Cultural Inﬂuences on Requirements Engineering Process

in the Context of Saudi Arabia

Tawfeeq Alsanoosy, Maria Spichkova and James Harland

School of Science, RMIT University, Melbourne, Australia

Keywords:

Requirement Engineering, Software Engineering, Software Development Process, Cultural Aspects.

Abstract:

Software development requires intensive communication between the requirements engineers and software

stakeholders, particularly during the Requirements Engineering (RE) phase. Therefore, the individuals’ culture

might inﬂuence both the RE process and the result. Our aims are to investigate the extend of cultural inﬂuences

on the RE process, and to analyze how the RE process can be adapted to take into account cultural aspects.

The model we present is based on Hofstede’s cultural theory. The model was applied on a pilot case study in

the context of the conservative Saudi Arabian culture. The results reveal 6 RE aspects and 10 cultural factors

that have a large impact on the RE practice.

1 INTRODUCTION

Requirements Engineering (RE) is one of the key

phases of the software development life cycle. RE

refers to the process of identifying and documenting

the stakeholder’s requirements. It was initially intro-

duced for use within Western culture before the ex-

pected beneﬁts began attracting the RE practitioners

in various other cultures, to adopt the RE process. The

RE process has a critical impact on software qual-

ity. It requires an intensive communication between

the requirements engineers and software stakehold-

ers, in order to elicit, validate, and reﬁne the require-

ments. This requires comprehensive communication

skills, and an understanding of the individuals’ behav-

ior and culture (Thanasankit, 2002; Hanisch and Cor-

bitt, 2007).

Culture has a great inﬂuence on how individu-

als and companies operate and how they adopt tech-

niques, methods, and practices to achieve their goals.

Culture shapes the way in which people think, com-

municate, understand, and select what is important

(Hofstede et al., 2010). Cultures have distinctive

beliefs, customs, and approaches to communication.

This diversity is inﬂuenced by the behavior practice

within these cultures. Hanisch et al. (2001) argued

that the social and cultural aspects of RE affect the

success of software development and it cannot be ig-

nored. The objectives of our research are to iden-

tify the cultural inﬂuences on the RE processes and

practices, and to analyze how the RE process can be

adapted to take into account cultural aspects.

But what do we usually understand under the term

culture? There are several deﬁnitions for it, speciﬁed

from the social psychology view. Hofstede, whose

work has been cited 70,392 times

, deﬁnes culture as

“collective programming of the mind which distin-

guishes members of one human group from another”,

cf. (Hofstede et al., 2010). We take this approach as

a background for our research not only because Hof-

stede conducted one of the most inclusive studies on

the inﬂuence of cultural values on the workplace, but

also because this is the most SE-related social psy-

chology study. Hofstede conducted the research with

employees working for IBM, covering more than 50

countries. This theory is widely accepted and adopted

in software engineering cultural studies (Borchers,

2003; Lim et al., 2015; Ayed et al., 2017).

Contributions: The RE process is originated

from Western culture and hence most of the standard

RE practices are driven by the United States. Suc-

cessfully applying such process/practice within dif-

ferent cultural context would, therefore considering

the cultural differences. This study investigates the

impact of the Saudi national culture on the practice of

RE process. We collected data, through interviewing

6 software practitioners and 2 software engineering

researchers, about the RE practices that were less or

more successfully adapted within the Saudi Arabian

culture. We then identiﬁed a model designed to il-

According to the Google Scholar, retrieved 3/11/2017

Alsanoosy, T., Spichkova, M. and Harland, J.

Cultural Inﬂuences on Requirements Engineering Process in the Context of Saudi Arabia.

DOI: 10.5220/0006770701590168

In Proceedings of the 13th International Conference on Evaluation of Novel Approaches to Software Engineering (ENASE 2018), pages 159-168

ISBN: 978-989-758-300-1

159

lustrate the interplay between the core of the RE pro-

cess and the Hofstede cultural dimensions, within the

Saudi culture. The results demonstrated the feasibil-

ity of our hypothesis as well as allowed us to identify

a missing hypothesis, which we added to the reﬁned

model. The current results of this study might be use-

ful for improving the RE process within the Saudi cul-

ture and any culture that has similar national cultural

proﬁles.

Outline: This paper is structured as follows. Sec-

tion 2 introduces the background of our research, fo-

cusing on RE processes and Hofstede’s cultural the-

ory. Our conceptual Model and hypotheses are pre-

sented in Section 3. Section 4 describes the research

methodology. The results of the case study are pre-

sented in Section 5. Section 6 introduces the pro-

posed framework. In Section 7, the related work is

discussed. Section 8 summarizes the paper.

2 BACKGROUND

2.1 Requirements Engineering Process

Zave (1997) deﬁnes RE as “the branch of software en-

gineering that is concerned with the real-world goals

for, functions of, and constraints on software systems.

