Evaluating the Attitudes and Trust of Sri Lankan Air Traffic Controllers towards Accepting Automation in terms of Electronic Flight Strips

Abstract

The adoption of Electronic Flight Strips (EFS) is a global trend aimed at streamlining the routine tasks for Air Traffic Controllers (ATCs), yet Sri Lanka continues to use traditional paper-based systems. This study investigates the potential acceptance of EFS automation by Sri Lankan ATCs, examining factors such as perceived usefulness, perceived ease of use, trust, and attitude toward automation within the Technology Acceptance Model framework. Data from 48 Sri Lankan ATCs revealed that all factors were positively correlated with automation acceptance, with perceived ease of use emerging as the most influential construct. A classification by work unit showed Approach Controllers perceived EFS as having lower usefulness compared to Tower and Area Controllers, likely due to their unique workflow. Additionally, ATCs with prior automation experience have demonstrated stronger positive attitudes, trust, and willingness to adopt EFS, emphasizing the role of experience in fostering automation acceptance. Furthermore, age and gender showed no significant impact on acceptance levels. These findings provide critical insights for EFS system designers and management to tailor training and implementation strategies, highlighting the importance of designing intuitive interfaces, building trust in safety, and leveraging experienced ATCs to champion adoption.

Full text

Suggested Citaon: Anupama Sudehani, B.L.G (2024). Evaluang the Atudes and Trust of Sri Lankan Air Trac Controllers
towards Accepng Automaon in terms of Electronic Flight Strips. University of Colombo Review (New Series III), 5(1), 26-
58.
© 2024 The Authors. This work is licenced under a Creave Commons Aribuon 4.0 Internaonal Licence which permits
unrestricted use, distribuon, and reproducon in any medium, provided the original work is properly cited.
Corresponding author: sudehani84@gmail.com
https://orcid.org/0009-0007-4340-6317
University of Colombo Review (Series III),
Vol.5, No.2, 2024
Evaluang the Atudes and Trust of Sri Lankan Air Trac Controllers towards
Accepng Automaon in terms of Electronic Flight Strips
B.L.G. Anupama Sudehani
Independent Researcher
ABSTRACT
The adopon of Electronic Flight Strips (EFS) is a global trend aimed at streamlining
the roune tasks for Air Trac Controllers (ATCs), yet Sri Lanka connues to use
tradional paper-based systems. This study invesgates the potenal acceptance
of EFS automaon by Sri Lankan ATCs, examining factors such as perceived useful-
ness, perceived ease of use, trust, and atude toward automaon within the Tech-
nology Acceptance Model framework. Data from 48 Sri Lankan ATCs revealed that
all factors were posively correlated with automaon acceptance, with perceived
ease of use emerging as the most inuenal construct. A classicaon by work unit
showed Approach Controllers perceived EFS as having lower usefulness compared
to Tower and Area Controllers, likely due to their unique workow. Addionally,
ATCs with prior automaon experience have demonstrated stronger posive at-
tudes, trust, and willingness to adopt EFS, emphasizing the role of experience in
fostering automaon acceptance. Furthermore, age and gender showed no signif-
icant impact on acceptance levels. These ndings provide crical insights for EFS
system designers and management to tailor training and implementaon strate-
gies, highlighng the importance of designing intuive interfaces, building trust in
safety, and leveraging experienced ATCs to champion adopon.
KEYWORDS:
Air Trac Control, Atudes, Automaon Acceptance, Electronic Flight Strips,
Perceived Ease of Use, Perceived Usefulness, Trust
DOI: https://doi.org/10.4038/ucr.v5i2.187

B.L.G. Anupama Sudehani 27
Introducon
Electronic Flight Strips (EFS) is a system that substitutes routine tasks of
loading, writing, and placing or sometimes data entering and printing of ight strips,
tasks that were executed manually by ATCs and demanded a considerable portion
of their time and eort from the main task of trac controlling. However, more
experienced controllers are versatile enough to multitask these functions eectively.
They usually spend a considerable amount of time in developing prociency in strip
handling techniques. Consequently, the question emerges regarding the feasibility
of introducing automation that supersedes their acquired skills while concurrently
simplifying their daily operations. The adoption of EFS has shown continuous
improvement in recent decades; however, there remains a notable resistance to its
implementation (Mackay and Fayard, 1999). This resistance is partly attributed to
the attitudes of conservative Air Trac Controllers (ATCs) who are skeptical of
automation (EUROCONTROL, 2000) and show varying levels of trust in such
systems (Muir and Moray, 1996; Lee and See, 2003). Attempts to introduce new
automation systems without considering user acceptance can cause signicant
challenges, wasting time, eort and money. Previous studies suggest that technology
acceptance is based on individuals’ own attitudes and beliefs (Gattiker, 1988). Hence,
the current study seeks to assess the potential adoption of automated EFS among
ATCs in Sri Lanka, who are familiar with paper strips, by evaluating their attitudes,
beliefs, and trust in emerging technology.
The study is structured as follows: this section presents the background
information, outlines the objective of the study and presents an extensive review
of relevant literature, highlighting previous work in the eld. This is followed by
section 2, which outlines the methodology employed in the study. Section 3 presents
the research ndings, and subsequently, section 4 discusses the limitations of the
outcomes and suggestions for future research.
Background of the Study
Flight progress strips, introduced as paper-based tools in the 1940s (Weihe,
1953), serve as external representations of aircraft under Air Trac Control (ATC).
They display critical ight details such as callsign, altitude, and speed, aiding
controllers in visualizing air trac and coordinating handovers. The steady increase
in air trac gradually compelled ATCs to adopt higher levels of automation in their
operations (Berndtsson and Normark, 1999). However, traditional pen and paper are
incapable of updating computer systems or radar screens when ATC directions are
sent to pilots. Thus, the concept of EFS, which is used to integrate ATC directives into
computer systems, was established (Hurter, 2012). This establishment was further
supported by numerous empirical ndings indicating that the act of writing on strips
and managing the strip board does not signicantly contribute to the primary tasks

28 UCR (Series III), Vol.5, No.2, 2024 Arcle
of air trac controllers, such as conict resolution and radar monitoring, with some
controllers also nding these activities time-consuming (Edwards et al., 1995). In
1992, the United Kingdom initiated a search for alternatives to replace traditional
paper strips with EFS (Hughes et al., 1992). The implementation of EFS in France
took place in the late 1990s (Surakitbanharn et al., 2015). The transition from the
initial concept of replacing paper ight strips to the implementation of digital ight
strips in European ATC facilities spanned a decade and a half. (Huber et al., 2020).
In order to enhance capacity and safety in ATC, digital strip systems have replaced
paper strips in lower airspace control centres across Europe (Huber et al., 2020). The
concept of EFS has evolved to a more advanced stage known as Portable EFS, which
harnesses the advantages of paper ight strips while incorporating the distinctive
features of EFS (Doble and Hansman, 2002).
In reality, some of the most complete ethnographic studies on ATC methods
and culture were conducted more than two decades ago, despite the fact that
Conversy and colleagues (2011) stated that practises still develop. Nevertheless,
the technologies discovered and validated by Western countries are also permeating
Eastern regions with the inuence of globalization. Southeast Asian countries are
also embarking on the modernization of their ATC systems. While these developing
nations are adopting technologies already used in other regions, there is a lack of
culturally specic or need-based guidelines for developing eective training plans
tailored to their unique market requirements (Surakitbanharn et al., 2015). When
deployed in foreign contexts, automated designs and training programmes established
for Western cultures may require adaptation and customisation. For instance, in honor
cultures like Turkey, factors such as distrust and disuse may necessitate additional
measures, like enhanced transparency, to encourage the adoption of automation
technologies. In contrast, dignity cultures like the US, with higher trust levels, may
require less encouragement. (Chien et al., 2020). Insucient or incorrect needs
assessment or attempting to implement an incompatible automation system for its
end users can result in the squandering of signicant nancial resources and eorts.
However, a prerequisite for the implementation of an automated EFS system among
a novel user base would entail their inherent willingness and favourable disposition
towards embracing such technology. The readiness of ATCs is a critical concern when
considering the human factors associated with adaptation of automated systems. This
concern arises from the argument that humans are ultimately responsible for systems
that have become increasingly complex and are no longer fully understandable or
controllable by humans alone (Vallor and Bekey, 2017).
End user attitudes towards accepting various automation products have been
investigated in past literature. One notable example is the development of a scale
designed to measure individual attitudes towards pilotless aircraft (Molesworth
et al., 2016). Attitudinal constructs such as perceived usefulness, trust, perceived

