Development of a Statistical Model to Predict Australian Flight Students’ Valuation of Aviation Safety
Development of a Statistical Model to Predict Australian Flight Students' Valuation of Aviation Safety
Michael Chiu 0
0 University of South Australia
Scholarly Commons Citation
Part of the Aviation and Space Education Commons; and the Aviation Safety and Security
A safer aviation industry, from training to air transport, is in everyone?s best interest.
The development of a positive safety culture as part of a robust Safety Management System is an
important part of assuring aviation safety (International Civil Aviation Organization [ICAO],
2013a). A clearer understanding of individual attitudes, and how they are composed, may be
insightful to help establish the context of safety culture.
A more focused look at individual attitudes, and their relationship to safety culture as a
whole, solicits the need to consider the practical application of the value of individual attitudes.
Flight schools, as one stakeholder group, have a vested interest in working with students that
have both a high chance of success in training and the least possible liability; a flight student?s
own attitude toward safety and engagement in safe behaviors are valued attributes.
Complicating the issue of culture and its effect on the overall operation of SMS is the
complex relationship between individual and collective thinking: the idea of a collection of
individual attitudes being equivalent to group-level culture is neither assumed nor well
understood. Much attention has shifted to the role of culture in an SMS that operates as intended
? and safety culture is a notable subset (Leib & Lu, 2013). However, the introduction of the
human element is notoriously difficult to isolate given the inherent entanglement of culture,
organizational structure, and processes (Guldenmund, 2007). This presents both a conceptual
and a methodological challenge, with research often emphasizing either individual attitudes or
group-level values without accommodating the other (Leung & Cohen, 2011).
Bearing in mind the interrelationship of culture, organizational structure, and processes as
well as the need to better understand individual/collective values, understanding safety culture
should begin by observing these phenomena as independently as possible. Research of this
nature has already been done: Adjekum (2014), Siu, Phillips, and Leung (2003), Freiwald,
LenzAnderson, and Baker (2013), Dutcher (2001), and others have assessed individual attitudes
toward safety and safety culture.
The purpose of this study was to explore the development of a predictive model to
determine which factors were more significant in a flight student?s valuation of safety. A better
understanding of the more relevant factors of one?s attitude toward safety can help aviation
organizations make informed decisions to direct resources to support individual flight students?
safety behaviors. Additionally, it may provide a dimension of assistance for the candidate
selection process at organizations that are space- and resource-constrained. Lastly, this study is
intended to stimulate future research linking safety attitudes with the development of a positive
Review of Literature
Safety culture itself is a complicated construct, and many have tried to define it with
varying degrees of specificity. The Commercial Aviation Safety Team of the European Strategic
Safety Initiative (Piers, Montijin, & Balk, 2009) defines safety culture as a set of safety related
attitudes, values, perceptions, and beliefs that are typically shared among all members within an
organization. In more generic terms, it represents the ?way of doing things,? and the priority of
safety within a particular group
. One component of safety culture also reflects the
commitment to engage in safety-related behaviors (Federal Aviation Administration [FAA],
2007), and Ek, Akselsson, Arvidsson, and Johansson (2007) found that commitment includes the
ability and willingness to prevent safety risk as well as the awareness to avoid the occurrence of
Although the concept of safety culture can be considered fairly nebulous, it is
undoubtedly a fundamental component of an organization?s ability to manage safety related
(Glendon & Stanton, 2000)
, and it is a key factor in determining whether an organization
is successfully managing safety (Bailey, 1997; Kennedy & Kirwan, 1998). ICAO (2006)
indicates that the level of operational safety can be maximized by a positive safety culture by
encouraging members within an organization to respect each other and create a harmonious
(Simon & Cistaro, 2009)
. Moreover, it enables all members within the
group to be treated fairly and equally by empowering responsibility and authority (Bos & Lu,
2007). Because of these properties, the operational risks and errors are more likely to be
anticipated if a positive safety culture is successfully installed in the organization
Halford & Goglia, 2008)
. Given this, the importance of a positive safety culture at an aviation
organization cannot be understated. It is understandable then that
von Thaden and Gibbons
) have indicated that safety culture related research is a valuable part of understanding of
organizational performance and procedures.
