TY - JOUR
T1 - Measuring the digital divide
T2 - A modified benefit-of-the-doubt approach
AU - Mahdiloo, Mahdi
AU - Andargoli, Amir E.
AU - Toloo, Mehdi
AU - Harvie, Charles
AU - Duong, Thach Thao
N1 - Funding Information:
The authors would like to thank five anonymous reviewers for their constructive comments which improved an earlier version of this paper. The research of Dr. Mehdi Toloo was supported by the Czech Science Foundation through project No. 19- 13946S .
Publisher Copyright:
© 2022 Elsevier B.V.
PY - 2023/2/15
Y1 - 2023/2/15
N2 - In this paper, a modified composite index is developed to measure digital inclusion for a group of cities and regions. The developed model, in contrast to the existing benefit-of-the-doubt (BoD) composite index literature, considers the subindexes as non-compensatory. This new way of modeling results in three important properties: (i) all subindexes are taken into account when assessing the digital inclusion of regions and are not removed (substituted) from the composite index, (ii) in addition to an overall composite index (aggregation of the subindexes), partial indexes (aggregated scores for each subindex) are also provided so that weak performances can be detected more effectively than when only the overall index is measured, and (iii) compared with current BoD models, the developed model has improved discriminatory power. To demonstrate the developed model, we use the Australian digital inclusion index as a real-world example.
AB - In this paper, a modified composite index is developed to measure digital inclusion for a group of cities and regions. The developed model, in contrast to the existing benefit-of-the-doubt (BoD) composite index literature, considers the subindexes as non-compensatory. This new way of modeling results in three important properties: (i) all subindexes are taken into account when assessing the digital inclusion of regions and are not removed (substituted) from the composite index, (ii) in addition to an overall composite index (aggregation of the subindexes), partial indexes (aggregated scores for each subindex) are also provided so that weak performances can be detected more effectively than when only the overall index is measured, and (iii) compared with current BoD models, the developed model has improved discriminatory power. To demonstrate the developed model, we use the Australian digital inclusion index as a real-world example.
KW - Benefit of the doubt
KW - Composite index
KW - Digital divide
KW - Digital inclusion index
KW - Non-compensatory subindexes
UR - http://www.scopus.com/inward/record.url?scp=85144419327&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85144419327&partnerID=8YFLogxK
U2 - 10.1016/j.knosys.2022.110191
DO - 10.1016/j.knosys.2022.110191
M3 - Article
AN - SCOPUS:85144419327
SN - 0950-7051
VL - 261
JO - Knowledge-Based Systems
JF - Knowledge-Based Systems
M1 - 110191
ER -