It is also concerned with the relationship of these fac-

tors to precise speciﬁcations of software behavior, and

to their evolution over time and across software fam-

ilies”. As highlighted by (Nuseibeh and Easterbrook,

2000), the above deﬁnition features three important

aspects of RE: (1) RE is concerned with real-world

goals, (2) its aim is to provide precise speciﬁcations

of the requirements, and (3) the deﬁnition emphasizes

the reality of the rapid change of user’s needs. Thus,

requirements speciﬁcation serves as a basis for the

development of the system, its design and architec-

ture, cf. Spichkova (2011). In this paper, the term re-

quirements engineer is used to refer to the practitioner

who works on specifying, analyzing, and document-

ing users’ requirements.

The RE process is usually described as the follow-

ing ﬁve core (sub-)processes/activities, cf. also Som-

merville and Lenarcic (2011); Abran et al. (2004);

Pandey et al. (2010):

• Requirements elicitation: The process of gath-

ering, identifying, extracting, and exposing the

stakeholders’ needs;

• Requirements analysis: The process of reﬁn-

ing stakeholders’ requirements and constraints by

deﬁning the procedure, input/ output data and ob-

ject of the required system;

• Requirements speciﬁcation: The process of cre-

ating a written description of stakeholders re-

quirements, needs and constraints in a complete,

accurate, understandable and consistent format;

• Requirements validation: The activity to ensure

that the requirement speciﬁcation is consistent,

clear and capture users’ needs and constraints;

• Requirements management: The process of

tracing changes to the requirements over the time

of system development and, in some cases, appli-

cation as well.

2.2 Hofstede’s Cultural Theory

Hofstede et al. (2010) proposed to focus on the fol-

lowing six dimensions of a national culture:

• Power Distance Index (PDI): The degree to

which the lesser authority individual accepts and

expects that power is distributed unequally;

• Individualism versus collectivism (IDV): The

degree to which people within a society collab-

orate with each other; Thus, in highly individu-

alistic societies would encourage individual au-

thority, achievement, and give the power to make

individual-decision.

• Masculinity versus Femininity (MAS): The de-

gree to how the social gender roles are distinct;

• Uncertainty Avoidance Index (UAI): The extent

of society members to feel either uncomfortable

or comfortable in chaotic or confused situations;

• Long- vs. Short-term Orientation (LTO): The

degree to which people within a society are linked

to its own past while dealing with the present and

the future challenges; and

• Indulgence versus Impulses (IND): The degree

to which people within a society have fun and en-

joy life without restrictions and regulations.

In this theory, each country has a numerical score

within the above dimensions to deﬁne the society of

this country. The score runs from 0-100 with 50 as an

average. The Hofstede’s rule for the score is that if a

score is above the average, the culture relatively high

on that dimension. For example, the Saudi Arab score

of the uncertainty avoidance index dimension is (80)

and it is considered high.

As this research used the Saudi Arabian context

for the pilot study, let us present an example of scores

for this country as per Hofstede et al. (2010). The

score of power distance is very high (95), which

means that Saudi people accept authority’s decisions

without any justiﬁcations and expect to be told what

ENASE 2018 - 13th International Conference on Evaluation of Novel Approaches to Software Engineering

160

to do. With a score of 25, Saudi Arabia is con-

sidered a collective society where people belong to

group that takes care of them. For the masculinity

dimension, a score of 60 indicates that conﬂicts are

resolved by letting the strong win and high authori-

ties are supposed to be decisive. Saudi Arabia scores

80 for uncertainty avoidance which means that Saudi

people prefer avoiding uncertainty situations. The

low score (36) of long-term indicates that establish-

ing truth is the main concern as well as respect for

traditions. Lastly, Saudi Arabia with a score of 52

does not clearly show any preference of indulgent or

restrained.

As our study argues that most of the RE standards

are driven by the United States culture and the adap-

tation of the RE process requires considering the cul-

tural differences, let us present a brief comparison be-

tween Saudi Arabia and the United States as per Hof-

stede’s score (see Figure 1). In short, Saudi Arabia

is remarkably different from the United States within

three dimensions: PDI, IDV, and UAI. For the PDI,

Saudi Arabia score is much higher than the United

States (40). This may indicate that high authority in

Saudi Arabia may have more privileges over the sub-

ordinates. In contrast, the United State’s scores very

high (90) on the IND dimension, which indicates that

the United States society is more individualist than

Saudi Arabia. Finally, the score of UAI in Saudi Ara-

bia is high, which may express the difﬁculties of cop-

ing with unexpected situations.

Figure 1: Hofstede’s Culture Scores for Saudi Arabia and

the United States.

3 CONCEPTUAL MODEL AND

HYPOTHESES

Table 1 presents our conceptual model. Each “?” in

the table means that the corresponding cultural di-

mension might have a major inﬂuence on the corre-

sponding RE activity, whereas each “X” means that

the corresponding cultural dimensions might have

less/no inﬂuence on the corresponding RE activity.