From both an operational and academic standpoint, safety culture and safety climate are
closely related concepts that share a similar abstract idea (Reichers & Schneider, 1990), but are
nonetheless distinguishable from each other. As suggested by Guldenmund (2007), they are
inseparable components in an organization that help measure the importance of safety by
adopting different approaches. However, the difference between the two concepts lies in that
safety climate is derived from safety culture and should be considered the dimension of safety
culture that can be observed directly in an organization
. Simply put, where
?safety culture? describes the overall collective attitude toward safety at an organization, ?safety
climate? can be a snapshot of safety perception at any given time. Along these lines,
Guldenmund (2007) notes that research on safety climate is basically research on attitudes.
From a methodological perspective, it is difficult to discern whether answers provided by
respondents reflect culture itself or attitudes from which to make inferences about culture, and
conclusions from this kind of methodology should be drawn carefully.
Safety culture can be divided into more discrete components, which provide both a
clearer understanding of the composition of safety culture as well as a methodological
framework for researching safety culture. These five safety sub-cultures consist of informed
culture, reporting culture, just culture, flexible culture, and learning culture (Reason, 2000).
Informed culture represents the personnel?s awareness of factors within an organization that
threaten safety, including but not limited to working environment, operational equipment, and
other staff members. Reporting culture is considered one?s willingness to report safety related
issues. It is built upon by establishing a blame-free environment and maintaining confidentiality
of the people who initiated the reports. A just culture includes a reporting environment where
unintentional errors are not punished. However, a line has to be drawn to establish unintentional
errors from deliberate acts of malice, and disciplinary actions should be taken if the latter occurs.
Flexible culture is indicated by the ability of an organization to change and adapt to a new
working environment (Reason, 1997). Finally, learning culture reflects an organization?s
eagerness to learn from mistakes and improve overall performance
it is important to notice that the five sub-cultures listed above are inter-dependent to each other,
and should one of them fail to be present, the other components would be significantly degraded.
While these various safety sub-cultures can provide a more concrete understanding of
what a safety culture is, it is another (and more complicated) matter to assess, predict, and
influence safety culture.
Assessing, Influencing, and Predicting Safety Culture
Considering the importance of safety culture to an organization, it is useful to discuss
how safety culture can be influenced, as well as which factors have been demonstrated to
meaningfully impact safety culture. This tends to be very difficult, as what predicts and/or
influences safety culture may vary depending on the industry and organization. Guldenmund
(2007) describes behavior at an organization ultimately as a product of the combination of
culture, structure, and processes, all of which are related:
...any organization?s culture cannot be isolated from its structure or processes. In
carrying out the processes and coping with difficulties groups of people develop a
culture, either despite of or because of some particular structure. (p. 737)
The relationship between culture and structure may be broadly relevant across industries,
especially those that depend on a strong safety culture such as aviation or medicine. Yet,
et al. (2012)
found that the number of workers? compensation claims to be a significant predictor
of safety culture at hospitals; one that does not exist in the aviation industry. It is not
independent from other organizational aspects, and may be industry specific, which provides
some explanation as to the difficulty of observing and measuring changes to safety culture.
Given that culture is both fostered and constrained by structure and processes, it is not
surprising that safety culture can be influenced by changes to operations (Ek et al., 2007). For
example, transitioning the focus of a business model to cost-effectiveness can result in an
alteration of organizational behavior and weaken the state of safety, possibly leading to the
occurrence of an incident or accident (Rasmussen, 1997). Even the roles of staff and different
natures of job positions can lead to distinctive safety culture among groups even within the same
organization. Adjekum (2014) found that the safety culture specifically of the front-line
operational personnel, including flight students and flight instructors, have a strong impact on
successful safety management at an aviation organization. Ek et al. (2007) demonstrated that the
top-level management?s commitment to safety can be inconsistent with the safety attitude of the
lower-ranked staff because of the difference in the associated responsibilities. On the other hand,
operational staff may perceive risks more accurately and exhibit better reporting behavior when
compared to administrative staff due to the exposure to safety risks. Factors such as staff duties
and proximity to risks and hazards can often be a direct function of management?s
decisionmaking at organizations. Differences in the strength of safety culture in different employee
groups and at different levels is consistent with Guldenmund?s (2007) description of the
culturestructures-processes model, considering their weights change at different organizational levels.
