the surveys while still obtaining segmentation-based insights that are consistent with the analysis of the full length version of the same survey. We use latent class analysis to cluster respondents based on their responses to a survey on human values. We first define the clustering performance based on stability and similarity measures for ten random subsamples relative to the complete set. We find foremost that the use of true binary scale can potentially reduce survey completion time while still providing sufficient response information to derive clusters with characteristics that resemble those obtained with the full Likert scale version. The main motivation for this study is to provide a baseline performance of a standard clustering tool for cases when it is preferable or necessary to limit survey scope, in consideration of issues like respondent fatigue or resource constraints.

M3 - Paper

ER -

Litong-Palima M, Albers KJ, Kano Glückstad F. Stability and Similarity of Clusters under Reduced Response Data. 2018. Afhandling præsenteret på The 32nd Annual Conference of the Japanese Society for Artificial Intelligence, Kagoshima, Japan.