B.L.G. Anupama Sudehani 29
safety and security showed an impact on early adoption of automated buses. Young
males who had prior experience with automated buses showed a higher inclination
towards using them during the initial stages (Rahim et al., 2023). The attitudes of
people towards automated vehicles serve as an indicator of their overall adoption
of technology, particularly among early adopters who perceive it as a positive trend
(Liljamo et al., 2018).
The literature on assessing attitudes of ATCs towards automation acceptance
is minimal worldwide. According to existing literature, the attitudes of ATCs towards
future automation demonstrate a notable degree of selectivity, particularly in their
acceptance of computer assistance (EUROCONTROL, 2000). The automation
acceptance of ATCs is inuenced by their attitudinal constructs such as trust in
automation, job satisfaction and attitudes towards the use induced by usability
and usefulness. Further, less than 8% of variability in automation acceptance can
be accounted for by conventional predictors such as age, trust, and job satisfaction
(Bekier et al., 2011). The acceptance of higher levels of automation decreased
among users after they were exposed to a task that involved automation assistance,
indicating that it is possible to modify automation acceptance. In a study conducted
using two advanced automation tools, utilizing Sheridan and Parasuraman’s (2005)
model of automation levels, it was observed that the willingness of ATCs to adopt the
new futuristic tools declined when the automated technology assumed control of the
decision-making process. Hence, active participation of the operator in the decision-
making process plays a signicant role in fostering the acceptance of automation
(Bekier et al., 2011).
The lack of studies assessing attitudes of ATCs towards accepting automation,
specically in the context of a paper-based ATC system in Sri Lanka, indicates a
gap in the existing literature. Understanding the attitudes of ATCs towards major
technological changes is crucial for successful implementation and adoption of
new systems. Cultural backgrounds can play a signicant role in shaping attitudes
and reactions towards automation, making it essential to investigate these factors
within the specic context of Sri Lanka. Conducting research in this area can provide
valuable insights for the development and implementation of automation systems
in the country. The study aims to evaluate the attitudes of Sri Lankan air trac
controllers towards EFS. As conrmed through personal communication with Sri
Lankan ATCs in April 2023, paper strips were in use at the time across all three units
of air trac control, including Tower control, Approach control, and Area control at
all airports throughout the country (Author, personal communication, April 2023).
However, the prospect of automating ight strips is not too distant and is driven
by two signicant factors. First, the rapid advancement of automation aligns with
the growth of air trac, making it increasingly feasible. Second, the recent paper
shortage in Sri Lanka further necessitates exploring alternative solutions, including

30 UCR (Series III), Vol.5, No.2, 2024 Arcle
the automation of ight strips. The recent economic crisis in the rst quarter of 2022
in Sri Lanka has had a tremendous impact on numerous sectors of the country’s
economy, resulting in a signicant shortage of paper. For instance, in March 2022 Sri
Lankan schools announced an indenite postponement of term exams. This scarcity
is mostly due to a lack of foreign funds to purchase an adequate supply of paper
(George et al., 2022; Tripathi et al., 2022), while concurrently, the decreasing cost of
display technology opens up new possibilities for utilizing display screens in various
innovative ways (Norman, 2013).
Objecve of the Study
The primary objective of this study is to examine the inuence of
psychological factors such as trust, perceived usefulness, perceived ease of use
and attitudes on automation acceptance within the context of EFS among Sri
Lankan ATCs. By examining the interplay between each other, this research aims
to enhance understanding of how these variables collectively shape the acceptance
of automation. Additionally, the study seeks to analyze and compare the eects of
dierent demographic factors on the acceptance of automation among ATCs. The
practical implications of this study are twofold: rstly, it can provide valuable
insights for EFS designers to improve their products and services; and secondly, it
can assist management and implementers in eectively managing the change process
associated with adopting automation in ATC systems.
The objectives of this study are restated as follows:
examine how beliefs, attitudes towards automation and trust in automation
impact the automation acceptance of Sri Lankan ATCs in terms of EFS; and
examine how demographic factors impact attitudes, beliefs, trust and
automation acceptance of Sri Lankan ATCs in terms of EFS.
Research Quesons
Flowing from the above stated objectives, this study aimed to address the
following two questions, focusing on Sri Lankan ATCs and their potential utilization
of Electronic Flight Strips. First, what is the relationship between the attitudes, beliefs
and trust of Sri Lankan Air Trac Controllers towards accepting Electronic Flight
Strips? Second, what impact do demographic factors (age, gender, work unit and
automation experience) have on attitudes, beliefs, trust and automation acceptance
of Electronic Flight Strips?
Literature Review
Variables Aecng Atudes Towards Accepng Automaon.
The factors inuencing the acceptance of automation can be understood
through the application of the Technology Acceptance Model (TAM). According to

B.L.G. Anupama Sudehani 31
TAM, there are two primary belief structures that determine attitudes towards the
intention and actual use of information technologies: Perceived Usefulness (PU) and
Perceived Ease of Use (PEOU) (Taylor and Todd, 1995). PU refers to the belief
that utilizing information technologies will enhance work performance, while PEOU
pertains to the belief that using information technologies requires minimal eort
(Venkatesh and Bala, 2008). Despite the empirical validation of TAM and similar
user acceptance models, researchers persistently seek to enhance their eectiveness
by introducing new external variables (Tarhini et al., 2013; Hu et al., 1999; Holden
and Karsh, 2010). Existing literature on the automation acceptance describe several
variables useful for understanding embracing of automation by ATCs. Attitude
towards automation (ATT) in general, was identied as a key factor inuencing
the acceptance of automation. However, the strength of this association diered
considerably depending on the specic function(s) that were targeted for automation
(Bekier et al., 2011). Albarracin et al. (2005) dene attitude as a psychological
tendency to evaluate an object or behavior with varying degrees of favor or disfavor,
encompassing subjective positive or negative assessments of people, events, or
objects (Robbins and Judge, 2008). Understanding negative attitudes can facilitate
their transformation into positive ones, promoting benecial behaviors (Horst,
1935). In the context of ATC, the concept of trust (T) in automation holds signicant
importance and cannot be overlooked. T can be dened as the attitude of an individual
towards the belief that an automated system will assist in accomplishing their
objectives, particularly in situations where uncertainty and vulnerability are present
(Lee and See, 2004). Existing literature has consistently shown that T plays a crucial
role in determining the level of acceptance towards automated systems (Riley, 1994;
Muir and Moray, 1996; Parasuraman and Riley, 1997; Lee and See, 2004). Factors
such as age and automation experience (Bekier et al., 2011) act as drivers towards
automation acceptance in air trac management. According to Davis et al. (1989)
experience of users with automation can be broken down into three areas namely: PU,
PEOU and quality. It is evident that individuals who have more positive experiences
with automation are more inclined to accept and adopt new technologies (Larson et
al., 2009; Dishaw and Strong, 1998).
TAM Model
The TAM model proposed by Davis et al. (1989), comprises ve components:
PU, PEOU, ATT, BI and actual system use. This study focuses on PU, PEOU, ATT
and BI. To gauge the acceptance, often referred to as Behavioral Intention (BI), of a
newly introduced technology, it entails exploring PU and PEOU of the technology.
PU is the belief that the introduction of new technology will positively impact
work performance. PEOU is the belief that minimal eort is required to use the
new technology (Venkatesh and Bala, 2008). ATT stands for one’s psychological

32 UCR (Series III), Vol.5, No.2, 2024 Arcle
inclination towards accepting automation which is either positive or negative. Both
PU and PEOU have a signicant direct impact on ATT. BI represents the extend of an
individual’s intention to use new automation. ATT has a stable association with BI in
TAM, and positive ATT is an indication of high acceptance (Yousafzai et al., 2007).
Trust
The role of T in the evaluation of automation acceptance has received limited
attention in previous studies. However, those that have examined this relationship
consistently report T as a signicant and positive predictor of favorable ATT of
individuals towards automation (Buckley et al., 2018; Choi and Ji, 2015; Kaur and
Rampersad, 2018). T is a direct determinant of BI (Carter and Bélanger, 2005). It has
been shown to determine BI or automation acceptance indirectly through PU and
PEOU (Pavlou, 2003).
Method
Design
Research Methodology
The primary objective of the study is to investigate how beliefs, attitudes and
trust inuence the Automation Acceptance (BI) among Air Trac Controllers in Sri
Lanka. Additionally, the study explores the inuence of demographic characteristics
on these attitudes and automation acceptance. Despite the potential benets, the
study opted not to employ a qualitative approach involving open-ended questions or
interviews. Conducting verbal interviews with a considerable number of participants
who have demanding rotating shift schedules posed signicant challenges, making it
unfeasible to meet the time constraints. In contrast, quantitative attitudinal data oers
the advantage of enabling longitudinal tracking of societal attitudes, facilitating the
examination of changes over time. Additionally, it allows for the comparison of
attitudinal dierences across dierent spatial contexts (Eirich and Corbett, 2009).
Hence, future research is facilitated by establishing a baseline prior to the adoption
of EFS, enabling subsequent investigations into attitudes during its operational use.
Thus, the research methodology employed a quantitative approach, utilizing a few
structured interviews and a survey questionnaire with closed end questions as the
primary research tools. The introduction of EFS as a prospective automation tool to
replace the current paper-based system is considered, drawing evidence from literature
and interviews conducted with ATCs in the UK and Sri Lanka. The interviews had
two main objectives: rstly, to identify any inconsistencies or discrepancies between
the functioning of the EFS system and the paper-based strip system in the current
operational setup, addressing any gaps between the past literature and the practical
implementation. Secondly, the interviews aimed to provide Sri Lankan ATCs with