There is certainly the potential for individual attitudes toward safety to vary, and, as indicated by
Cooper (2000), this possible deviation in values and attitudes can become an obstacle to
maintaining a safe organization. It is worth noting that the linkage of safety culture, structures,
and processes, suggests that, to at least some extent, safety culture can be consciously influenced.
With this consideration, safety training can play a meaningful role in influencing safety culture.
Safety training at an aviation organization is largely dependent on the legal requirements
of the aviation regulator of its state. However, since the development of the Safety Management
Manual (SMM), a comprehensive Safety Management Systems (SMS) guidebook (ICAO, 2006),
SMS, including safety training, has experienced sustained growth across global aviation
organizations. This has been accelerated by the development of international safety management
standards and recommended practices (SARPs): ICAO Annex 19: Safety Management
. Though ICAO standards constitute a non-legally binding body of policies and
procedures, they are generally treated as regulatory in nature by the vast majority of participating
states, which have developed their own legally-binding legislation mirroring the ICAO SARPs.
The current 3rd edition of the Safety Management Manual (SMM) covers a wide range of SMS
details, including the theoretical basis for safety risk management, the importance of safety
culture and organizational factors, and the practical guidelines for implementation and operation
of a Safety Management System. This includes a well-developed discussion of the importance,
need, and recommendations for Safety Promotion, especially the establishment of SMS related
training (ICAO, 2013a). Many states have adopted SMS training regulations, which are similar
to (if not nearly the same) as what is outlined in the SMM.
Safety Management Systems training should, according to the SMM, cover a variety of
safety aspects such as organizational policy and objectives, the safety risk management process,
and methods of communication. It should also include training in safety philosophy and theory.
Lastly, it must include practical aspects of participating in the organization?s SMS, such as the
process for submitting hazard reports, etc. (ICAO, 2013a).
There can be little doubt as to the informative component of SMS training; that is, the
exchange of information directly related to the SMS. This includes awareness of the
organization?s expectations of individuals at all levels, how to participate in a hazard reporting
system, and the stated policies and objectives of the leadership regarding safety. While safety
training is expected to positively influence safety culture (ICAO, 2013a), little research has
explored the connection between increased training and individual valuation of safety.
The Role of the Individual in Safety Culture
Safety culture is a description of the collective. While it does not make sense to describe
an individual as having a ?safety culture,? the role of the individual in the development of safety
should not be ignored. Dutcher, Carrick, and Smith (2003) found that individual attitudes are
very relevant to understanding the development of safety culture and, just as importantly, may be
a more accurate assessment of safety culture than traditional statistics of accident and incident
rates. Given the importance of safety culture as a whole on an organization?s successful safety
management, and with recognition that safety culture represents in part a collection of individual
(potentially differing) attitudes, it is not unreasonable to suggest that research explore safety
culture valuation at the individual level.
Individual demographic factors have been connected to aspects of safety culture. Ek et
al. (2007) used factors such as gender, age, time in company, and time in position as potential
factors that had an impact on how an individual employee perceived aspects of safety at an air
traffic control organization. Though few significant effects were demonstrated between groups,
the study investigated individuals? perceptions of the organization as a whole rather than address
their own behavior.
Many studies have connected gender, age, etc. to safety and risk attitudes. Liu, Liu,
Wang, Zhang, and Wang (2013), and
Yavuz and Welch (2010)
, demonstrate examples of gender
differences significantly affecting individuals? engagement in high-risk behavior and how they
perceive safety measures in urban transit, respectively. The latter study even went so far as to
suggest the need for development of gender-specific safety enhancements so as to maximize the
safety benefit as perceived by men and women. Similarly, age has been demonstrated to be a
meaningful distinction with safety attitudes.
Siu et al. (2003)
found significant differences
between age groups of construction workers in Hong Kong with regard to safety attitude and
safety performance. Sawacha, Naoum, and Fong (1999) found age, job experience, and training
all to have strongly significant effects on safety performance. Dutcher (2001) found that pilot
experience improved their perception of the importance of safety training.
