Testing for cultural measurement equivalence in research on domestic and international tertiary students' fear of crime
| Date | 01 September 2017 |
| Published date | 01 September 2017 |
| DOI | 10.1177/0004865815604197 |
| Subject Matter | Articles |
Australian & New Zealand
Journal of Criminology
2017, Vol. 50(3) 397–418
!The Author(s) 2015
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0004865815604197
journals.sagepub.com/home/anj
Article
Testing for cultural
measurement equivalence
in research on domestic and
international tertiary students’
fear of crime
Lin Xiong
Discipline of Psychology, College of Arts, Victoria University,
Melbourne, Victoria, Australia
Christopher Nyland
Department of Management, Monash University, Melbourne, Australia
Kosmas X Smyrnios
School of Management, RMIT University, Melbourne, Victoria, Australia
Abstract
Objectives: Education institutions routinely instruct students on how to remain safe from
crime. We hold that this instruction and much of the associated practice might be problematic,
because none of the researchers who have contrasted the fears and the victimization avoidance
strategies of domestic and international students have tested for cultural measurement equiva-
lence. This study aims to examine, whether cultural measurement equivalence exists when
domestic and international tertiary students respond to fear of crime-related measures.
Methods: This cross-sectional study involved 1170 tertiary students across four Melbourne-
based universities, Australia. Multiple group confirmatory factor analyses with covariance and
mean structures, using structural equation modeling, were used to test whether the same
constructs were measured across international and local tertiary students.
Results: The two cohorts hold the same conceptual frame of reference when responding to
the measurement items. However, the cohorts display different true score values in relation
to a number of questionnaire items associated with fear of crime, perceptions of safety, and
avoidance behavior.
Conclusions: This study suggests that researchers need to render testing for cultural meas-
urement equivalence standard practice, when undertaking cross-cultural studies of student
safety and that such practice should also be incorporated into student safety programs.
Corresponding author:
Lin Xiong, Discipline of Psychology, College of Arts, Victoria University, Level 2, Building 4, Werribee Campus,
Melbourne, Victoria 3030, Australia.
Email: lin.xiong@vu.edu.au
Keywords
Cultural measurement equivalence, fear of crime, perceived risk, perception of safety,
avoidance behavior
Introduction
In 2014, almost 5 million tertiary students were studying outside their home country.
Given that many of these sojourners are highly vulnerable, researchers have sustained an
ongoing effort seeking to clarify the factors that determine which students go where and
why (Maceady & Tucker, 2011). As part of this effort, drawing on an online survey
involving participants from over 80 countries, the British Council (2012) reports that
through 2006–2012, the importance of safety from crime moved from being the 17th to
the 5th most important factor determining where students study. Two years later,
Hobsons (2014) reported safety had become the fourth most important determinant.
The British Council (2012) concluded that the increased weighting students now accord
safety is primarily due to their growing utilization of social media to gain information
that supplements the advertising of education agents and the widespread publicity
accorded violent attacks on international students studying in Australia.
When students are asked if they believe themselves to be vulnerable to racial vilifica-
tion, physical attack, and robbery research reveals that international students score
higher than their domestic counterparts (Babacan et al., 2010; Larsen, Payne, &
Tomison, 2011; Marginson, Nyland, Sawir, & Forbes-Mewett, 2010; Nyland, Forbes-
Mewett, & Marginson, 2010). Such research includes works that have been described as
‘sobering, scholarly, and thoughtful’ (Tight, 2011, p. 250). These findings inform the
instruction and practice of many of the educators, counselors, and law enforcement
officers charged with keeping students safe from crime (Nelson, 2014).
Notwithstanding, findings generated by most, if not all, of these studies are necessarily
problematic, because none have determined whether domestic and international students
have a common understanding of the questions asked by researchers. In brief, it appears
that no previous study has tested the research instruments utilized for cultural measure-
ment equivalence (CME) prior to analyzing interview or survey data this being the
‘possibility that interpretations of psychological measurements, assessments, and obser-
vations are [or are not] similar if not equal across different ethnocultural
populations’(Trimble, 2014, p. 1). Furthermore, that researchers, teachers, and practi-
tioners who address student safety should test for CME is evidenced by the fact that
theoretical and empirical studies across a number of fields have identified substantial
cross-cultural differences in student understanding of what they are asked and taught
(De Vita, 2002). This divergence can be attributed to a range of factors including dif-
fering cultural/ethnic backgrounds and language skills (Poyrazli & Lopez, 2007), level of
familiarity with the host environment (Forbes-Mewett & Nyland, 2008; Marginson
et al., 2010), level of social support (Li & Kaye, 1998), and differing coping strategies
(Khawaja & Dempsey, 2008).
In testing for CME, we draw on Vandenberg and Lance (2000, p. 5) who advise that
when conducting cross-cultural or multi-group studies, scholars should determine
whether participants from different cultural and/or ethnic backgrounds: (a) interpret a
398 Australian & New Zealand Journal of Criminology 50(3)
To continue reading
Request your trial