B.L.G. Anupama Sudehani 33
an accurate understanding and insight into the functioning of the EFS system. The
data analysis and ndings were based on the quantitative data collected through the
survey. Ethical approval for conducting interviews and surveys was obtained from
the Craneld University Research Ethics System (CURES), UK, in 2023, prior to
commencing the study.
Research Model
The present study adopts the TAM as a fundamental framework to examine
the relationship between ATT and BI. TAM incorporates PU and PEOU as two
belief structures that inuence ATT and subsequently impact BI. As discussed under
literature review, T is identied as a construct that inuences BI, and this construct
impacts ATT. To explain the behavior of automation acceptance broadly, we add T
to the TAM model and develop study hypotheses for the extended model (Figure 1).
Figure 1: Research Model. The black arrows indicate the hypothesized relationships.
The area within the dotted lines is adapted from original TAM model by Davis et al.
(1989)
Hypothesis 1. ATT has positive eect on EFS Automation acceptance (BI)
and is signicant
Hypothesis 2. T has positive impact on EFS Automation acceptance (BI) and
is signicant

34 UCR (Series III), Vol.5, No.2, 2024 Arcle
Hypothesis 3. PU positively inuences ATT to use EFS and is signicant
Hypothesis 4. PEOU has positive eect on ATT to use EFS and is signicant
Hypothesis 5. T has positive impact on ATT to use EFS and is signicant
Research Design
The main objective of this study was to determine what impact the
identied variables PU, PEOU, ATT and T have on BI of Sri Lankan ATCs in
terms of EFS, thereby providing guidance to EFS designers and management to
plan accordingly. To achieve this, an extensive literature review was conducted,
forming the basis for hypothesis development and analysis. The review also
guided the design of interview protocols and supported the creation of two
scenarios to simulate the functionality of the EFS system, ensuring participants,
who had no prior experience with the system, could gain meaningful insights
into its features. Subsequently, a questionnaire was developed and distributed
to a group of participants for testing purposes. A pilot study was then carried
out to assess and rene the survey instrument, followed by reliability testing to
assess the consistency and stability of the instrument in measuring the intended
constructs or variables. Once a sucient number of responses were collected, the
data were analyzed, hypotheses were tested, and conclusions were drawn based
on the ndings.
Figure 2. Research Design

B.L.G. Anupama Sudehani 35
Parcipants
The study was conducted with the sole air navigation service provider in
Sri Lanka, which included diverse ATC groups (Tower, Approach, Area) in various
locations totaling 64 (Tower-28, Approach-16, Area-20) operational level air trac
control ocers. Survey invitations were distributed to the entire population, and
48 responses were received out of a target population of 64. Table 1 presents the
characteristics of the sample of 48 participants.
Table 1. Demographic Characteristics of Sample (N=48)
Characteristic Group Count Percentage
Gender Male 30 62.5%
Female 18 37.5%
Age 22-43 35 72.9%
44-65 13 27.1%
Work Unit
Tower 21 44.8%
Approach 16 33.3%
Area 11 22.9%
Automation Experience
No Experience 510.4%
Less than 10 years 25 52.1%
More than 10
years
18 37.5%
The majority of participants were male ATCs (62.5%), were between the age
range of 22-43 years (72.9%), worked at Tower (44.8%), and had less than 10 years
of experience in automation (52.1%). Only 5 (10.4%) ATCs among respondents had
no prior experience in automation. The rest of the ATCs consisted of 16 (33.3%)
Approach Controllers, 11 (22.9%) Area Controllers, 18 (37.5%) female ATCs and 18
(37.5%) ATCs with more than 10 years of automation experience.
Materials
Scenarios
An information leaet was prepared to accompany the questionnaire,
providing relevant details about EFS for individuals with limited prior knowledge.
This information leaet included two scenarios derived from the literature review
and interview insights with ATCs from Sri Lanka and the UK, outlining dierent
ways of managing traditional paper strips and EFS to achieve the same work task
or goals by the controller (Appendix A). In addition to insights from interviews and
literature, factors such as benets, drawbacks, and human factor implications were
considered in an unbiased manner when developing the scenarios.

36 UCR (Series III), Vol.5, No.2, 2024 Arcle
Scenario-1 elucidates the handling of routine tasks within a paper-based
system, while scenario-2 demonstrates the task handling within an EFS system,
providing a balanced perspective on both approaches. Prior to completing the
questionnaire, participants were provided with the online information leaet
(Appendix B) via emails which they were instructed to review in advance.
Procedure
Interviews
The ethical approval was rst acquired to conduct interviews with ATCs and
to gather data via a questionnaire. Prior to their participation, the interviewees, who
voluntarily took part in the study, were requested to provide informed consent by
signing the ethics consent form. The interviews were conducted through an online
platform, using structured questions with each session lasting between 15 to 20
minutes. The work tasks identied in the literature (Mackay, 1999; Huber et al.,
2020), were further validated by experienced ATC ocers to address any variations
that may have arisen due to temporal gaps. The handling of paper strips was validated
by ATC ocers in Sri Lanka who still utilize paper-based strips, while the EFS
handling was validated by ATC ocers in the UK.
Survey
The primary aim of this survey-based questionnaire was to evaluate beliefs,
attitudes and trust towards the acceptance of automation (EFS) among ATCs in Sri
Lanka. The survey instrument utilized a self-administered questionnaire comprising
20 closed-ended questions in English language. The questionnaire was constructed
in Qualtrics (Qualtrics, 2023) and distributed to respondents online for convenience,
accessibility, and attractiveness. Further, for greater accessibility the questionnaire
link and scenario leaet were published in Sri Lankan ATCs’ common “Whatsapp”
group. On average, respondents took approximately 4 to 6 minutes to complete the
questionnaire. Data collection lasted from 19th to 27th of June 2023. Out of the
total 64 questionnaires submitted, 54 responses were received. Only 48 were fully
completed and included in the study.
The questionnaire was divided into several sections (Appendix C). The initial
section of the questionnaire was dedicated to gathering demographic information,
including age, gender, ATC work unit and years of automation experience. The
second section was dedicated to exploring PU, the third section examined PEOU
and the fourth section investigated T in Automation. The fth section assessed the
ATT using questions proposed by Davis et al. (1989). The nal section focused on
BI. The four questions under PU, all four questions under PEOU, and the question
under BI were formulated using the basic TAM questionnaire (Turner, 2008). The
questionnaire utilized a ve-point Likert scale ranging from “Strongly Agree” to

B.L.G. Anupama Sudehani 37
“Strongly Disagree” to assess the level of agreement with the questions pertaining
to PU, PEOU and T. The section on BI included a single-item question, which also
used ve-point Likert scale from “Strongly Agree” to “Strongly Disagree”. Although
single-item measures are not always optimal, they are acceptable for certain variables
(Scarpello and Campbell, 1983).
Analysis
To evaluate whether the Likert scale represents a one-dimensional
measurement, its internal consistency was assessed using Cronbach’s alpha
(Cronbach, 1951). Descriptive statistics were used to analyze individual item scores,
including frequency and percentages, as well as scale scores using the mean and
standard deviation. Subsequently, a sensitivity analysis was performed to identify
and remove outliers from the data. The analysis was then carried out using the
outlier-free dataset. Since small sample sizes (n<50), Shapiro-Wilk test (Shapiro and
Wilk, 1965) was conducted for each variable distribution under the ve variables
in each statistical analysis. The results showed the presence of at least one or two
variables with non-normal distributions. Hence, non-parametric statistical methods
were mainly employed, due to nonnormality of independent variables PEOU, ATT,
T, BI and the ordinal nature of the dependent variable BI.
Each hypothesis (H1 to H5) was examined using Spearman Correlation
(Spearman, 1904) to determine how PU, PEOU, ATT, T inuence BI.
Equal sample sizes of 11 were considered from each group to analyze whether
the work unit (Tower, Approach, Area) has an impact on attitudinal variables (PU,
PEOU, ATT and T), as the Area Controller count was 11. Non-parametric Kruskal-
Wallis test (Kruskal and Wallis, 1952) was conducted for the comparison of three
independent samples. Signicant ndings were subsequently further compared using
the non-parametric Mann-Whitney test (Mann and Whitney, 1947).
Additionally, the relationships between the attitudinal variables (PU, PEOU,
ATT and T) and factors (age, gender and automation experience) were examined
using independent sample t-tests and non-parametric Mann-Whitney test. To test
automation experience on the variables, the group categorized as “No experience”
was excluded from the analysis concerning the inuence of inconsistencies and the
notably small sample size (N=5). Moreover, the presence of an Area Controller
without any prior automation experience is dicult to comprehend.
An alpha level of .05 was considered for a condence interval of 95% for
all statistical tests. Analysis was carried out using IBM SPSS Statistics (IBM Corp.,
2023).