Referring to the Hofstede’s scores, we believe that

three of Hofstede’s cultural dimensions (PDI, IDV,

and UAI) have major impacts on the practice of RE

process in Saudi Arabia, while the other dimensions

might have a minor impact. Therefore, we formulated

the following hypotheses (H):

• H1: The high value/score of PDI in Saudi Arabia

signiﬁcantly impacts on the RE activities.

• H2: The high value/score of IDV in Saudi Arabia

signiﬁcantly impacts on the RE activities.

• H3: The high value/score of UAI in Saudi Arabia

signiﬁcantly impacts on the RE activities.

• H4: The slightly low value/score of MAS in Saudi

Arabia may impact on the RE activities.

• H5: The slightly high value/score of LTO in Saudi

Arabia may impacts on the RE activities.

• H6: The IND dimension has no impact on the RE

activities in Saudi Arabia.

Table 1: Inﬂuence of Cultural Dimensions on RE Activities.

RE Activity

Cultural Dimensions

PDI IDV MAS UAI LTO IND

Elicitation ? ? ? ? ? X

Analysis ? ? ? ? ? X

Speciﬁcation X X X ? ? X

Validation ? ? ? X ? X

Management ? ? X X X X

4 EXPERTS’ VIEW: CASE STUDY

We conducted a qualitative study through a pilot case

study with interviewing Saudi software practitioners

and academics. Qualitative research is used to inves-

tigate a social behavior from the view of the study

participants with attempting to achieve a holistic un-

derstanding of the situation. The study was con-

ducted following the process suggested by Runeson

et al. (2012), where the process contains the follow-

ing steps: First, we deﬁned the research questions and

contracted the study design. Then, we developed the

case study procedure and protocol for data collection.

Next, the ﬁrst author conducted in-depth interviews

in Saudi Arabia. Finally, the data analysis procedure

was applied to the data.

The pilot case studies provide a powerful mech-

anism to study the aspects that support or prevent

innovative behavior in software engineering practice

(Monteiro et al., 2016). Here, we used a pilot case

study to understanding the Saudi cultural inﬂuence

on RE process.The following research questions were

formulated:

Cultural Inﬂuences on Requirements Engineering Process in the Context of Saudi Arabia

161

RQ1: What requirements engineering activities

are adopted by the Saudi requirements engineers?

RQ2: What cultural aspects and traditions in-

ﬂuence the requirements engineering practice in

Saudi Arabia?

4.1 Data Collection

To collect the data, we interviewed eight participants:

six software practitioners, one senior academic, and

one lecturer in a Saudi university. The snowballing

technique was used to recruit the participants. Partici-

pants were asked to propose other persons who could

probably contribute to the study. Neuman (2002) sug-

gests that snowballing sampling is often used if the

sample size is small or if there is a preference to se-

lect well-informed participants. In addition, the snow-

ball recruiting technique was beneﬁcial for this study

because social networking, in a collectivist society, is

the main source of information Hofstede et al. (2010).

The practitioners interviewees were selected

based the following criteria:

• The participants had to be involved in the RE pro-

cess during systems development;

• Working in a Saudi software company;

• Be employed in a medium- to a large-sized orga-

nization; and

• Engaged in eliciting the requirements from Stake-

holders.

The academics interviewees were selected based

the following criteria:

• The participants had to be involved in the RE pro-

cess during systems development; and

• Working in a Saudi university.

The social nature of this study leaded us to

adopt semi-structured interviews. The questions were

mixed of open-ended and close-ended types includ-

ing 30 questions, focusing on the RE process adopted

in Saudi Arabia and the cultural factors inﬂuence the

RE process from Saudi Arabian context. When the

participants accepted the invitation to participate, the

following documents were sent to them one day be-

fore the face-to-face interview: the interview ques-

tions, the explanatory statement and the consent form.

The average duration of the interviews was about 90

minutes. Some interviews held in Arabic where other

in English, based on the interviewees’ preferences.

4.2 Data Analysis

A thematic analysis method was used to analyze data

from the interview transcripts, following the guide-

lines described by Braun and Clarke (2006). As per

Marks and Yardley (2003), the beneﬁt of thematic

analysis is that it allows to combine analysis of the

frequency of codes with analysis of their meaning in

context, to overcome the major criticisms of content

analysis. The process of thematic data analysis goes

through six phases:

1. Familiarization with data;

2. Generating initial codes;

3. Searching for themes;

4. Reviewing themes;

5. Deﬁning and naming themes; and

6. Producing a report.

All interviews were transcribed, and the interviews

conducted in Arabic were also translated to English.

All recordings were reviewed and coded manually

using coding and pattern matching techniques Miles

et al. (1984). Codes that emerged in each interview

were compared. Next, we analyzed the interview data

to identify common points and ideas. Finally, we

grouped the emerged 16 themes into two main groups:

RE aspects and cultural aspects.