Research has also explored behavior on the individual level with respect to safety
subcultures, such as reporting culture. Chiang, Hsiao, Lin, and Lee (2011) found a correlation
between level of reporting culture and the level of safety culture: hospital nurses who reported
more frequently and in greater detail were found to be those with a higher level of safety
valuation. Similarly, engagement (or participation) has also been found to be an important
predictor of a strong safety culture, as a stronger reporting culture and aviation safety
information seeking were demonstrated to be outcomes of high engagement (Chiang et al., 2011;
Freiwald et al., 2013).
While there have been many studies that have looked at individual attitudes? relationship
to safety concepts, predicting the strength of safety culture itself is extraordinarily challenging.
Frazier, Ludwig, Whitaker, and Roberts (2013) suggests that ??it is difficult to find a single
survey that exhibits any predictive quality in actual safety performance or statistics? (p. 18). The
present study does not attempt to predict safety culture, only an individual?s valuation of safety;
further research will be necessary to explore the link between attitudes of individuals and safety
The present study sought to develop a predictive model for a flight student?s valuation of
safety as a function of research-based predictors of gender, amount of flight experience, extent of
safety training, and engagement. In addition, to consider the influence of these factors in
context, this study also sought to determine the baseline cultural level of safety valuation to
establish whether a significant difference existed between local individuals and those with safety
training and flight experience.
This was a quantitative study that sought to develop a predictive model for valuation of
safety on the basis of five possible factors: age, flight experience, gender, depth of safety
training, and engagement. To answer RQ1, the study used a one-way ANOVA to determine
whether the safety valuation of flight students/recently trained pilots differed significantly from
local individuals who had no experience in aviation. In addition, to answer RQ2, significant
factors were used to develop the best possible model of safety valuation using linear regression.
Two groups of participants were used in this study. One represented flight students and
pilots having recently completed training. Qualified participants in this group were required to
be at least 18 years of age, identify as culturally Australian, have not more than 250 hours of
combined dual/solo flight experience (to ensure more recent training and student level of
experience), and have received aviation safety training since starting their flight program. The
second group of participants (for the purpose of RQ1) were local Australians. Qualified
individuals had to self-identify as representing Australian culture and have no flight training or
professional aviation background. A priori power analysis (using G*Power) indicated that
achieving statistical power of 0.8 with a 0.4 (large) effect size at a .05 significance level with two
groups required a minimum total sample of 52 individuals.
Surveys for the aviation group of participants were disseminated through social media.
This was to accommodate the inconsistent schedule of flight students and to appeal to that
particular age group. Sharing a survey link through social media allowed for its distribution via
researchers? social and professional networks to reach a higher number of potential participants.
This recruitment method also afforded the opportunity to find participants not connected to any
one particular flight school. Surveys for the local group were solicited in person in the city center
of a major Australian city. Participants were recruited in a public space by researchers and asked
to participate in the study using an iPad. This was done on a weekend day to increase the
likelihood that the sample would accurately represent city locals.
The aviation participants were given a survey with the following components:
1. Basic non-identifying demographic questions ? including gender, age, and flight
experience. All were self-disclosed by participants and both age and flight experience
were treated as continuous variables.
2. Safety valuation questions ? 15 survey items addressed safety valuation. This portion of
the survey tool was adapted from Liao (2015) and included three sets of five items,
touching on safety subcultures of just culture, reporting culture, and learning culture.
These were assessed using a five-point Likert scale ranging from strongly disagree to
3. Engagement questions ? five survey items adapted from
Thorp et al. (2012)
individuals? level of engagement with their training organization/employer. These items
were also evaluated using a five-point Likert scale.
4. Safety Training questions ? six survey items addressed the perceived quality of training
that the students received. The safety training questions did not specify the time, form
and location of training, only the participant?s self-identified belief of the extent to which
they received some form of training on that particular subject. The subjects from each
question were adapted from ICAO (2013a) and an adapted four-point Likert scale ranging
from not covered to covered in depth was used to solicit responses.
