115
1 INTRODUCTION
1.1 Background to the study
Serious incidents can come with catastrophic
consequences for staff and passenger onboard ships.
Maritime personnel are naturally the group most
affected by the safety aspect at sea. On the one hand,
they are constantly exposed to potential shipping
related risks and on the other hand, they are the
group that continuously have to carry out repetitive
safety training activities.
All in all, it can be said that a tireless focus on
maritime safety work is necessary. In order to
counteract the possibility of drowsiness in safety work
and training activities it is important to engage
research in the field of maritime safety training.
The commitment to safe and sustainable
operations and environments in working life faces a
growing need for new perspectives, not least when it
comes to shipping. Traditionally, research and
development of safety and quality management
systems has mainly focused on analyzes of adverse
events and accidents. With this perspective, two
problems occur. Firstly, the perspective encourages
the choice to mainly react to accident- and incident
statistics, which is an approach best suited in
businesses with many recurring negative events. This
means that after the negative events decrease, the risk
indicators also decrease, which tends to lead to a false
sense of safety and a reduced commitment to safety
work (Hollnagel, 2010: 281-283). This focus has also
led to a pursuit of the human factor, while the system
in which an incident has occurred may remain
unchanged.
Currently, most industries, especially high-risk
industries such as shipping, are facing growing
complexity, while at the same time increased
Perceived Quality of Safety Training Onboard Ship –
The Swedish Case
C. Hult
Kalmar Maritime Academy, Linnaeus University,
Kalmar, Sweden
ABSTRACT: This is a statistic study on the perceived quality of safety training. The study is based on
respondents from the Swedish register of seafarers. The results show that the experience of safety training is
generally positive among the seafarers. The level of positivity is influenced by personal circumstances (such as
onboard position and age) and by external factors in the work environment (such as ship type, trade area and
flag state). However, the effect of onboard position diminishes when age and external factors are controlled for.
In this situation, only the positive effects for cabin crew and age remain with strength and significance. Most
ship types show strong positive effect, except for road ferries and vessels for max 12 passengers. When it comes
to trade area it is only sheltered trade that shows a significant positive effect. Moreover, a homeland flag in the
stern leads to a significant increase in the perceived quality of the safety training. These findings are discussed
in the concluding part of the paper.
http://www.transnav.eu
the
International Journal
on Marine Navigation
and Safet
y of Sea Transportation
Volume 17
Number 1
March 2023
DOI: 10.12716/1001.17.01.
11
116
demands for efficiency, quality, sustainable
operations and working environment must be met.
All of this is happening in the face of an unpredictable
future of increased focus on digitization and
automation in shipping activities. Given this
background it is necessary to study the safety work at
sea from all different angles.
1.2 The objectives of this study
The purpose of the study presented in this paper is to
investigate Swedish seafarer’s experiences, or
perceptions, of safety training onboard ships.
Specifically, the objectives of the study are to
investigate (i) whether perceived quality of onboard
training differs due to age among the Swedish
seafarers, (ii) whether the perceived quality differs
due to trade area, (iii) whether the perceived quality
differs due to vessel type (iv), whether the perceived
quality differs due to flag state, (v) and whether the
perceived quality of onboard training differs by
gender.
2 THEORETICAL BASIS FOR THE STUDY
2.1 Earlier research on attitudes to work
Quantitative research on seafarers’ attitudes to work
is not abounded. There are, however, a number of
studies from different parts of the world (e.g., Guo et
al. 2005, Guo et al. 2010, Pan et al. 2011, Sencila et al.
2010, Turker & Er 2007). A particular research interest
can be seen within the cruise sector, probably due to
the link between employee job situation and customer
satisfaction in service occupations (e.g., Larsen et al.
2012, Testa 2001, Testa & Mueller 2009, Testa et al.
2003). Moreover, there are some attitude studies on
Swedish seafarers with quantitative approaches,
where some focusing on job satisfaction onboard
merchant ships (Olofsson 1995, Werthén 1976), and
several recent studies that focus on commitment to
work and occupation (Hult 2012, Hult & Snöberg
2013, Hult & Österman 2015; Österman & Hult 2016;
Sandberg et al. 2020).
There is one recent study with explicit focus on
safety training for seafarers (Viktorelius et al. 2023).