4.3 Validity and Reliability

Validity and reliability of research aim to ensure

the trustworthiness of a study. We considered the

three validity aspects of exploratory research recom-

mended by Yin (2013): First, we constructed this

study through in-depth reviewing of the literature.

Second, we reviewed some of the participants’ soft-

ware documentation that obtained during the inter-

view as a second source of evidence. Finally, we es-

tablished a chain of evidence from the beginning of

the search through selecting the study interviewees,

using interview process and transcribing and inter-

preting the data. External validity achieved by using

Braun’s guidelines for coding and analyzing the data

Huberman and Miles (1985). The reliability achieved

by documenting a rich description of the research

method and the interview protocol (Neuman, 2002).

ENASE 2018 - 13th International Conference on Evaluation of Novel Approaches to Software Engineering

162

5 RESULTS AND DISCUSSION

5.1 Requirement Engineering Process

To collect the responses to the ﬁrst research question,

the interviewees were asked to describe the RE pro-

cesses performed in all software development projects

they conducted. From the participants’ responses, we

identiﬁed core aspects that describe the state of RE

practice in the Saudi context, related to the 5 RE ac-

tivities (see Section 2.1).

5.1.1 Requirements Elicitation

Requirements elicitation was named by all partici-

pants as an important part of the process. The inter-

viewees generally elicit requirements (1) either from

software stakeholders directly, or (2) by searching

about the problem domains, or (3) by asking stake-

holders for examples of similar systems. Four re-

quirement elicitation techniques were noted: inter-

views, focus groups, observations and prototypes.

The interviewees believed that it was important

to understand the vision of the top managers or soft-

ware owner before eliciting the requirements from the

end users, especially for a new system. The goals,

scope, objectives and main issues from high authority

people were used as guidelines to develop software

that meets the company’s vision, where the end users’

needs are less important.

Almost 80% of the interviewees agreed that a

face-to-face interview was the best method used for

eliciting customers’ requirements. Adopting inter-

view as an elicitation method assisted the study par-

ticipants to

• acquire a comprehensive knowledge about users’

requirements;

• verify the requirements during the session;

• help clients to express their needs freely;

• minimize the time to establish a trust; and

• build a good relationship with the customers.

One participant reported one serious problem in

implementing software sponsored by the government:

requirements were identiﬁed by representative stake-

holders from the government and no end users partic-

ipated the elicitation session.

5.1.2 Requirement Analysis

Another RE activity reported by almost all partici-

pants was requirements analysis. The main objective

of this activity is to (1) detect and resolve conﬂicts

and (2) to know how the requirements should be im-

plemented.

The participants expressed that analyzing the re-

quirements not only involve examining the software

functionality, but also the technical, business and cul-

tural aspects. As an example of the business aspect

was that an end user may request to delete records

in the business sheet, whereas according the interna-

tional ﬁnancial and accounting standard, such records

cannot be removed. An example of technical as-

pects was that an end user might request a feature that

threatens the company’s security. For cultural aspects,

an end user might ask for changing the interface lan-

guage. To analyze the requirements, two interviewees

used a scenario-based analysis method, which is a de-

scription of the steps needed to complete a given task.

5.1.3 Requirement Validation

As indicated by interviewees, requirements validation

was not an independent phase, as validation activi-

ties are either embedded in the requirement analysis

phase, or not conducted at all.

5.1.4 Requirement Documentation

All participants from industry did not adopt any rec-

ommended requirements documentation standard and

did not follow any guidelines in constructing the doc-

umentation template. Instead, the requirements doc-

umentation was structured based on the requirement

engineer’s experience or following a documentation

template used for some previous projects. The partic-

ipants gave the following three reasons for not adopt-

ing any requirements documentation standards:

• The requirement engineer and the client are satis-

ﬁed with the current requirements documentation;

• None of the existing documentation standards has

been released in an Arabic requirements docu-

mentation version; translating or using English

documentation standards was not convenient.

• The software teams are highly qualiﬁed (by their

own deﬁnition) and, therefore, there is no need for

a high level description of the process.

However, from the point of view of academic par-

ticipants, following the international documentation

standards (e.g., IEEE) with a customization to the

company standard is seen as a more preferable so-

lution than structuring the software documentation

based on the requirement engineer’s experience.

Cultural Inﬂuences on Requirements Engineering Process in the Context of Saudi Arabia

163

5.1.5 Requirement Change Management

Interestingly, the participants did not perceived re-

quirements change management as a part of the RE

process. 2 out of 5 participants from the industry

adopted tools or techniques for tracking requirements

changes. This demonstrates that adaptation of the

good practices and standards of RE within software

development is still in progress in Saudi Arabia. The

ﬁndings concur with Alnafjan (2012).