38 UCR (Series III), Vol.5, No.2, 2024 Arcle
Results
Response Rate
A total of 48 complete responses were received from a target population of
64, resulting in a 75% response rate. A sample size of 10% to 30% is a satisfactory
representation of parent population (Mugenda and Mugenda, 2012). Further, Roscoe’s
Rule of Thumb (Roscoe, 1975) recommends sample sizes between 30 to 500 which
represent 10% of the target population as acceptable. Therefore, sample size 48 with
75% representation was considered as satisfactory for the statistical analysis.
Validity and Reliability Analysis
A construct is considered reliable if its Cronbach’s alpha is above 0.70
(George and Mallery, 2003). The reliability of the construct in this study was assessed
using this criterion, resulting in a high alpha value (α = 0.94), which exceeds the 0.70
threshold. The Trust scale with four items found unacceptable (α= -.13). Upon review,
items T3 and T4 were found to contribute to the low alpha value. Removing these
items signicantly improved the scale’s reliability, raising the alpha to an acceptable
level of 0.79. Since Cronbach’s alpha does not provide reliability estimates for single
items, BI was not considered individually. The results are summarized in Table 2.
Table 2. Reliability Statistics
Constructs No. of Items Alpha (α) Internal Consistency
PU 4 .88 Good
PEOU 4 .78 Acceptable
ATT 3 .88 Good
T 2 .79 Acceptable
Overall 14 .94 Excellent
Outlier Treatment
The box plot diagrams in descriptive analysis revealed the existence of outliers
in all the four independent variables and as well as in the dependent variable. The
values below Q1 (rst Quartile) - 1.5 * IQR (Interquartile Range) or above Q3 (Third
Quartile) + 1.5 * IQR were considered outliers. A total of 5 outliers were observed,
accounting for approximately 10% of the entire dataset, and several of them were
present in multiple variables. As these outliers could not be attributed to sampling
error (as all population members received invitations) or data entry mistakes (as
the questionnaires were self-administered), a sensitivity analysis was performed to
determine their retention. An independent sample t-test was performed with and

B.L.G. Anupama Sudehani 39
without the outliers to evaluate the eect of automation experience, comparing
groups with “Less than 10 years” and “More than 10 years” of experience. Results
showed no signicant dierences between groups across variables (PU, PEOU, ATT,
T and BI) when outliers were included. However, signicant dierences emerged for
ATT, T and BI after removing the outliers. A Mann-Whitney test, which is robust to
outliers, yielded similar results: no signicant dierences with outliers but signicant
dierences in ATT, T and BI after their removal. Due to the eect of outliers on the
primary ndings and eect sizes, they were excluded from the main analysis.
Parcipants’ Prole
The sample demographic presented in table 1 revealed that 73% of the sample
consisted of individuals aged 22 to 43 years, indicating an overrepresentation of this
age group. Similarly, the male representation in the sample was approximately 63%
and 62% possessed less than 10 years of experience with automation. Work unit
wise, Tower Controllers had a slight over representation accounting for 45% of the
sample.
Stascal Analysis
Descripve Analysis
PEOU obtained the highest mean score of 4.05 (SD=0.69), indicating a
positive perception of EFS’s Ease of Use. Notably, PEOU3 scored highest (M=4.15,
SD=0.94), showing condence in acquiring EFS skills. Highest percentage agreement
(62.5%) also accounted under PEOU for PEOU1, indicating that many respondents
perceive they can easily learn EFS handling. T had the lowest mean (M=3.77,
SD=0.81). However, respondents positively perceived automation as trustworthy,
with T2 recording a mean score of 3.52 (SD=0.94), indicating a low perception of
automation safety. PU had the lowest item score (PU3, M=3.44, SD=0.99) as well as
the lowest percentage for “agree” (37.5%) with highest percentage being “neutral”
(17%), suggesting EFS’s limited usefulness in decision-making and inability of
the respondents to comprehend whether or not EFS contribute in decision making.
Around 80% of the respondents have positive attitudes towards automation (ATT,
M=3.99, SD=.91) and are willing to accept automation (BI, M=3.96, SD=1.07).
Table 3 below and Table (iv) in Appendix D depict the descriptive statistics for scale
values and scale items, respectively.

40 UCR (Series III), Vol.5, No.2, 2024 Arcle
Table 3. Descriptive statistics for Scale Values
Variable Min Max M SD
PU 1.00 5.00 3.78 .89
PEOU 1.50 5.00 4.05 .69
ATT 1.33 5.00 3.99 .91
T1.5 5.0 3.77 .81
BI 1 5 3.96 1.07
Note: n=48 for all variables
Eect of Work Unit, Age, Gender and Automaon Experience
The work unit of the respondents had no signicant eect on PEOU, ATT,
T and BI (Table 5). However, PU diered signicantly across the three units (H (2)
= 7, p = .03). Specically, Tower and Area Controllers reported similar levels of
perceived usefulness of EFS, while Approach Controllers rated EFS as less useful
than both Tower (p = .012) and Area Controllers (p = .046) with large and medium
eect sizes r=.54 and r=.42, respectively. Table 4 and 5 depict descriptive statistics
for all variables per work unit and Kruskal Wallis test results, and the results of Mann
Whitney tests.
Table 4. Descriptive statistics and Kruskal’s Wallis test results for variables per
Work unit
Variable
Work Group
H (2) p
Tower Approach Area
Median M SD Median M SD Median M SD
PU 4.25 4.32 .51 4.00 3.61 .67 4.25 4.25 .47 7.03*
PEOU 4.50 4.38 .32 4.25 4.07 .78 4.00 4.25 .47 1.16 (ns) .56
ATT 4.33 4.30 .55 4.00 4.33 .49 4.33 4.42 .45 .44 (ns) .80
T4.00 4.00 .39 3.50 3.91 .83 4.00 4.23 .56 1.57 (ns) .46
BI 4.00 4.45 .52 4.00 4.27 .65 4.00 4.45 .52 .55 (ns) .76
Note: n= 11 for all groups
*p<.05 ns: non-signicant

B.L.G. Anupama Sudehani 41
Table 5. Results for Mann Whitney tests for work units under PU
Variable Work Group Mean Rank U z p r
PU
Tower 14.95 22.5* -2.52 .012 .54
Approach 8.05
Area 14.23 30.5* -1.99 .046 .42
Approach 8.77
Note: n= 11 for all groups
*p<.05
There were no signicant dierences in PU, PEOU, ATT, T and BI across
the two age groups or the two gender groups. Table 6 and 7 summarize descriptive
statistics and results of t- tests and Mann Whitney tests (per age and gender group),
respectively.
Table 6. Descriptive statistics and results of independent t-tests per Age and
Gender Group
Group
Type Variable Group n M SD Mean
Dierence
Lower
CI95%
Upper
CI95%
t
(41) pCohen’s
d
Age
PU 22-43 31 3.93 .74 -.11 -.58 .35 -.50
(ns) .62 .67
44-65 12 4.04 .45
PEOU 22-43 31 4.17 .50 -.06 -.43 .31 -.33
(ns) .74 .58
44-65 12 4.23 .63
T22-43 31 3.81 .72 -.24 -.73 .26 -.96
(ns) .34 .72
44-65 12 4.04 .72
Gender
ATT Males 26 4.24 .55 .03 -.33 .39 .16
(ns) .87 .58
Females 17 4.22 .61
TMales 26 4.00 .68 .32 -.12 .77 1.47
(ns) .15 .71
Females 17 3.68 .75
ns: non-signicant

42 UCR (Series III), Vol.5, No.2, 2024 Arcle
Table 7. Descriptive statistics and Mann-Whitney test results per Age and
Gender Group
Group
Type
Variable Group nMedian M SD Mean
Rank
U z p
Age
ATT 22-43 31 4.00 4.14 .58 20.03 125
(ns)
-1.71 .09
44-65 12 4.50 4.47 .48 27.08
BI 22-43 31 4.00 4.23 .62 21.44 168.5
(ns)
-.54 .59
44-65 12 4.00 4.33 .65 23.46
Gender
PU Males 26 4.25 4.05 .75 24.6 151.5
(ns) -1.74 .08
Females 17 3.75 3.82 .51 17.91
PEOU Males 26 4.12 4.17 .52 21.65 212
(ns) -.23 .82
Females 17 4.25 4.21 .56 22.53
BI Males 26 4.00 4.31 .55 22.71 202.5
(ns) -.52 .60
Females 17 4.00 4.18 .73 20.91
ns: non-signicant
Automation experience had no signicant eect on PU and PEOU (Table
10). However, ATT, T and BI of respondents signicantly diered between the
two experience groups. Respondents with over 10 years of automation experience
demonstrated signicantly higher positive attitudes toward automation (p = .005),
greater trust (p = .01), and a stronger inclination to accept EFS (p = .04) compared to
those with less experience, with moderate eect sizes r=.45, r=.41 and small eect
size r=.32, respectively. Table 8 and 9 summarize descriptive statistics and results of
t- tests and Mann Whitney tests per automation experience group.
Table 8. Descriptive statistics and independent t-tests results for PU and PEOU per
Automation Experience Group
Variable Group n M SD Mean
Dierence
Lower
CI95%
Upper
CI95%
t (36) pCohen’s
d
PU <10
years
23 3.84 .69
-.31 -.74 .11 -1.50
(ns) .14 .63
>10
years
15 4.15 .51
PEOU <10
years
23 4.11 .49
-.27 -.62 .07 -1.62
(ns) .11 .51
>10
years
15 4.38 .54
ns: non-signicant