Based on interviews and observations, this study
shows that there are shortcomings in how simulations
are carried out and evaluated. The shortcomings have
a lot to do with shortage of time combined with
increased demands which make it difficult to engage
in safety training. It emerged that the exercises are
often carried out on rule-based incentives rather than
with focus on organizational learning and knowledge
needs. Even the generally advocated structure of
preparatory briefing and final debriefing proved to be
deficient in the cases studied.
2.2 This study on safety training onboard ships
The results from the final study presented in previous
section, concerning safety training for seafarers, calls
for further studies. Therefore, this study will focus on
seafarers’ attitudes towards safety training on board
ships, in where we may find some answers related to
the shortcomings described above. Thus, the aim for
this study is to investigate the perceived quality of
safety training among Swedish seafarers using a
quantitative approach. The details of this approach
will be outlined in the following section.
3 METHOD
3.1 The method and data
The method required from the presented objective is
quantitative, based on a targeted survey. The survey
construction began in January 2022 at the same time
as a dialogue with the Swedish Seafarers Register
concerning access to postal addresses for registered
seafarers was initiated. After conducting an ethics
review according to the data protection regulation, a
large file with 12 022 e-mail addresses was obtained.
The survey was sent out to all addresses on March 24,
2022.
The survey was closed on April 12, 2022 after two
reminder mailings. The number of respondents turn
out to be 2 365 individuals, which gave a response
rate of 19.67%. Although the number of respondents
is unusually large, the percentage response rate was
relatively low. Therefore, we must ask ourselves to
what extent a survey response from 2 365 individuals
can be considered as representative of all 12 022
individuals in the Swedish Seafarers Register.
According to a traditional non-response analysis,
we should assume that the actual respondents are
likely to have a greater interest in- and perhaps higher
ambitions with the safety training compared to those
who didn’t respond. In our case, a non-response
analysis is problematic also for other reasons. There
were incoming remarks from several receivers of our
survey who did not know they were in the register.
Some claimed that they did not even consider
themselves as belonging to the occupational category.
Nevertheless, the following analysis rests on the
unusually high number of 2 365 actual seafarers.
3.2 Processing of data and analysis
The Statistical Package Social Science (SPSS) were
used throughout the analysis. The dependent variable
of perceived quality of safety training – from now
referred to as ‘perceived safety training quality’ or
‘PSTQ’. The variable was constructed as an index
using internal reliability control. The dependent
variable was based six attitude questions on which
respondents were asked to take a position by selecting
a fixed option on a five-point Likert Scale, from
strongly agree to strongly disagree. A completed
factor analysis indicated that the six aspects form a
uniform component with very high reliability (0.906)
where the highest possible value is 1.0 and the lowest
allowed 0.7. The appropriate indicators for PSTQ
turned out as shown in Table 1.
The indicators were then recoded so that 0 denotes
the option that entails the lowest PSTQ and 4 the
highest. The indicators were then summarized in the
index of PSTQ. To facilitate interpretation of the
117
results, the index was divided by its maximum value
and multiplied by 100. The PSTQ-index is thus
permitted to vary between 0 and 100, where 100 is the
highest perceived training quality possible.
Table 1. Indicators of perceived training quality for seafarers
________________________________________________
To what extent do you agree or disagree with the following
statements when thinking about your current (or recent)
work at sea?
________________________________________________
I experience the safety training onboard as well thought out.
I experience the safety training onboard as varied.
I find the safety training onboard stimulating.
I find the safety training onboard interesting.
I experience the safety training as realistic.
The safety training makes me feel safe in my work at sea.
________________________________________________
Table 2 presents the mean value and standard
deviation for the index. The high mean value and the
low standard deviation indicates that seafarers are
quite united in their perceived quality of safety
training. Cronbach’s Alpha is a test of the internal
correlation among the indicators in each index – the
higher the value (between 0 and 1), the more reliable
the index. Table 2 shows that the index turn out
stable.
Table 2. The Perceived Training Quality – Swedish seafarers
in 2022
________________________________________________
Mean value (0-100) 67.98
Standard deviation 18,57
Cronbach’s Alpha 0.644
________________________________________________
4 RESULTS
4.1 The perceived quality of safety training at sea
In the following, it will be shown how the perceived
training quality (PSTQ) relates to age, gender, trade
area, ship type, flag and position. At the very end, a
multivariate analysis is displayed where the variables
precented below that show initial significance will be
included. Significance levels and numbers of
respondents for each category are given below the
tables.
