A Hybrid Structural Equation Modeling and Fuzzy-set Qualitative Comparative Analysis Framework for Using Limited Samples to Relate Cycling Performance to Behavioral Traits
Jafari, Mahsa LU ; Persaud, Bhagwant and D'agostino, Carmelo LU
(2025)
In Traffic Safety Research
9.
- Abstract
The objective of this investigation was to explore a hybrid application of Structural Equation Modelling (SEM) and Fuzzy-set Qualitative Comparative Analysis (FsQCA) to assess the impact of safety-related behavioural traits on the performance of bicyclists using a limited sample of data as a case study. These behavioural characteristics of cyclists are fundamental when specific cyclist groups are targeted for safety countermeasures. The data were collected from individuals who completed a survey and participated in bicycle simulator experiments. Each of the items in the survey aimed to describe how keen the rider is to take risks. The motivation for the study is the recognition that traditional statistical methods have limitations... (More)
The objective of this investigation was to explore a hybrid application of Structural Equation Modelling (SEM) and Fuzzy-set Qualitative Comparative Analysis (FsQCA) to assess the impact of safety-related behavioural traits on the performance of bicyclists using a limited sample of data as a case study. These behavioural characteristics of cyclists are fundamental when specific cyclist groups are targeted for safety countermeasures. The data were collected from individuals who completed a survey and participated in bicycle simulator experiments. Each of the items in the survey aimed to describe how keen the rider is to take risks. The motivation for the study is the recognition that traditional statistical methods have limitations related to the model structure and the type of variables they can analyze, especially where sample sizes are limited. To address these limitations and to estimate complex relationships between variables, SEM can be used to assess the individual effect of each variable on the response variable(s). In the real-world context, however, a combination of variables can affect response variables. To address this issue, this study used the hybrid SEM-FsQCA approach, in which SEM was applied as the first step in analyzing the latent behavioural variables. Then, using the outputs of the previous step, FsQCA was applied to assess the effect of combinations of variables on the performance of cyclists. Based on SEM, none of the factors significantly affected performance, likely due to the low sample size. However, when applying FsQCA, it was observed that combinations of factors significantly affect performance. This hybrid approach was seen to be promising for the case study application and, in that context, to have the potential, even with smaller samples, to significantly contribute to a deep understanding of the safety-related behavioral traits, which can be used in designing targeted countermeasures to improve their safety.
(Less)
- author
- Jafari, Mahsa
LU
; Persaud, Bhagwant
and D'agostino, Carmelo
LU
- organization
- publishing date
- 2025-01-01
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- behaviroal traffic safety, bicycle safety countermeasures, cyclist safety, fuzzy-set qualitative comparative analysis, structural equation modeling
- in
- Traffic Safety Research
- volume
- 9
- article number
- e000095
- publisher
- Lund University, Faculty of Engineering
- external identifiers
-
- scopus:105007952191
- ISSN
- 2004-3082
- DOI
- 10.55329/qjgq4392
- language
- English
- LU publication?
- yes
- additional info
- Publisher Copyright: © 2025, Lund University Faculty of Engineering. All rights reserved.
- id
- 134da837-ad15-4189-ba44-5cb8d70d89ac
- date added to LUP
- 2025-08-14 15:44:08
- date last changed
- 2025-08-15 10:07:18
@article{134da837-ad15-4189-ba44-5cb8d70d89ac,
abstract = {{<p>The objective of this investigation was to explore a hybrid application of Structural Equation Modelling (SEM) and Fuzzy-set Qualitative Comparative Analysis (FsQCA) to assess the impact of safety-related behavioural traits on the performance of bicyclists using a limited sample of data as a case study. These behavioural characteristics of cyclists are fundamental when specific cyclist groups are targeted for safety countermeasures. The data were collected from individuals who completed a survey and participated in bicycle simulator experiments. Each of the items in the survey aimed to describe how keen the rider is to take risks. The motivation for the study is the recognition that traditional statistical methods have limitations related to the model structure and the type of variables they can analyze, especially where sample sizes are limited. To address these limitations and to estimate complex relationships between variables, SEM can be used to assess the individual effect of each variable on the response variable(s). In the real-world context, however, a combination of variables can affect response variables. To address this issue, this study used the hybrid SEM-FsQCA approach, in which SEM was applied as the first step in analyzing the latent behavioural variables. Then, using the outputs of the previous step, FsQCA was applied to assess the effect of combinations of variables on the performance of cyclists. Based on SEM, none of the factors significantly affected performance, likely due to the low sample size. However, when applying FsQCA, it was observed that combinations of factors significantly affect performance. This hybrid approach was seen to be promising for the case study application and, in that context, to have the potential, even with smaller samples, to significantly contribute to a deep understanding of the safety-related behavioral traits, which can be used in designing targeted countermeasures to improve their safety.</p>}},
author = {{Jafari, Mahsa and Persaud, Bhagwant and D'agostino, Carmelo}},
issn = {{2004-3082}},
keywords = {{behaviroal traffic safety; bicycle safety countermeasures; cyclist safety; fuzzy-set qualitative comparative analysis; structural equation modeling}},
language = {{eng}},
month = {{01}},
publisher = {{Lund University, Faculty of Engineering}},
series = {{Traffic Safety Research}},
title = {{A Hybrid Structural Equation Modeling and Fuzzy-set Qualitative Comparative Analysis Framework for Using Limited Samples to Relate Cycling Performance to Behavioral Traits}},
url = {{http://dx.doi.org/10.55329/qjgq4392}},
doi = {{10.55329/qjgq4392}},
volume = {{9}},
year = {{2025}},
}