The second participant group, consisting of local Australians that had no aviation
experience, received only the 15-item safety valuation questions. Survey questions are listed in
IBM SPSS Statistics 24 was used for all analysis. For the aviation group, a total of 27
qualified responses were received. This included two female and 25 male participants. Table 1
summarizes the means and standard deviations for the remaining demographic items as well as
safety valuation, safety training, and engagement.
Summary of Pilot Group Responses
To ensure reliability of multiple survey items addressing single concepts, Cronbach?s
alpha reliability coefficients were obtained for Safety Valuation and Engagement. Safety
Training was not included because safety training items were specifically addressing different
aspects of training which may be emphasized differently at organizations. The coefficients for
Safety Valuation and Engagement were determined to be .700 and .768, respectively.
Comparison of Aviation and Non-Aviation Participants
There were 30 participants in the non-aviation group representing local valuation of
safety. The means and standard deviation for Safety Valuation (from Table 1) for the aviation
group was found to be 3.93 and .340, respectively. In addition, the mean and standard deviation
of the non-aviation group of participants was found to be 3.84 and .530, respectively, as
indicated in Figure 1.
A test of homogeneity of variances yielded a Levene Statistic of 1.817 corresponding to a
.183 significance level, suggesting the assumption of homogeneity of variance was not violated.
The mean difference between the two groups was .101, which corresponded to an
Fstatistic of .495 and a significance level of .485, which was not significant at the .05 level.
Developing the regression model began with pre-selection of factors. A correlation
matrix was developed among Safety Valuation and the five other potential factors to explore
correlations and possible multicollinearity. Table 2 lists the observed correlations and
significance values between Safety Valuation and Gender, Age, Flight Hours, Safety Training,
and Engagement. For the purpose of analysis, Gender was coded as 1 = Female, 2 = Male.
Correlations Potential Factors and Safety Valuation
1. Safety Valuation
4. Flight Hours
5. Safety Training
*p < .05
With p = .027, engagement was the only potential predictive factor that exhibited
significance at the .05 level. In addition, there were no observed correlations between any of the
potential factors at the .05 significance level, although Age and Engagement had a -.394
correlation with a significance level of .057.
Because of the lack of correlation between Safety Valuation and Age, Gender, Flight
Hours, and Safety Training, only Engagement was selected for a regression model. Therefore, a
regression were met. Chiu et al.: Predictive Model for Australian Flight Students' Safety Valuation
simple linear regression was developed to predict Safety Valuation based on Engagement. A
residuals plot for Engagement is shown in Figure 2, indicating the assumptions of linear
Figure 2. Residuals plot for Safety Valuation with Engagement as a predictor
Continuing the single-predictor regression model, Figure 3 shows the linear relationship between
Engagement and Safety Valuation and the subsequent best-fit line.
Published by Scholarly Commons, 2019
A significant regression equation was found (F(1,22) = 5.594, p < .027) with an
of .203. The participants? predicted Safety Valuation was determined to be 3.014 + .214
(Engagement value) with both variables being measured on a 1-5 scale. The effect size
calculated to be .254, corresponding to a medium to large effect size.
The first research question in this study sought to determine whether a significant
difference existed between the Safety Valuation of the aviation group (flight students and recent
pilot graduates) and local individuals. The study found that the Safety Valuation of Australian
flight students/recently graduated pilots was not significantly different from local Australian
individuals. A difference could not be shown despite this group received specific safety training
and operate in a known high-risk, high-consequence field.
Furthermore, the study (RQ2) tried to develop a predictive model for Safety Valuation
based on research-guided potential factors. This study could not identify Age, Flight Hours,
Gender, or Safety Training as predictors of the Safety Valuation of Australian flight
students/recently trained pilots. As such, a single-variable linear regression using Engagement
exhibits little to no predictive quality with regard to determining an individual?s valuation of
safety. However, as a single variable, Engagement was found to account for 20.3% of the
change in Safety Valuation. These conclusions are further discussed in the following section.
There are several limitations of this study that are relevant to note. First, there are several
methodological weaknesses that should be considered. Primarily, a larger sample size may have
helped clarify some relationships between Safety Valuation and the possible predictive factors.