Table 3, display eventual gender differences when
it comes to perceived quality of the safety training.
Here it becomes obvious that there are basically no
differences at all when it comes to gender. The gender
variable will therefore not be used in the final
multivariate analysis.
Table 3. Gender and the perceived training quality
________________________________________________
B coefficient Significance
________________________________________________
Constant man 68,059 0,000
Woman -0,199 0,847
________________________________________________
Significance levels: *** = 0.001 level, ** = 0.01 level, * = 0.05 level.
Number of respondents according to gender: 1386 men and 431
women.
Table 4 shows that the effect of age comes with the
highest level of significance. Age displays a positive
linear relationship, with + 0.175 per year on PSTQ.
Table 4. Age and the perceived training quality
________________________________________________
B coefficient Significance
________________________________________________
Constant youngest 60,196 0,000
Age (effect per year) 0,175*** 0,000
________________________________________________
Significance levels: *** = 0.001 level, ** = 0.01 level, * = 0.05 level.
Number of respondents 1812.
In Table 5 it appears that sea personnel in near
costal trade and European trade express a
significantly lower PSTQ than personnel in sheltered
trade.
Table 5. Trade area and perceived training quality
________________________________________________
B coefficient Significance
________________________________________________
Constant Sheltered trade 70,326 0,000
Near Costal trade -3,002** 0,007
European trade -4,004** 0,002
Ocean trade -1,304 0,473
________________________________________________
Significance levels: *** = 0.001 level, ** = 0.01 level, * = 0.05 level.
Number of respondents for respective trade area: Sheltered trade
429, Near Costal trade 847, European trade 404, Ocean trade 139.
It appears in Table 6 that personnel from all ship
types listed below express lower PSTQ than the
constant, whish in this case are other ship types. The
category ‘Other types of ships’ refers to those who did
not specify a ship type in the questionnaire.
Table 6. Vessel type and perceived training quality
________________________________________________
B coefficient Significance
________________________________________________
Constant Other vessels 71,245 0,000
Tankers -4,485* 0,012
Dry cargo -10,639** 0,002
Ro-Ro -2,431 0,220
Passenger ship -2,652* 0,046
Ro-Pax -4,566** 0,003
Offshore -6,713* 0,031
Road ferry -7,642* 0,023
Max12pass -5,923 0,092
________________________________________________
Significance levels: *** = 0.001 level, ** = 0.01 level, * = 0.05 level.
Number of respondents for respective ship type: Tankers 181, Dry
cargo 33, Ro-Ro 131, Passenger ship 756, Ro-Pax 325, Offshore 41,
Road ferry 35, Max12pass 32, Other vessels 268.
Table 7 shows that personnel who sail under the
Swedish flag express a significantly higher perceived
safety training quality compared to those who sail
under another flag. The difference is clearly
significant, with which the flag variable can be used in
the final multivariate analysis.
Table 7. Flag and perceived training quality
________________________________________________
B coefficient Significance
________________________________________________
Constant foreign flag 64,889 0,000
Swedish flag 3,503** 0,010
________________________________________________
Significance levels: *** = 0.001 level, ** = 0.01 level, * = 0.05 level.
Number of respondents according to Flag: Swedish flag 1610, Other
flag 213.
In Table 8, a multivariate analysis of the variables
age, trade aria, vessel type, flag and onboard position
is presented in two steps against the dependent
variable, the perceived quality of the safety training.
In Step I, the effect of position is shown. Here it
appears that commanders and cabin attendants report
significantly higher perceived safety training quality
com-pared to security personnel (who are in the
constant). It can be added that also chief engineer,
restaurant personal and hotel administration are very
close to the lowest significance level (all marked in
bold). It can also be noted that all positions are above
the constant in perceived training quality.
In Step II, Table 8, it is first shown what happens to
the values for different positions when all other
explanatory variables are entered. All the values for
118
the positions are changing when the effect of age,
trade area, vessel type and flag are brought into the
equation. The positive effect of being a commander
loses its previous significance, even if the coefficient
remains high. In terms of position, it is now only the
positive effect of being a cabin attendant that remain
significant concerning the perceived safety training
quality (PSTQ) after controlling for the additional
variables in the model.