5.2 Cultural Inﬂuences

While analyzing the collected data, we identiﬁed 10

cultural aspects inﬂuencing the RE practices:

1. Deference to elderly people and to high authori-

ties;

2. Autocratic decision-making;

3. Limited trust;

4. Belief in expertise;

5. Relationships;

6. Empathy with client;

7. Letting the strongest win;

8. Gender segregation;

9. Dress code; and

10. Using English for requirements documentation.

We grouped these points into 5 cultural categories:

We cover 4 out of 6 Hofstede’s dimensions (power

distance, uncertainty avoidance, collectivism, mas-

culinity) and introducing a new category to present

speciﬁc cultural factors that cannot ﬁt the Hofstede’s

dimensions. Based on the results of the conducted

interviews, Table 2 illustrates the inﬂuence of Saudi

Arabia national culture on the core activities within

the RE process. Each Xin the Table 2 means that the

corresponding RE activity is inﬂuenced by the corre-

sponding cultural dimension. In this model, we ex-

cluded two of the Hofstede’s cultural dimensions and

added a speciﬁc cultural factors. we excluded the LTO

dimension because of our data did not explore any

inﬂuences or dependencies of this dimension on the

Saudi RE process and we excluded the indulgence di-

mension because it does not have any connection with

the RE process. Also, we added the speciﬁc cultural

factors as it emerged during analyzing the study par-

ticipants’ interviews. We named it speciﬁc because it

is exclusively distinct to the Saudi culture.

Table 2: Inﬂuence of Saudi culture on RE Activities.

RE Activity

Cultural Dimensions

PDI IDV MAS UAI Speciﬁc

Elicitation X X X X

Analysis X X X

Speciﬁcation X X

Validation X X

Management X X

5.2.1 Power Distance Index

As mentioned in Section 2.2, the score of power dis-

tance index (PDI) in Saudi Arabia is deﬁned as 95

(very high). The following three points ﬁt this dimen-

sion: deference to elderly people and to high authori-

ties, autocratic decision-making, and lack of trust.

Deference to elderly people and to high author-

ities: As per (Hofstede et al., 2010), in the countries

with a high PDI, showing respect for parents and older

people is a basic and lifelong virtue. The data collec-

tion highlighted the fact that due to the Saudi Arabian

culture, demonstrating respect was habitually prac-

ticed. A total of ﬁve participants reported that defer-

ence of customers, especially elderly and high author-

ities, is seen by them as a standard practice. One of

the participants stated that ‘most of the time, I cannot

argue with elderly or high authority employees’.

Autocratic decision-making: Hofstede et al.

(2010) stated that “power distance will affect the de-

gree of centralization of the control and decision-

making structure and the importance of the status of

the negotiators’. In that context, the analysis of the

collected data demonstrated that the decision-making

was signiﬁcantly consultative and autocratic rather

than democratic. The reason for this is a centralized

decision-making in Saudi Arabia, i.e., all decisions

are taken by high authorities.

Three participants believed that ‘The manager will

agree on what he thinks it is beneﬁcial to the business.

All about business. If it is beneﬁcial, he will accept

it’. and therefore the study participants accept and

agree one what is decided by the top managers. How-

ever, one participant claimed a critical drawback to

the centralized decision as ‘If he [decision maker] is

satisﬁed, we do not care about the end users’ needs’.

Limited trust: According to Hofstede et al.

(2010), high PDI cultures suffer from lack of trust. A

total of ﬁve participants mentioned that lack of trust in

both Saudi and non-Saudi requirements engineer was

a problematic issue. During the elicitation session,

clients tended to not openly discuss the main core of

the business issues, instead, clients may ask the re-

quirements engineer about: the number of projects he

was involved in, years of experience, race, the educa-

ENASE 2018 - 13th International Conference on Evaluation of Novel Approaches to Software Engineering

164

tion and technique background, etc. Surprisingly, the

level of trust is normally established based on the re-

quirements engineering’s nationality or based on the

strength of the relationship that built during the RE

phase. One participant commented that some nation-

alities were highly trusted just because Saudi people

believe in them. Also, three participants agreed that

establishing a good relationship with customers sig-

niﬁcantly increased the trust level. However, if the

level of trust is not maintained on the early stage of

the software development life cycle, the RE process

might end up with an absolute failure to achieve its

purposes.

5.2.2 Uncertainty Avoidance

The score of uncertainty avoidance in Saudi Arabia is

deﬁned as 80 (very high), cf. Section 2.2. According

to Hofstede, in high uncertainty cultures, strong belief

in expertise and their knowledge is common Hofstede

et al. (2010).

The professional informers assume that Saudi cus-

tomers accept their recommendations and suggestions

not because they are experts in the ﬁeld, but because

these recommendations are important and valuable.

According to the interviewees, presenting facts about

the best way to solve the problem can change software

stockholders’ mind easily. However, this expresses

only the view point and self-estimates of requirements

engineers rather than provides a complete real picture,

where the belief in expertise (including the belief in

one’s own expertise) plays a large role.

5.2.3 Collectivism

Saudi Arabia is considered a collective society (score

25, cf. Section 2.2), where two following points play

an important role: building a relationship and empa-

thy with users.