While this is unlikely the case for Flight Hours and Safety Training (which had near-zero
correlations to Safety Valuation), the additional factors of Age and Gender may significantly
explain a portion of Safety Valuation with a larger sample. Along these lines, a greater sample
size could help refine the portion of Safety Valuation explained by Engagement. Due to these
limitations, this study should not be generalized to a greater population, especially in
consideration of the cultural aspects discussed later in this section.
Another limitation of this study was the use and definition of Safety Valuation. For the
purpose of this study, Safety Valuation was determined by the average surveyed strength of three
safety culture sub-cultures (just culture, reporting culture, and learning culture). Although this
study does not assume an inherent connection between Safety Valuation and safety culture (i.e. a
group of individuals with a high safety valuation does not imply a positive safety culture), it does
frame Safety Valuation in terms of relevant safety culture concepts applied on an individual
level. Of course, further study is necessary to determine the connection (if any) between an
individual?s valuation of safety and a group?s safety culture, especially considering the
introduction of complex organizational processes and structures as components of a safety
Lastly, this study was conducted in the context of the Australian flight training
environment culture, and respondents self-identified as being representative of Australian
culture. As to whether or not the sample genuinely reflected Australian culture is a greater
methodological challenge that was beyond the scope of this study. Nonetheless, culture likely
played a major role in the uniqueness of the results of this study, and this is further discussed
later in this section. Recruitment of participants contributed to this limitation. Distribution of the
survey to the pilot group via social media could have potentially excluded pilots who are not
active on social media or were not in networks associated with the researchers. Furthermore, the
in-person recruitment in a popular public space may have failed to identify certain segments of
the local population.
The Importance of Engagement
One notable remark from the results of this study is the significance of Engagement as a
predictor of safety valuation. Of the five possible predictors (Age, Flight Hours, Safety Training,
Gender, and Engagement) of Safety Valuation, only Engagement was found to have a significant
relationship. Much further study is necessary to understand the impact of what is likely a major
predictor of an individual?s safety valuation. Research topics, such as the genesis of an
individual?s engagement, what influences engagement, and the relationship between collective
engagement and safety culture, are just three examples of directions for future research to better
understand engagement and its implications to ensuring aviation safety.
The Value of Safety Training
This study found no significant relationship between the quality of safety training and
safety valuation. Although this is a somewhat surprising result it is by no means conclusive with
regard to an individual?s valuation of safety. Even in the case that safety training genuinely does
not explain one?s valuation of safety, it likely still plays a critical role in the development of
practical safety behaviors and the development of safety culture. Certainly, valuation of safety is
not a prerequisite for safety compliance and safety action, and even in the absence of an effect on
an individual?s valuation of safety, it can provide information on procedures that impact
behavior. The quality, quantity, frequency, and retention of safety training need to be the subject
of intense further research, to help determine the effect it has on recipients and overall benefit to
This study was conducted in the context of Australian culture. Replication of this study
in other cultural environments is strongly recommended. Results of this study could not
demonstrate a significant difference in valuation of safety between local Australian individuals
and those who had aviation training, including safety training. However, in another culture, one
with perhaps a lower cultural level of safety valuation, effects of safety training, flight
experience, and age might be demonstrated to be significant. In this regard, it is possible that a
high level of cultural safety valuation may be a double-edged sword; a high level of inherent
safety valuation is desirable but perhaps little can be done to increase it further, whereas in
another cultural environment, the base cultural safety valuation might be low but could be
actively constructed. The significance of cultural considerations in research ? especially in a
global industry ? cannot be overstated. Much additional research is necessary to determine how
culture impacts valuation of safety, the development of a safety culture, and safety assurance of
Adjekum, D. K. (2014). Safety culture perceptions in a collegiate aviation program: A systemic
assessment. Journal of Aviation Technology and Engineering, 3(2), 44-56.
Bailey, C. (1997). Managerial factors related to safety program effectiveness: An update on the
Minnesota Perception Survey. Professional Safety, 8, 33-35.
Bos, P., & Lu, C-t. (2007, August/September). Safety management systems: A primer. AAAE
Airport Magazine, 44-48.