B.L.G. Anupama Sudehani 43
Table 9. Descriptive statistics and Mann-Whitney test results for ATT, T and BI per
Automation Experience Group
Variable Gender
Group
nMedian M SD Mean
Rank
U z p
ATT <10
years
23 4.00 4.06 .56 15.52
81** -2.83 .005
>10
years
15 4.67 4.58 .43 25.60
T<10
years
23 3.50 3.67 .61 15.85
88.5* -2.58 .01
>10
years
15 4.00 4.30 .68 25.10
BI <10
years
23 4.00 4.09 .67 16.85
111.5* -2.02 .04
>10
years
15 5.00 4.53 .52 23.57
* p <0.05, **p<0.01
Correlaons Between Variables
A random sample of size 43 (<50) free of outliers was selected, and Shapiro-
Wilk test was conducted to test the normality. Results showed that distributions for
means of PEOU (W=.94, p=.02), ATT (W=.90, p=.001), T (W=.94, p=.03) and BI
(W=.77, p<.001) departed signicantly from normal while there was no evidence for
the means distribution of PU (W=.95, p=.07) to be non-normal. Due to non-normality
of variables PEOU, ATT, T and BI and BI being ordinal, the study conducted a
Spearman correlation analysis to determine statistically signicant relationships
between the variables.
Pearson correlation test is robust with respect to skewness, nonnormality
(Pearson, 1931; Dunlap, 1931; Havlicek and Peterson, 1976) and scale type (Havlicek
and Peterson, 1976). Thus, the study also conducted a Pearson correlation analysis to
ascertain the validity of the ndings. The results of Pearson correlation test revealed
the same correlations among the variables with diering magnitudes (Appendix E)
to the results of Spearman test given in Table 10, hence increasing the validity of the
ndings. Correlation strengths were interpreted using Dancey and Reidy’s (2004)
correlation categorization.

44 UCR (Series III), Vol.5, No.2, 2024 Arcle
Table 10. Descriptive Statistics and Spearman Correlations for Study Variables
Variables M SD 12345
PU 3.96 0.67 -
PEOU 4.19 0.53 .63** -
ATT 4.23 0.57 .54** .62** -
T3.87 0.72 .49** .50** .46** -
BI 4.26 0.62 .64** .66** .59** .54** -
Note:n=43 for all variables
**p<.01
The correlation analysis results conrmed all proposed hypotheses, as shown in
Table 11. Specically, PU demonstrated signicant positive correlations with PEOU,
BI, ATT and T, supporting H3. PEOU also had signicant positive correlations with
ATT, BI and T, conrming H4. Additionally, ATT and BI were positively correlated,
supporting H1. Finally, T showed signicant positive correlations with both BI and
ATT, thus supporting H2 and H5.
Table 11. Results of Correlation Analysis
Hypothesis Hypothesis Result rp
H1 ATT has positive eect
on BI
Accepted .59** <.001
H2 T has positive impact on
BI
Accepted .54** <.001
H3 PU positively inuences
ATT
Accepted .54** <.001
H4 PEOU has positive eect
on ATT
Accepted .62** <.001
H5 T has positive impact on
ATT
Accepted .46** <.01
**p<0.01
Summary of Findings
The research ndings revealed that the work unit of respondents had no
signicant eect on PEOU, ATT, T and BI. Tower and Area Controllers perceived
the usefulness of the EFS system similarly, whereas Approach Controllers perceived
it as less useful. Age and gender did not show signicant dierences in PU, PEOU,
ATT, T and BI. However, respondents with more than 10 years of automation
experience displayed a more positive attitude and greater trust towards automation,

B.L.G. Anupama Sudehani 45
and they were more willing to accept the EFS system compared to those with less
experience. The study conrmed strong positive correlations between PU and PEOU,
signicant moderate positive correlation between PU and ATT, and strong positive
correlation between PEOU and ATT. Additionally, ATT showed a moderate positive
correlation with BI, and T had a moderate positive correlation with ATT and BI. All
ve hypotheses were satised based on these results.
Discussion
Discussion of Results and Study Implicaons
The main purpose of this study was to determine what impact the identied
variables PU, PEOU, ATT and T have on BI of Sri Lankan ATCs in terms of
EFS and thereby provide guidance to EFS designers and management to plan
accordingly. To accomplish this goal, two scenarios were created to educate Sri
Lankan ATCs about the strengths and limitations of EFS and traditional paper
strips handling. Subsequently, feedback was obtained from ATCs at various
operational levels (Tower, Approach, Area) to evaluate their beliefs, attitudes,
and levels of trust in accepting automation by means of TAM with T incorporated
into it. The results from correlation analysis revealed that all the relationships we
used in TAM were positively and signicantly correlated (supporting hypotheses
H1, H3 and H4) including the relationships with external variable T (supporting
H2 and H5).
T had a signicant impact on BI which is consistent with previous
research ndings (Riley, 1994; Muir and Moray, 1996; Parasuraman and Riley,
1997; Lee and See, 2004). T showed a moderate, positive association with BI
which further replicates the important relationship of T and BI in Trust and TAM
model introduced by Gefen et al. (2003). EFS is primarily an eciency tool
rather than a decision support tool. Half of the respondents did not accept it as a
decision-making tool. Prior research on ATCs indicates that Trust is a signicant
predictor of automation acceptance, provided the tool does not assume decision-
making authority from ATC in safety-critical situations (Bekier et al., 2011).
Further, only one in eight ATCs in the population strongly trusted automation
as safe. Combined, these results justify the existence of association between T
and BI and the moderate strength of the relationship. Hence, designers should
prioritize building trust in the safety aspects of EFS and persuading ATCs about
its controllability and supportive nature, keeping the controller in the loop for
decision making in safety-critical circumstances. This relationship further
suggests that for low-level automation, BI is not only aected by ATT but also by
T, and hence should be incorporated in TAM.

46 UCR (Series III), Vol.5, No.2, 2024 Arcle
Apart from the indirect inuence of PU and PEOU on BI through ATT,
they showed positive moderate correlation with BI which depicted a direct
inuence that aligns with previous research ndings that state PU (Lee et al.,
2003, Davis and Davis, 1989) and PEOU (Luarn and Lin, 2005; Wei et al., 2009)
signicantly impact BI. Beliefs inuence intentions both directly, and indirectly
through attitude (Bagozzi, 1982). Thus, though this revelation is beyond the
scope of this study, PU and PEOU equally and highly inuence BI than the direct
impact of ATT which is again not explained in TAM. Hence, designers should
emphasize the usefulness and ease of use in designing EFS to ATCs.
T had a positive moderate correlation with ATT which is in line with many
prior research. This correlation is slightly weaker than the eect of PU and PEOU
on ATT. However, this indicates that PU and PEOU alone do not adequately
describe ATT. Thus, this nding reassures that T should be incorporated into the
TAM model as suggested by many other researchers (Wu and Chen, 2005).
ATT had a signicant positive association with BI of moderate strength
which is reassuring prior research ndings (Wu and Chen, 2005). This may be
partly due to the indirect channeling of PU, PEOU and T through ATT which
are direct inuencers of BI (according to the ndings of current study). The
respondents have positively perceived EFS as useful in their performance
improvement and easy to use, which, together with their positive trust towards
the automation contributed to their positive attitude towards EFS. Furthermore,
ATT was found to have a stronger correlation with BI compared to T.
Overall, ATCs show a stronger preference for technological features.
These include improved performance or perceived usefulness (PU) and ease of
use (PEOU). Trust (T) in safety features is less inuential in shaping their attitude
(ATT) toward EFS. Ease of use (PEOU) is the most dominant factor in their
acceptance of automation (BI).
Among the demographic factors, age and gender had no signicant
impact on any of variables whereas work unit and automation experience showed
signicant inuence on some of the variables that can be explained as follows.
Work unit was an inuential demographic factor on PU. Tower and
Area Controllers perceived the usefulness (PU) of the EFS system at a similar
level, whereas Approach Controllers perceived it as less useful. In other words,
Approach Control perceived a signicantly lower level of improvement of their
performance due to introduction of EFS. A previous study on Australian ATCs
states that they perceived automated tools to be useful when those tools oered
assistance and support to the controller (Langford et al., 2022). The nding may
be explained by the level of exposure of each unit to the automation or the extent
to which each Sri Lankan ATC units use handwritten strips. Approach Control
serves as the intermediary unit in the control sequence, positioned between