Building a relationship: In a collectivism cul-

tures, building a relationship in workplaces is very

important, similar to family relationships. Interviews,

as a face-to-face method to elicit the requirements,

permit the software engineers to create an emotional

relationship with the client, which was indicated as a

positive culture aspect. Six participants highlighted

that building a good relationship means having an in-

depth understanding of users’ problem domain as well

as establishing a certain level of trust. For example,

during the requirements elicitation building a relation-

ship with the clients can be used as facilitator to over-

come barriers (e.g, lack of user involvement) or chal-

lenges in general such as project over-budget.

Empathy with users: Achieving user satisfaction

was continually mentioned as the reason to accept

new changes, which can be explained in terms of feel-

ing empathy with the client to fulﬁll his needs. Four

participants agreed that ‘He is a client. We cannot

make him unhappy. The client should be satisﬁed’.

5.2.4 Masculinity

Solving conﬂicts by letting the strongest win was of-

ten mentioned in the transcript. As per Hofstede’ the-

ory, the resolution of conﬂicts in a masculine society

(Saudi Arabia is considered a masculine society with

score 60) is solved by letting the strongest win Hofst-

ede et al. (2010).

In Saudi context, autocratic behavior was indi-

cated when the interviewee spoke about solving con-

ﬂicts between end users and high authorities. The

ﬁndings showed that most of the conﬂicts were re-

solved by agreeing with the top managers’ prefer-

ences rather than giving end users an opportunity to

justify this need. Moreover, we ﬁnd that this cultural

aspect is highly related to the PDI dimension. In high

PDI dimension, such as Saudi Arabia, the decision is

made based on the higher authorities’ opinion.

5.2.5 Speciﬁc Cultural Factors

The study identiﬁed three points that are very for

Saudi culture: gender segregation, dress code, and

English language, where the ﬁrst two can also be seen

as a very special application of masculinity dimen-

sion.

Gender segregation: A total of ﬁve participants

asserted that communicating with Saudi women dur-

ing the RE process, even during the elicitation ac-

tivity, was not an issue. The study participants ap-

proached the female end users in their workplace.

Even though the workplace is separated into male

and female sections, the professionals’ participants

elicited the requirements through either face-to-face

meetings or using communication applications such

skype or e-mail. However, only one participant re-

ported that gathering the requirements from female

was difﬁcult due to shyness and self-respect of the

Saudi women.

Dress code: Some participants highlighted that

wearing a suit for meeting with clients is very impor-

tant. They perceived this rule as a cultural aspect.

English language: Even though the ofﬁcial writ-

ten language in Saudi Arabia is Arabic, all partici-

pants indicated that software documentation is usu-

ally written in English, especially for government

businesses. Since Saudi people have diverse levels

of the speaking and writing English language skills,

adopting English language to communicate and docu-

ment the requirements might introduce miscommuni-

Cultural Inﬂuences on Requirements Engineering Process in the Context of Saudi Arabia

165

cation and misunderstanding between the clients and

requirements engineers and, therefore, it might affect

the overall RE process.

6 PROPOSED FRAMEWORK

Based on the expert participants’ overview, we de-

signed a framework of the national cultural inﬂuence

on the practice of RE process as presented in Figure 2.

We are planning to replicate the study and compare

the results with several cultures, in order to propose

a generalized framework that identiﬁes the inﬂuence

of national culture on the practices of RE process.

For this purposes, we extended and reﬁned the initial

model (presented in Table 1) to explore the inﬂuence

of the cultural background on the RE process within

other country/culture context, as presented in Table 3.

Thus, based on the Hofstede’s country/culture score

for a particular dimension, we can explore (1) what

RE activities have to be adapted and (2) which of the

culture factors might have an inﬂuence on the RE pro-

cess.

Table 3: Inﬂuence of Cultural Dimensions on RE Activities.

RE Activity

Cultural Dimensions

PDI IDV MAS UAI LTO IND Speciﬁc

Elicitation ? ? ? ? X X ?

Analysis ? ? ? ? ? X ?

Speciﬁcation X ? X ? ? X ?

Validation ? ? ? X ? X ?

Management ? ? X X X X ?

Based on the initial results, we formulated a num-

ber of new hypotheses for our future work:

• H1: The high PDI has an impact on the practice

of RE activities.

• H2: The low IDV has an impact on the practice

of RE activities.

• H3: The high UAI has an impact on the practice

of RE activities.

• H4: The high/low MAS has an impact on the

practice of RE activities.

• H5: The high LOT cultural dimension has an im-

pact on the practice of RE activities.

• H6: The IND dimension has no impact in the

practice of RE activities.

• H7: There are speciﬁc cultural values impact

the practice of RE activities, within each coun-

try/culture context.