Chiang, H-Y., Hsiao, Y-C., Lin, S-Y., & Lee, H-F. (2011). Incident reporting culture: Scale
development with validation and reliability and assessment of hospital nurses in Taiwan.
International Journal for Quality in Health Care, 23(4), 429-436.
Cooper, M. D. (2000). Towards a model of safety culture. Safety Science, 36, 111-136.
Dutcher, J. (2001). Attitudes toward flight safety at a regional gliding school (Atlantic). The
University of Newcastle, Australia.
Dutcher, J., Carrick, K., & Smith, S. (2003). Is your HPIAM programme really effective? The
University of Newcastle, Australia.
Ek, A., Akselsson, R., Arvidsson, M., & Johansson, C. R. (2007). Safety culture in Swedish air
traffic control. Safety Science, 45(7), 791-811. https://doi.org/10.1016/j.ssci.2006.08.017
Federal Aviation Administration (FAA). (2007). Introduction to Safety Management Systems for
Airport Operators (FAA Advisory Circular AC 150/5200-37A). Washington, DC: U.S.
Government Printing Office.
analysis of a safety culture survey. Journal of Safety Research, 45, 15-28.
Freiwald, D., Lenz-Anderson, C., & Baker, E. (2013). Assessing safety culture within a flight
training organization. Journal of Aviation/Aerospace Education & Research, 22(2),
Guldenmund, F. W. (2007). The use of questionnaires in safety culture research ? an evaluation.
International Civil Aviation Organization (ICAO). (2006). Safety Management Manual (SMM)
(1st ed.). (Doc. 9859 AN/474). Montreal, Canada.
(3rd ed.). (Doc. 9859 AN/474). Montreal, Canada.
International Civil Aviation Organization (ICAO). (2013a). Safety Management Manual (SMM)
International Civil Aviation Organization (ICAO). (2013b). International Standards and
Recommended Practices. The Convention on International Civil Aviation (Annex 19).
Kennedy, R., & Kirwan, B. (1998). Development of a hazard and operability-based method for
identifying safety management vulnerabilities in high risk systems. Safety Science, 30(3),
Leib, S., & Lu, C-t. (2013). A gap analysis of airport safety using ICAO SMS perspectives: A
field study of Taiwan. Journal of Aviation Technology and Engineering, 2(2), 63-70.
Leung, A. K.-Y., & Cohen, D. (2011). Within- and between-culture variation: Individual
differences and the cultural logics of honor, face, and dignity cultures. Journal of
Personality and Social Psychology, 100(3), 507-526. https://doi.org/10.1037/a0022151
Liao, M.-Y. (2015). Safety culture in commercial aviation: Differences in perspective between
Chinese and Western pilots. Safety Science, 79, 193-205.
Liu, C., Liu, W., Wang, Y., Zhang, Z., & Wang, P. (2013). Patient safety culture in China: A
case study in an outpatient setting in Beijing. BMJ Quality & Safety, 23(7), 556-564.
Piers, M., Montijin, C., & Balk, A. (2009, March 6). Safety culture framework for the ECAST
SMS-WG. Presented at the meeting of the European Commercial Aviation Safety Team
of the European Strategic Safety Initiative (ESSI), Cologne, Germany.
Rasmussen, J. (1997). Risk management in a dynamic society: A modelling problem. Safety
Science, 27, 183-213.
Reason, J. (1997). Managing the risks of organizational accidents. Aldershot, UK: Ashgate.
Reason, J. (2000). Safety paradoxes and safety culture. Injury Control & Safety Promotion, 7(1),
Reichers, A. E., & Schneider, B. (1990). Climate and culture: An evolution of constructs. In B.
Schneider (Ed.), Organizational climate and culture (pp. 5-39). San Francisco, CA:
Sawacha, E., Naoum, S., & Fong, D. (1999). Factors affecting safety performance on
construction sites. International Journal of Project Management, 17(5), 309-315.