B.L.G. Anupama Sudehani 47
the Tower and Area Controllers. Typically, ight strips are generated through
existing Flight plan printing system with data input either from the Tower or
Area. Thus, limited time spent on data entry or manual writing of strips and more
often getting a printed strip can be suggested as a probable cause for Approach
unit as not seeing EFS as useful in further improving their performance compared
to other two units. However, there is no existing literature to either support or
refute this explanation. This nding may be specic to Sri Lankan context based
on the way they handle operations.
Automation experience had no signicant eect on PU and PEOU. Some
studies state that prior experience in specic technology positively inuences
PU and PEOU (Kim and Malhotra, 2005; Dishaw and Strong, 1998) while some
shows no eect on PU (Jackson et al., 1997). The lack of signicant inuence of
prior automation experience on PU and PEOU may be attributed to the absence
of prior experience with the specic target automation (EFS). Participants’
perceptions of usefulness and ease of use were likely based on the two provided
scenarios, which enabled similar perceptions among all participants without
direct interaction with the actual EFS system. However, automation experience
was signicant in three aspects (ATT, T, BI). ATCs with relatively more
automation experience showed a more positive attitude towards automation,
trusted automation more and were more positive on accepting automation. These
ndings agree with previous literature (Bekier et al., 2011; Kim and Malhotra,
2005). An explanation to this can be found in Bekier et al. (2011) who claimed
that ATCs with prior highly positive automation experience are likely to accept
new automation if it serves a “supportive” function. Other studies also state that
more positive prior experience with automation leads to acceptance (Larson et
al., 2009; Dishaw and Strong, 1998).
Age was not found signicant enough to have any impact on attitudes,
belief, trust and acceptance. This contrasts with the majority of prior studies,
which suggest that older ATCs or older users exhibit less willingness to accept
automation compared to younger ones (Bekier et al., 2011; Hudson et al., 2017;
Ivanov et al., 2018). One probable cause may be the categorization of age (22-
43 years & 44-65 years) in the study was not eective enough to dierentiate
between the young and old. However, this result aligns with previous study done
on ATCs regarding the introduction of a hypothetical conict detection tool in
which age did not signicantly aect automation acceptance (Bekier et al., 2011).
The fact that both males and females showed no signicant dierence
in PU, PEOU, ATT, T or BI is quite unexpected. Previous studies, though not
specically conducted in the context of ATC show that generally males have
more acceptability towards new technology than females (Rahim et al., 2023;
Schoettle and Sivak, 2014). A study done on ATC context, has found gender to have

48 UCR (Series III), Vol.5, No.2, 2024 Arcle
no eect on conict resolution performance, workload and situation awareness
(Trapsilawati et al., 2022). Gill and Grint (1995) argue that women are often
expected to adopt traditionally masculine ways of interacting with technology.
Thus, one possible explanation for this observation is that the demands of the ATC
job role outweigh gender dierences.
Limitaons of the Study
The study ndings can be limited since ATC population is a very specic,
small community of 64 of which only 75% participated in the study. In subsampling,
the sample sizes were further reduced, hence it might have had some impact on the
results to dier from past research ndings. Another limitation is that the attitudes
and trust levels were based on the two scenarios that were introduced to them and a
majority of them had not yet encountered an EFS. Thus, the results might vary if they
had exposure to an operational EFS system.
Future Direcons
More insights may be drawn if future research could consider a longitudinal
study to examine the changes over time while tracking changes in attitudes while EFS
is in operational use. Another possible suggestion for future research is to investigate
whether the three main ATC work units (Tower, Approach and Area) perceive
usefulness of automation dierently in the global context. Moreover, research should
be conducted in future to consider how other factors such as cultural context and also
personality traits of ATCs aect beliefs, attitudes, trust and automation acceptance.
Conclusion
Attitudes toward EFS are inuenced not only by Perceived Usefulness and
Perceived Ease of Use but also signicantly by Trust. Therefore, the designers should
focus on building trust in the safety aspects of EFS and persuading ATCs about its
controllability and supportive nature, particularly in safety-critical situations where
the controller’s involvement is crucial. Designers must emphasize the usefulness
and ease of use of EFS to ATCs, as these aspects have a higher inuence on attitude
toward EFS than direct trust impact. However, Perceived Usefulness and Perceived
Ease of Use alone are not sucient to fully describe attitude, indicating the need
to incorporate trust in the TAM model for a more comprehensive understanding.
The dominant attitude construct among ATCs is a minimum eort in use of EFS.
This highlights the signicance of ensuring ease of use and eciency in the design
of EFS systems. Management should leverage ATCs with substantial automation
experience as valuable assets to promote the adoption of EFS. Additionally, they
should emphasize the utility of EFS to approach control by showcasing various other
features such as integration to digital ight plans, available lists of upcoming trac,

B.L.G. Anupama Sudehani 49
ability to replay EFS display along with radar replay for incident analysis (Author,
personal communication, 2023) not just the elimination of paper strips.
Conict of Interest
The author has no conict of interest to declare.
References
Albarracin, D., Johnson, B. T., & Zanna, M. P. (Eds.). (2005). The Handbook
of Attitudes (1st ed.). New York: Psychology Press. https://doi.
org/10.4324/9781410612823
Bagozzi, R. P. (1981). Attitudes, intentions, and behavior: A test of some key
hypotheses. Journal of Personality and Social Psychology, 41(4), 607–
627. https://doi.org/10.1037/0022-3514.41.4.607
Bekier, M., Molesworth, B. R. C., & Williamson, A. (2011). Dening the drivers for
accepting decision making automation in air trac management. Ergonomics,
54(4), 347-356. doi: 10.1080/00140139.2011.558635.
Berndtsson, J., & Normark, M. (1999). Coordinative functions of ight strips: Air
trac control work revisited. In Proceedings of GROUP’99, International
Conference on Supporting Group Work (November 14–17, 1999). Embassy
Suites Hotel, Phoenix, Arizona, USA. Association for Computing Machinery.
https://doi.org/10.1145/320297.320308
Buckley, L., Kaye, S.-A., & Pradhan, A. K. (2018). Psychosocial factors associated
with intended use of automated vehicles: A simulated driving study.
Accident Analysis & Prevention, 115, 202-208. https://doi.org/10.1016/j.
aap.2018.03.021.
Carter, L., & Bélanger, F. (2005). The utilization of e-government services: Citizen
trust, innovation and acceptance factors. Information Systems Journal, 15(1),
5-25.
Chien, S.Y., Lewis, M., Sycara, K., Kumru, A., & Liu, J.-S. (2020). Inuence of
Culture, Transparency, Trust, and Degree of Automation-on-Automation Use.
IEEE Transactions on Human-Machine Systems, 50(3), 205-214. doi: 10.1109/
THMS.2019.2931755.
Choi, J. K., & Ji, Y. G. (2015). Investigating the importance of trust on adopting an
autonomous vehicle. International Journal of Human-Computer Interaction,
31(10), 692-702. https://doi.org/10.1080/10447318.2015.1070549.
Conversy, S., Gaspard-Boulinc, H., Chatty, S., Valès, S., Dupré, C., & Ollagnon, C.
(2011). Supporting air trac control collaboration with a Table Top system. In
Proceedings of the ACM 2011 conference on Computer supported cooperative
work (pp. 425-434). Hangzhou, China: ACM. DOI: 10.1145/1958824.1958891.
Cronbach, L. J. (1951). Coecient alpha and the internal structure of tests.
Psychometrika, 16(3), 297–334. https://doi.org/10.1007/BF02310555

50 UCR (Series III), Vol.5, No.2, 2024 Arcle
Dancey, C., & Reidy, J. (2004). Statistics Without Maths for Psychology: Using SPSS
for Windows. London: Prentice Hall.
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User Acceptance of Computer
Technology: A Comparison of Two Theoretical Models. Management Science,
35(8), 982–1003. https://doi.org/10.1287/mnsc.35.8.982
Davis, F. D., & Davis, F. (1989). Perceived Usefulness, Perceived Ease of Use, and
User Acceptance of Information Technology. MIS Quarterly, 13(3), 319-340.
https://doi.org/10.2307/249008
Dishaw, M. T., & Strong, D. M. (1998). Experience as a moderating variable in a
task-technology t model. In Proceedings of the Fourth Americas Conference
on Information Systems (pp. 722-724).
Doble, N. A., & Hansman, R. J. (2002). Preliminary design and evaluation of
portable electronic ight progress strips. In Proceedings of the 21st Digital
Avionics Systems Conference (pp. 7C2-7C2). Irvine, CA, USA. doi: 10.1109/
DASC.2002.1052923.
Dunlap, H. F. (1931). An empirical determination of means, standard deviations
and correlation coecients drawn from rectangular distributions. Annals of
Mathematical Statistics, 2, 66–81.
Edwards, M. B., Fuller, D. K., & Vortac, O. U. (1995). The role of ight progress
strips in en route air trac control: A time-series analysis. International
Journal of Human-Computer Studies, 43(1), 1–13.
Eirich, F., & Corbett, K. (2009). Measuring and understanding attitudes. Social
Research: Method Guides.
EUROCONTROL. (2000). ATCO attitudes towards future automation concepts: A
literature review. Brussels, Belgium: Author.
Gattiker, U. E. (1988). Technology adaptation: A typology for strategic human
resource management. Behaviour & Information Technology, 7(4), 345-359.
https://doi.org/10.1080/01449298808901882
Gefen, D., Karahanna, E., & Straub, D. (2003). Trust and TAM in online shopping:
An integrated model. MIS Quarterly, 27(1), 51–90.
George, A.S.H., George, A.S., & Baskar, T. (2022). Sri Lanka’s Economic Crisis:
A Brief Overview. Partners Universal International Research Journal, 1(2),
9–19. https://doi.org/10.5281/zenodo.6726553
George, D., & Mallery, P. (2003). SPSS for Windows step by step: A simple guide and
reference (11.0 update, 4th ed.). Boston, MA: Allyn & Bacon.
Gill, R., & Grint, K. (1995). Introduction. In The Gender-Technology Relation:
Contemporary Theory and Research (1st ed., pp. 1-28). Bristol: Taylor &
Francis. https://doi.org/10.4324/9781315041032
Havlicek, L. L., & Peterson, N. L. (1976). Robustness of the Pearson correlation
against violation of assumption. Perceptual and Motor Skills, 43, 1319–1334.