7 RELATED WORK

To analyze cultural aspects, many social psycholo-

gists propose theoretical frameworks that illustrate the

key cultural factors that distinguish one culture from

another, cf. e.g., Hofstede et al. (2010), Trompenaars

and Hampden-Turner (2011), and Hall (1989). How-

ever, from our best knowledge, our work is the ﬁrst

one focusing on the cultural aspects within the RE

process in Arabian culture. In our earlier works, we

analyzed individual and social requirement aspect of

software development, cf. Alharthi and Spichkova

(2016); Alharthi et al. (2015), as well as requirements

for global product development, focusing on the di-

versity, which comes from the differences in the regu-

lations, laws and cultural aspects for different coun-

tries or organizations, cf. Spichkova et al. (2015);

Spichkova and Schmidt (2015). In our current work,

we went further and investigated cultural aspects in

details, focusing on the Arab culture.

Several studies have investigated the impact of

culture and cultural differences on RE processes. For

example, Damian and Zowghi (2003) analyzed the

challenges faced by stakeholders during the RE pro-

cess in distributed multi-site organizations. By con-

ducting in-depth interviews, the study showed that

culture and cultural differences have negative impacts

on RE activities. However, the existing studies do not

propose any solution that can be generalized and ap-

plied in the future. The goal of our work is to solve

this problem.

Thanasankit (2002) conducted a study on whether

the speciﬁcs of Thai culture might have inﬂuences on

the RE process, in particulate, the concept of power

distance. According to the authors, the communica-

tion and decision-making process in Thailand is inﬂu-

enced by the unequal power distribution and the level

of uncertainty. Based on Thai cultural aspects, the

researchers suggested some strategies to improve RE

practices. These strategies focus on Thai culture only

and cannot be generalized. However, the results of the

study concur with our hypothesis that the cultural in-

ﬂuence on the RE process might be signiﬁcant and it

should be taken into account to improve the RE prac-

tices. In contrast to Thanasankit (2002), our research

adopted the six dimensions of the Hofstede’s cultural

theory on the RE process and the results of the pi-

lot case study demonstrated the feasibility of our ap-

proach.

Isabirye and Flowerday (2008) evaluated the

existing requirements elicitation techniques and

methodologies in the context of rural culture in South

Africa, coming to the conclusion that the stakehold-

ers’ culture plays a signiﬁcant role in undertaking

ENASE 2018 - 13th International Conference on Evaluation of Novel Approaches to Software Engineering

166

Figure 2: Framework of RE process in a cultural context.

the RE process, which concurs with our hypotheses.

Borchers (2003) analyzed the impact of culture on

the software engineering techniques in Japan, India,

and the United States. The study also demonstrated

that cultural differences have a great inﬂuence on soft-

ware engineering process. In our work, we go further

and focusing on the development of the correspond-

ing framework.

8 CONCLUSION

RE process is highly sensitive to cultural aspect and

requires extensive communication skills. Several

studies have recognized that a success RE process

does not only depend on the technique or method used

but also depends on understanding the individuals’

behavior and culture. In this paper, we investigate the

inﬂuence of cultural background on performing the

RE process from a conservative culture, Saudi Ara-

bia.

We conducted interviews with software practition-

ers and software engineering researchers in Saudi

Arabia. Using thematic analysing technique with

eight participants, this study identiﬁes 10 cultural fac-

tors play a role on the RE practice and 5 RE aspects

from the Saudi context. As a result, we introduce a

model of RE processes with a cultural context. In this

model, we map the Hofstede’s cultural theory to the

core parts of the RE process to investigate the possible

dependencies and inﬂuences.

Hofstede’s culture dimension allows us to ﬁnd rel-

evant correlation regarding the impact of the cultural

background on RE practice. The result of this study,

therefore, would help the Saudi RE practitioners and

other RE practitioners, having the similar national

cultural proﬁle, to identify in advance the potential

cultural implications that might be faced.

Future work: As the next steps, we are going

to expand the scope of the study (1) to cover other

countries, (2) to gather the information on the soft-

ware stakeholders’ view points, as they are a very

important actors within the RE process, (3) to cover

not only small-to-medium companies but also a large

ones, as this might provide additional aspects. We

are planning to extend this work to a large number of

participants and comparing the impact within several

cultures.

ACKNOWLEDGEMENTS

We would like to thank participants for kindly accept-

ing to participate in this research project. We appreci-

ate their cooperation and engagement.

The ﬁrst author is supported by a scholarship from

Taibah University in Saudi Arabia.

Cultural Inﬂuences on Requirements Engineering Process in the Context of Saudi Arabia

167

REFERENCES

Abran, A., Moore, J. W., Bourque, P., Dupuis, R., and Tripp,

L. L. (2004). Guide to the software engineering body

of knowledge: 2004 version SWEBOK. IEEE Com-

puter Society.

Alharthi, A. and Spichkova, M. (2016). Individual and so-

cial requirement aspects of sustainable elearning sys-

tems. In International Conference on Engineering Ed-

ucation and Research, pages 1–8. WSU.

Alharthi, A., Spichkova, M., and Hamilton, M. (2015). Re-

quirements engineering aspects of elearning systems.