Section 1: Basic Demographic Information (given only to the aviation participant group)
Section 2: Safety Valuation Questions (five-point Likert scale; adapted from Liao (2015); given
to both participant groups in random order)
1. I understand which behaviours are acceptable and unacceptable with respect to safety.
2. I believe I would be treated fairly if my actions were unsafe.
3. I think it is important for people to learn that their actions have consequences and that
they will be handled fairly.
4. I believe that behaviours are unsafe, not people themselves.
5. My relationships with people do not affect what safety issues I disclose (or would
6. I would be willing to disclose my own violations of safety regardless of if my behavior
had been observed.
7. I would not hesitate to report someone?s safety violations, even if believe they are more
important or powerful than me.
8. I would report and unsafe action that I observed, even if I knew or was close with the
person doing it.
9. I feel that reporting safety issues contributes to a safer environment for everyone.
10. I would report my own rule violations even if no one has noticed them.
11. I pay attention to safety information when I encounter it.
12. It is important to me that safety information is provided by companies/venues that I visit.
13. I try to find ways to enhance my safety knowledge on my own time.
14. If I have a safety concern, I am comfortable trying to resolve it.
15. I am willing to learn safety lessons from others.
Section 3: Engagement
(adapted from Thorp et al. ; given only to the aviation participant
group in random order)
1. My instructor or someone at my flight training organization, cares about me as a person.
2. My flight training organization has employees that encourage my development.
3. My fellow students are committed to doing quality work.
4. In the past year, I feel I have had opportunities in flight training to learn and grow.
5. I have specific ?school friends? that I know through flight training.
Section 4: Safety Training Topics (adapted from ICAO [2013a]; using an adapted four-point
Likert scale including not covered, mentioned, somewhat covered, and covered in depth; given
only to the aviation participant group in random order)
Frazier , C. B. , Ludwig , T. D., Whitaker , B. G. , & Roberts , D. S. ( 2013 ). A hierarchical factor
Glendon , A. I. , & Stanton , N. A. ( 2000 ). Perspectives on safety culture . Safety Science , 34 ( 1- 3 ),
Schein , E. H. ( 2004 ). Organizational culture and leadership (3rd ed .). San Francisco, CA: Jossey-Bass.
Simon , S. I. , & Cistaro , P. A. ( 2009 ). Transforming safety culture: Grassroots led/management supported change at a major utility . Professional Safety , 28 - 35 .
Siu , O.-l. , Phillips , D. R. , & Leung , T.-w. ( 2003 ). Age differences in safety attitudes and safety performance in Hong Kong construction workers . Journal of Safety Research , 34 ( 2 ), 199 - 205 . https://doi.org/10.1016/s0022- 4375 ( 02 ) 00072 - 5
Stolzer , A. J. , Halford , C. D. , & Goglia , J. J. ( 2008 ). Safety management systems in aviation . Burlington, VT: Ashgate.
von Thaden , T. L. , & Gibbons , A. M. ( 2008 , July) . The safety culture indicator scale measurement system (SCISMS) . DOT/FAA/AR-01-G/015, Office of Aviation Research and Development, U.S. Department of Transportation, Federal Aviation Administration, Washington, DC.
Thorp , J. , Baqai , W. , Witters , D. , Harter , J. , Agrawal , S. , Kanitkar , K. , & Pappas , J. ( 2012 ). Workplace engagement and workers' compensation claims as predictors for patient safety culture . Journal of Patient Safety , 8 ( 4 ), 194 - 201 . https://doi.org/10.1097/pts.0b013e3182699942
Uttal , B. ( 1983 , October 17). The corporate culture vultures . Fortune Magazine.
Weick , K. E. ( 1987 , Winter). Organizational culture as a source of high reliability . California Management Review , 29 ( 2 ), 112 - 127 . https://doi.org/10.2307/41165243
Yavuz , N. , & Welch , E. W. ( 2010 ). Addressing fear of crime in public space: Gender differences in reaction to safety measures in train transit . Urban Studies , 47 ( 12 ), 2491 - 2515 . https://doi.org/10.1177/0042098009359033
1. Organisational Safety policies
2. Organisation Goals and Objectives
3. Safety Risk Management Principles
4. Hazard Reporting Systems
5. Safety Evaluation and Auditing
6. Lines of Communication for Disseminating Safety Information