B.L.G. Anupama Sudehani 51
Holden, R. J., & Karsh, B. (2010). The Technology Acceptance Model: Its past and
its future in health care. Journal of Biomedical Informatics, 43(1), 159-172.
https://doi.org/10.1016/j.jbi.2009.07.002
Horst, P. (1935). Measuring complex attitudes. The Journal of Social Psychology,
6(3), 369-374. https://doi.org/10.1080/00224545.1935.9919744
Huber, S., Gramlich, J., & Grundgeiger, T. (2020). From paper ight strips to digital
strip systems: Changes and similarities in air trac control work practices.
In Proceedings of the ACM on Human-Computer Interaction, 3(CSCW1),
Article 28. https://doi.org/10.1145/3392833
Hudson, J., Orviska, M. & Hunady, J. (2017). People’s attitudes to robots in caring
for the elderly. International Journal of Social Robotics, 9(2), 199-210.
Hughes, J. A., Randall, D., & Shapiro, D. (1992). Faltering from ethnography to
design. In Proceedings of the 1992 ACM conference on Computer-Supported
Cooperative Work (pp. 115-122). doi: 10.1145/143457.143469.
Hu, P. J., Chau, P. Y. K., Sheng, O. R. L., & Tam, K. Y. (1999). Examining the
Technology Acceptance Model using physician acceptance of telemedicine
technology. Journal of Management Information Systems, 16(2), 91–112.
Hurter, C., Lesbordes, R., Letondal, C., Vinot, J.-L., & Conversy, S. (2012).
Strip’TIC: exploring augmented paper strips for air trac controllers.
In AVI ‘12: Proceedings of the International Working Conference on
Advanced Visual Interfaces, May 2012 (pp. 225–232). https://doi.
org/10.1145/2254556.2254598
IBM Corp. (2023). IBM SPSS Statistics [Computer software]. https://www.ibm.
com/products/spss-statistics
Ivanov, S. (2018). Tourism beyond humans – robots, pets and Teddy bears.
Paper presented at the International Scientic Conference “Tourism and
Innovations”, 14-15th September 2018, College of Tourism – Varna, Varna,
Bulgaria.
Retrieved from https://ssrn.com/abstract=3215437.
Jackson, C. M., Chow, S., & Leitch, R. A. (1997). Toward an understanding of
the behavioral intention to use an information system. Decision Sciences,
28(2), 357–389.
Kim, S. S., & Malhotra, N. K. (2005). A longitudinal model of continued IS
use: An integrative view of four mechanisms underlying postadoption
phenomena. Management Science, 51(5), 741–755.
Kaur, K., & Rampersad, G. (2018). Trust in driverless cars: Investigating key factors
inuencing the adoption of driverless cars. Journal of Engineering and
Technology Management. https://doi.org/10.1016/j.jengtecman.2018.04.006.
Kruskal, W. H., & Wallis, W. A. (1952). Use of ranks in one-criterion variance
analysis. Journal of the American Statistical Association, 47(260), 583–
621. https://doi.org/10.1080/01621459.1952.10483441

52 UCR (Series III), Vol.5, No.2, 2024 Arcle
Langford, K., Kille, T., Lee, S.-Y., Zhang, Y., & Bates, P. R. (2022). “In
automation we trust” - Australian air trac controller perspectives of
increasing automation in air trac management. Transport Policy, 125,
352-362. https://doi.org/10.1016/j.tranpol.2022.07.001.
Larsen, T.J., Sørebø, A.M., & Sørebø, Ø. (2009). The role of task-technology
t as users’ motivation to continue information system use. Computers in
Human Behavior., 25, 778-784.
Lee, J. D., & See, K. A. (2004). Trust in automation: designing for appropriate
reliance. Human Factors, 46, 50–80. https://doi.org/10.1518/
hfes.46.1.50.30392
Lee, D., Rhee, Y., & Dunham, R. B. (2009). The role of organizational and
individual characteristics in technology acceptance. International Journal
of Human-Computer Interaction, 25(7), 623-646.
Liljamo, T., Liimatainen, H., & Pöllänen, M. (2018). Attitudes and concerns on
automated vehicles. Transportation Research Part F: Trac Psychology
and Behaviour, 59(Part A), 24-44. https://doi.org/10.1016/j.trf.2018.08.010
Luarn, P., & Lin, H. H. (2005). Toward an understanding of the behavioral
intention to use mobile banking. Computers in Human Behavior, 21(6),
873-891.
Mackay, W. E. (1999). Is paper safer? The role of paper ight strips in air trac
control. ACM Transactions on Computer-Human Interaction, 6(4), 311-
340. https://doi.org/10.1145/331490.331491.
Mackay, W. E., & Fayard, A. L. (1999, October). Designing interactive paper:
lessons from three augmented reality projects. In Proceedings of IWAR (Vol.
98, pp. 81-90).
Mann, H. B., & Whitney, D. R. (1947). On a test of whether one of two random
variables is stochastically larger than the other. The Annals of Mathematical
Statistics, 18(1), 50–60. https://doi.org/10.1214/aoms/1177730491
Molesworth, B. R. C., & Koo, T. R. (2016). The inuence of attitude towards
individuals’ choice for a remotely piloted commercial ight: A latent class
logit approach. Transportation Research Part C: Emerging Technologies,
71, 51-62. ISSN 0968-090X. https://doi.org/10.1016/j.trc.2016.06.017.
Muir, B. M., & Moray, N. (1996). Trust in automation. Part II. Experimental studies
of trust and human intervention in a process control simulation. Ergonomics,
39(3), 429-460.
Mugenda, O. & Mugenda, A. (2012). Research methods dictionary. Nairobi: Applied
Research & Training Services (Arts Press).
Norman, D. A. (2013). The design of everyday things (Revised and expanded edition).
New York, NY: Basic Books.

B.L.G. Anupama Sudehani 53
Pearson, E. S. (1931). The analysis of variance in the case of non-normal variation.
Biometrika, 23, 114–133.
Parasuraman, R., & Riley, V. A. (1997). Humans and automation: Use, misuse,
disuse, abuse. Human Factors, 39(2), 230-253.
Pavlou, P. A. (2003). Consumer acceptance of electronic commerce: integrating trust
and risk with the Technology Acceptance Model. International Journal of
Electronic Commerce, 7(3), 101-134.
Qualtrics. (2023). Qualtrics [Computer software]. https://www.qualtrics.com
Rahim, A. N., Fonzone, A., Fountas, G., & Downey, L. (2023). On the Attitudes
Toward Automation in Determining the Intention to Use Automated
Buses in Scotland. Transportation Research Record, 0(0). https://doi.
org/10.1177/03611981231159116
Riley, V. (1994). A theory of operator reliance on automation. In Mouloula, M. &
Parasuraman, R. (Eds.), Human performance in automated systems: Recent
research and trends (pp. 8-14). Hillsdale, NJ: Erlbaum.
Robbins. S. P., & Judge, T. A. (2008) Essentials of Organizational Behaviour (9th
ed.). Upper Saddle River, NJ: Pearson Prentice Hall.
Roscoe, J. T. (1975). Fundamental research statistics for the behavioral sciences
(2nd ed.). Holt, Rinehart, and Winston.
Scarpello, V., & Campbell, J. P. (1983). Job satisfaction: Are all the parts there? Personal
Psychology, 36(3), 577-600. https://doi.org/10.1111/j.1744-6570.1983.
tb02236.x
Schoettle, B., & Sivak, M. (2014). A survey of public opinion about autonomous and
self-driving vehicles in the US, the UK, and Australia. University of Michigan,
Ann Arbor, Transportation Research Institute.
Shapiro, S. S., & Wilk, M. B. (1965). An analysis of variance test for normality
(complete samples). Biometrika, 52(3/4), 591–611. https://doi.org/10.1093/
biomet/52.3-4.591
Sheridan, T. B., & Parasuraman, R. (2005). Human-Automation Interaction. In R. S.
Nickerson (Ed.), Review of Human Factors and Ergonomics (Vol. 1, pp. 89-
129). Santa Monica, CA: Human Factors and Ergonomics Society.
Spearman, C. (1904). The proof of the goodness of t of certain frequency distributions,
and the calculation of their correlation coecients. Philosophical Magazine,
7(8), 568–573. https://doi.org/10.1080/14786440409463335
Surakitbanharn, C. A., Landry, S. J., & Rukijkanpanich, J. (2015). Development
of training for new air trac control technology implementation. Procedia
Manufacturing, 3, 2738-2745. ISSN 2351-9789. https://doi.org/10.1016/j.
promfg.2015.07.690