In 24th Australasian Software Engineering Confer-

ence, pages 132–133. ACM.

Alnafjan, K. (2012). An empirical investigation into the

adoption of Software Engineering Practice in Saudi

Arabia. International Journal of Computer Science

Issues (IJCSI), 9(3):328–332.

Ayed, H., Vanderose, B., and Habra, N. (2017). Agile Cul-

tural Challenges in Europe and Asia: Insights from

Practitioners. In 39th International Conference on

Software Engineering, pages 153–162. IEEE.

Borchers, G. (2003). The software engineering impacts

of cultural factors on multi-cultural software devel-

opment teams. In 25th International Conference on

Software Engineering (ICSE), pages 540–547.

Braun, V. and Clarke, V. (2006). Using thematic analysis in

psychology. Qual. Res. Psychol., 3:77–101.

Damian, D. E. and Zowghi, D. (2003). RE challenges in

multi-site software development organisations. Re-

quirements Engineering, 8(3):149–160.

Hall, E. T. (1989). Beyond culture. Anchor.

Hanisch, J. and Corbitt, B. (2007). Impediments to re-

quirements engineering during global software devel-

opment. European Journal of Information Systems,

16(6):793–805.

Hanisch, J., Thanasankit, T., and Corbitt, B. (2001). Under-

standing the cultural and social impacts on require-

ments engineering processes-identifying some prob-

lems challenging virtual team interaction with clients.

ECIS 2001 Proceedings, page 43.

Hofstede, G., Hofstede, G. J., and Minkov, M. (2010). Cul-

tures and Organizations - Software of the Mind: Inter-

cultural Cooperation and its Importance for Survival.

McGraw-Hill.

Huberman, A. M. and Miles, M. B. (1985). Assessing lo-

cal causality in qualitative research. The self in social

inquiry: Researching methods, pages 351–381.

Isabirye, N. and Flowerday, S. (2008). A model for eliciting

user requirements speciﬁc to south african rural areas.

In Annual Conference of the South African Institute

of Computer Scientists and Information Technologists,

pages 124–130. ACM.

Lim, S. L., Bentley, P. J., Kanakam, N., Ishikawa, F., and

Honiden, S. (2015). Investigating country differences

in mobile app user behavior and challenges for soft-

ware engineering. IEEE Trans. on Software Engineer-

ing, 41(1):40–64.

Marks, D. F. and Yardley, L. (2003). Research Methods for

Clinical and Health Psychology. SAGE Publications.

Miles, M. B., Huberman, A. M., and Saldana, J. (1984).

Qualitative data analysis: A sourcebook. Beverly

Hills.

Monteiro, C. V., da Silva, F. Q., and Capretz, L. F. (2016).

The innovative behaviour of software engineers: Find-

ings from a pilot case study. In Proceedings of the

10th ACM/IEEE International Symposium on Empir-

ical Software Engineering and Measurement, page 7.

ACM.

Neuman, L. W. (2002). Social research methods: Qualita-

tive and quantitative approaches.

Nuseibeh, B. and Easterbrook, S. (2000). Requirements en-

gineering: a roadmap. In Conference on the Future of

Software Engineering, pages 35–46. ACM.

Pandey, D., Suman, U., and Ramani, A. (2010). An effec-

tive requirement engineering process model for soft-

ware development and requirements management. In

Advances in Recent Technologies in Communication

and Computing (ARTCom), pages 287–291. IEEE.

Runeson, P., H

ost, M., Rainer, A., and Regnell, B.

(2012). Case Study Research in Software Engineer-

ing - Guidelines and Examples. Wiley.

Sommerville, I. and Lenarcic, J. (2011). Software engineer-

ing. Pearson/Addison-Wesley, 9th edition.

Spichkova, M. (2011). Architecture: Requirements + de-

composition + reﬁnement. Softwaretechnik-Trends,

31(4).

Spichkova, M. and Schmidt, H. (2015). Requirements engi-

neering aspects of a geographically distributed archi-

tecture. In 10th International Conference on Evalua-

tion of Novel Approaches to Software Engineering.

Spichkova, M., Schmidt, H. W., Nekvi, M. R. I., and Mad-

havji, N. H. (2015). Structuring diverse regulatory

requirements for global product development. In Re-

quirements Engineering and Law, pages 57–60. IEEE.

Thanasankit, T. (2002). Requirements engineering – explor-

ing the inﬂuence of power and thai values. European

Journal of Information Systems, 11(2):128–141.

Trompenaars, F. and Hampden-Turner, C. (2011). Rid-

ing the waves of culture: Understanding diversity in

global business. Nicholas Brealey Publishing.

Yin, R. K. (2013). Case study research: Design and meth-

ods. Sage publications.

Zave, P. (1997). Classiﬁcation of research efforts in require-

ments engineering. 29(4):315–321.

ENASE 2018 - 13th International Conference on Evaluation of Novel Approaches to Software Engineering

168