Moving down to the added explanatory variables,
we can note that increasing age still has a significant
positive and linear effect on PSTQ. It also appears that
it is only sheltered trade that shows a significant
positive effect compared to near costal trade (which is
in the constant). Regarding ship types, only the effects
for offshore and road ferry are without significance.
All ship types show a positive effect except ships with
a maximum of 12 passengers which display a
coefficient significantly below the constant (which are
dry cargo ships).
At the bottom of the variable list, it appears that
the Swedish flag still has a positive and significant
effect on the perceived safety training quality. Below
the variable list the R-Square is presented. The R-
Square can be understood as the model's explanatory
value, where a value of 1 would mean that all
variation in the material is fully explained (which in
most cases would be impossible). When it comes to
analysis of survey responses, one can be completely
satisfied that the explanatory value has increased
from .019 in Stage I to .049 in Stage II.
Table 8. Final multivariate regression on perceived training
quality
________________________________________________
Step I Step II
________________________________________________
Coefficient Sig Coefficient Sig
________________________________________________
Constant 59,821 ,000 42,114 ,000
Commander 10,910* ,030 9,217 ,066
Chief Mate 5,340 ,306 5,342 ,304
Second Mate 5,443 ,299 5,743 ,272
Chief Engineer 9,449 ,069 7,424 ,153
First Engineer 1,045 ,851 1,248 ,822
Engineer 5,501 ,314 5,277 ,330
Crew deck 7,808 ,123 6,763 ,178
Crew machine 6,534 ,213 6,381 ,219
Restaurant pers 8,651 ,086 8,771 ,078
Cabin attendant 13,095** ,013 12,702* ,015
Hotel admin 10,045 ,068 10,646 ,051
Other positions 7,054 ,176 6,929 ,180
Age - - ,143*** ,000
Sheltered trade - - 3,301* ,012
European trade - - ,177 ,896
Ocean trade - - 3,676 ,062
Tankers - - 8,143* ,020
Ro-Ro - - 8,319* ,021
Passenger ship - - 7,036* ,037
Ro-Pax - - 8,031* ,020
Offshore - - 7,984 ,070
Road ferry - - ,452 ,922
Max12pass - - -7,683* ,028
Other vessels - - 11,230*** ,001
Swedish flag - - 3,539* ,023
R-Square ,019 ,049
________________________________________________
Levels of significance: *** = 0.001-level ** = 0.01-level, * = 0.05-level.
Number of respondents for respective position: Commander 331,
Chief Mate 119, Second Mate 108, Chief Engineer 130, First
Engineer 51, Engineer 62, Crew deck 270, Crew machine 112,
Restaurant 334, Cabin attendant 95, Hotel admin 58, Other positions
122. Constant in Step I: Security. In Step 2: Security, Lowest age,
Near Costal trade, Dry Cargo ship, Flag other than Swedish.
4.2 Conclusion, future work and recommendations
The overall conclusion of the above analysis is that the
experience of the quality of safety training is
influenced partly by personal circumstances such as
position and age, and partly by external factors in the
work environment. But it is also clear that the effect of
the personal circumstances fades when the external
factors are entered. In that situation, it is only the
positive effects for cabin crew and age that remains
with both strength and significance.
Moving to the external factors and the trade area
category, it is only sheltered trade that show a
significant positive effect on the perceived training
quality. A possible explanation for this is that the
safety training is perceived to be easier and more
manageable on smaller ships with a smaller crew
where everyone knows each other well.
Finally, it may be added that Table 8 in Step II can
be used to predict the positivity in the experience of
safety training for different types of cases. Below, such
a prediction is exemplified with the case of a 60-year-
old cabin attendant onboard an ocean-going
passenger ship under the Swedish flag:
Constant: 42,114
+ 60 years old (0,143*60): 8,58
+ Cabin attendant: 12,702
+ Passenger ship: 7,036
+ Ocean trade: 3,676
+ Swedish flag: 3,539
__________________________________
= 77,647
Thus, the positive experience of safety training for
a 60-year-old cabin attendant onboard an ocean-going
passenger ship under the Swedish flag should
therefore, statistically calculated, be 77.647 on a scale
of 0–100. Then, of course, it is not certain that this type
of case exists in reality.