54 UCR (Series III), Vol.5, No.2, 2024 Arcle
Tarhini, A., Hone, K., & Liu, X. (2013). User acceptance towards web-based learning
systems: Investigating the role of social, organizational and individual factors
in European higher education. Procedia Computer Science, 17, 189–197.
Taylor, S., & Todd, P. A. (1995). Understanding information technology usage: A test
of competing models. Information Systems Research, 6(2), 144-176.
Trapsilawati, F., Prastiwi, P. B., Vista, Y., Myesha, Z., Herliansyah, M. K., &
Wijayanto, T. (2022). Investigating trac and controller factors in spatial
multitasking: The context of air trac conict resolution. International
Journal of Transportation Science and Technology, 11, 536–544.
Tripathi, S., Sharma, K., & Pandya, R. (2022). A study of the economic crisis and its
impacts with special reference to Sri Lanka. Towards Excellence, 14(4).
Turner, M., Kitchenham, B., Budgen, D., & Brereton, P. (2008). Lessons learnt
undertaking a large-scale systematic literature review. In EASE’08: Proceedings
of the 12th international conference on Evaluation and Assessment in Software
Engineering (pp. 110-118).
Vallor, S., & Bekey, G.A. (2017). Articial intelligence and the ethics of self-learning
robots. In P. Lin, K. Abney, & R. Jenkins (Eds.), Robotics 2.0. New York:
Oxford University Press.
Venkatesh, V., & Bala, H. (2008). Technology Acceptance Model 3 and a research
agenda on interventions. Decision Sciences, 39(2), 273-315.
Wei, T. T., Marthandan, G., Chong, A. Y. L., Ooi, K. B., & Arumugam, S. (2009).
What drives Malaysian m-commerce adoption? An empirical analysis.
Industrial Management & Data Systems, 109(3), 370-388.
Weihe, V. I. (1953). Computer applications in air trac control. In Proceedings of
the Eastern Joint Computer Conference (pp. 18–22). IEEE.
Wu, I. L., & Chen, J. L. (2005). An extension of Trust and TAM model with TPB in
the initial adoption of on-line tax: An empirical study. International Journal of
Human-Computer Studies, 65, 784-808.
Yousafzai, S. Y., Foxall, G. R., & Pallister, J. G. (2007). Technology acceptance: a
meta‐analysis of the TAM: Part 1. Journal of Modelling in Management, 2(3),
251-280. https://doi.org/10.1108/17465660710834453

B.L.G. Anupama Sudehani 55
Supplementary Material
Appendix A.
Table (i). Identication of the current dierences in handling of EFS compared to
that of Paper Strips
Work goal of controller Paper strips handling EFS handling
Adding a new aircraft to
controlled trac
Select a suitable coloured paper strip
depending on arrival or departure,
Added automatically in to pending
bays or can manually populate with
call sign and type of aircraft with
the ability to ll in further details
Insert it in a strip holder picked from
a container,
Place it in the active strip bay
Inform or remind stand
by controller or another
controller of an action
related to an aircraft
Keep the relevant strip noticed
by keeping a hand on it until the
relevant action is coordinated and
fullled
Highlight the relevant strip or
specic eld in the strip that needs
further action
Entering Clearances Manual writing using an ink pen
from one hand while holding the
strip holder from the other hand
An SID or STAR is populated
automatically or by clicking on
the clearance box it allows to
enter manually
Maintain aircraft in the
order of landing
Might use both hands to remove
and place the strip holders in correct
order of landing on the strip bay
system functionality allows the
controller to drag and drop the
strips across the EFPS display
including establishing a landing
order, departure sequence,
highlighting ground conictions
Handing over the
responsibility of an
aircraft to another
controller in controlling
sequence
If another controller is taking over
control of same position the strips
will remain at the same place in the
same strip bay
A transfer bay between adjacent,
intermediate or nal controllers
Or A pending strip is populated and
exist in the Flight plan system
or A strip transfer box to specify
whom to transfer
If a controller in next controlling
unit is taking over then the same
strip holder is physically being
handed over or
the next controller writes up a new
paper strip with required details
and insert in a holder and place on
his strip bay
Note: Adapted partly from “Table 2. Direct comparison between controllers’
interactions with paper and digital ight strips to achieve their goals.” Huber, S., Gramlich,
J., & Grundgeiger, T. (2020). From paper ight strips to digital strip systems: Changes
and similarities in air trac control work practices. Proceedings of the ACM on Human-
Computer Interaction, 4(CSCW1), 28, 1–21. https://doi.org/10.1145/3392833, p. 28:15.
Other sources: Interviews with Sri Lankan ATCs and UK ATCs.

56 UCR (Series III), Vol.5, No.2, 2024 Arcle
Appendix B.
Scenario 1 (An ATC working in Conventional Paper Strip System) and
Scenario 2 (An ATC working in Electronic Fight Strip System)
Figure (i). Scenario 1 (An ATC working in Conventional Paper Strip System)
Figure (ii). Scenario 2 (An ATC working in Electronic Fight Strip System)

B.L.G. Anupama Sudehani 57
Appendix C.
Components of the Questionnaire for data collection
Table (ii) Questionnaire Part 1- Demographics
Variable Value
Age 0: 22-43 1: 44-65
Gender 0: Male 1: Female
Work unit 0: Tower 1: App 2: ACC
Automation Experience 0: No experience
1: Less than 10 years
2: More than 10 years
Table (iii) Questionnaire Part 2- Variables in the Research Model
Variable Item SD D N A SA
PU PU1 Using Electronic Flight Strips can help me
improve my trac controlling performance
1 2 3 4 5
PU2 Electronic Flight Strips help me be more
productive in job tasks
1 2 3 4 5
PU3 Electronic Flight Strips improve my
decision making in trac controlling tasks
123 4 5
PU4 I would nd Electronic Flight Strips system
useful in my job
1 2 3 4 5
PEOU PEOU1 Electronic Flight Strips will be easy for me
to learn
1 2 3 4 5
PEOU2 Electronic Flight Strips make it easy for me
to achieve my desired work tasks
1 2 3 4 5
PEOU3 I am condent that I can easily acquire
Electronic Flight Strip handling skills
1 2 3 4 5
PEOU4 Electronic Flight Strips are easy to use 1 23 4 5
TT1 I trust automation as an Air Trac Controller 1 2 3 4 5
T2 I feel automation is safer 1 2 3 4 5
T3 Automation is unpredictive 1 2 3 4 5
T4 Automation is unreliable 1 2 3 4 5
ATT ATT1 Using Electronic Flight Strips is a good idea 1 23 4 5
ATT2 Using Electronic Flight Strips is a wise idea 1 2 3 4 5
ATT3 Using Electronic Flight Strips is a pleasant
trend
1 2 3 4 5
BI BI I presently wish to use the Electronic Flight
Strip System regularly at work 1 2 3 4 5
Note: Scale denitions are as follows. SD-Strongly Disagree, D-Disagree, N-Neither
Agree nor Disagree, A-Agree, SA-Strongly Agree

58 UCR (Series III), Vol.5, No.2, 2024 Arcle
Appendix D.
Table (iv). Descriptive statistics for Scale Items
Item MSD Mode S.D. D. N. A. S.
A. Scale Scale %
PU1 3.71 1.09 4 3 4 724 10 Agree 50
PU2 4.02 1.06 4 235 20 18 Agree 41.7
PU3 3.44 .99 4 2 5 17 18 6 Agree 37.5
PU4 3.96 1.03 4 2 2 820 16 Agree 41.7
PEOU1 4.04 .74 4 1 0 6 30 11 Agree 62.5
PEOU2 4.04 .97 4 2 2 326 15 Agree 54.2
PEOU3 4.15 .94 5 1 2 6 19 20 S. A. 39.6
PEOU4 3.96 .87 4 0310 21 14 Agree 43.8
T1 4.02 .84 4 04 4 27 13 Agree 56.3
T2 3.52 .94 4 1 6 14 21 6 Agree 43.8
ATT1 4.08 1.01 4 2 2 422 18 Agree 45.8
ATT2 3.90 1.12 4 3 2819 16 Agree 39.6
ATT3 4.00 .90 4 1 2 7 24 14 Agree 50
BI 3.96 1.07 4 3 2424 15 Agree 50
Note: Scale denitions are as follows. SD-Strongly Disagree, D-Disagree, N-Neither
Agree nor Disagree, A-Agree, SA-Strongly Agree
Appendix E.
Table (v). Descriptive Statistics and Pearson Correlations for Study Variables
Variables n M SD 123 4 5
PU 43 3.96 0.67 -
PEOU 43 4.19 0.53 .64** -
ATT 43 4.23 0.57 .56** .56** -
T43 3.87 0.72 .47** .51** .45** -
BI 43 4.26 0.62 .60** .61** .59** .50** -
**p<.01