ACKNOWLEDGEMENT
I would like to express my gratitude towards the financial
support from the Swedish Mercantile Marine Foundation. I
would also like to thank Martin Viktorelius for help with
the initial survey design and collection of data. Moreover, I
like to thank the 2 365 respondents to our survey. Without
their input this study would not have existed.
REFERENCES
Guo, Jiunn-Liang, Gin-Shuh, Liang, & Ye, Kung-Don. (2005).
Impact of Seafaring Diversity: Taiwanese Ship-officers'
Perception. Journal of the Eastern Asia Society for
Transportation Studies, 6, 4176-4191.
Guo, Jiunn-Liang, Ting, Shih-Chan, & Lirn, Taih-Cherng.
(2010). The Impacts of Internship at Sea on Navigation
Students. Maritime Quarterly, 19(4), 77-107.
Hollnagel, E. (2010). Epilogue: RAG - The Resilience
Analysis Grid. In J. Pariès, J. et al. (eds) Resilience
Engineering in Practice: A Guidebook. Ashgate
Publishing Ltd, UK. 275-296.
Hult, Carl. (2012). Seafarers’ commitment to work and
occupation. In C. Hult (Ed.). Swedish Seafarers and the
Seafaring Occupation 2010 – A study of work-related
attitudes during different stages of life at sea. Kalmar:
Kalmar Maritime Academy, Linnaeus University.
119
Hult, Carl, & Snöberg, Jan. (2013). Wasted Time and Flag
State Worries. Marine Navigation and Safety of Sea
Transportation, 121-128: CRC Press.
Hult, Carl, & Österman, Cecilia (2015). Swedish Seafarers’
Occupational Commitment in Light of Gender and
Family Situation. Safety of Marine Transport: Marine
Navigation and Safety of Sea Transportation. 41-47.
Larsen, Svein, Marnburg, Einar, & Øgaard, Torvald. (2012).
Working onboard – Job perception, organizational
commitment and job satisfaction in the cruise sector.
Tourism Management, 33(3), 592-597.
Olofsson, Martin. (1995). The work situation for seamen on
merchant ships in a Swedish environment: Chalmers
University of Technology.
Pan, Yu-Huei, Lin, Chi-Chang, & Yang, Chung-Shang.
(2011). The Effects of Perceived Organizational Support,
Job Satisfaction and Organizational Commitment on Job
Performance in Bulk Shipping. Maritime Quarterly,
20(4), 83-110.
Sandberg, C., Hult, C., Österman, C., Praetorius, G. (2020).
The Committed Service Crew : The Impact of Passenger
Proximity on Organizational Commitment and Job
Satisfaction. TransNav, International Journal on Marine
Navigation and Safety of Sea Transportation. 14 (3). 595-
600.
Sencila, V, Bartuseviciene, I, Rupšiene, L, & Kalvaitiene, G.
(2010). The Economical Emigration Aspect of East and
Central European Seafarers: Motivation for Employment
in Foreign Fleet. TransNav, International Journal on
Marine Navigation and Safety of Sea Transportation,
4(3), 337-342.
Testa, Mark R. (2001). Organizational commitment, job
satisfaction, and effort in the service environment. The
Journal of Psychology, 135(2), 226-236.
Testa, Mark R, & Mueller, Stephen L. (2009). Demographic
and cultural predictors of international service worker
job satisfaction. Managing Service Quality, 19(2), 195-
210.
Testa, Mark R, Mueller, Stephen L, & Thomas, Anisya S.
(2003). Cultural fit and job satisfaction in a global service
environment. MIR: Management International Review,
129-148.
Turker, F, & Er, ID. (2007). Investigation the root causes of
seafarers’ turnover and its impact on the safe operation
of the ship. International Journal on Marine Navigation
and Safety of Sea Transportation, 1(4), 435-440.
Viktorelius, M, Sellberg C, Hult, C, & Praetorius, G. (2023)
Säkerhetsträning av Fartygsbesättningar för Resilience:
Kalmar Maritime Academy, Linnaeus University..
Werthén, Hans-Erik. (1976). Sjömannen och hans yrke: en
socialpsykologisk undersökning av trivsel och
arbetsförhållanden på svenska handelsfartyg. (Doctoral
thesis), University of Gothenburg.
Österman, C., Hult, C. (2016). Administrative burdens and
over-exertion in Swedish short sea shipping. Maritime
Policy & Management. 43 (5). 569-579.
