Inﬂuence of Human Personality in Software Engineering

A Systematic Literature Review

Anderson S. Barroso

1,2

, Jamille S. Madureira da Silva

2,3

, Michel S. Soares

and Rogerio P. C. do Nascimento

Computing Department, Tiradentes University, Aracaju, Brazil

Computing Department, Federal University of Sergipe, São Cristóvão, Brazil

Computing Department, Federal Institute of Sergipe, Aracaju, Brazil

Keywords:

Personality Traits, Human Factors, Software Engineering, MBTI, Big Five, FFM.

Abstract:

Personality of software engineering professionals has been a continuous element of interest in academic re-

search. Researchers have applied different models of personality analysis in various software engineering

areas to identify improvement points, to promote job satisfaction and to better organize teams. This paper

aims to conduct a study, by means of a systematic literature review (SLR), to evaluate personality models ap-

plied in software engineering and to understand how human personality inﬂuences professional’s work. Three

main models, most frequently used, were identiﬁed (MBTI, BIG 5 and FFM) to evaluate software engineering

professionals. There is evidence of the inﬂuence of personality on the activities performed. However, some

results have suggested that the study of personality is not an easy task to be performed, because there are

contradictions in ﬁndings that challenges the validity of studies.

1 INTRODUCTION

Modern society has increasingly demanded quality

and productivity in all branches of industry and prod-

uct development. Software Engineering is no excep-

tion. It has been recognized that excellent individuals

in a software team makes a huge inﬂuence on the ﬁnal

products quality, and that one developer can be orders

of magnitude more productive than other (Brooks,

1987) (Sudhakar et al., 2012) (Meyer et al., 2014).

Therefore, it is important to know how to create ef-

fective teams that will more effectively and efﬁciently

produce high quality software products (Richardson

et al., 2012) (Yilmaz et al., 2016).

According to Pressman and Maxim (Pressman and

Maxim, 2014), software is developed and used by

people, and supports interaction between people, thus

characteristics such as human behavior and coopera-

tion are fundamental to their development. The con-

cern with human aspects becomes of fundamental im-

portance for success of a software project.

Recently, Spinellis and Androutsellis prepared

an analysis of software development activities in

different companies (Spinellis and Androutsellis-

Theotokis, 2014). They concluded that the success

of a team depends on the targets set for its members,

the required controls, pre-deﬁned standards and busi-

ness rules that directly inﬂuence the work of software

engineering teams.

Considering these factors, one needs to identify

technical and managerial issues of each individual in

order to arrange the best teams. One factor that stands

out is to evaluate the relationship between human per-

sonality and the activity to be performed by each pro-

fessional. According to Capretz (Capretz, 2003), hu-

man personality represents how people’s behavior in-

ﬂuences decision-making and the ability to assimi-

late information, generating an important debate in

academia and industry.

In order to understand if human personality inﬂu-

ences activities of software engineering profession-

als, this paper proposes a Systematic Literature Re-

view (SLR) (Kitchenham, 2004) with the objectives

of identifying, summarizing, and analyzing all mod-

els that have been proposed or have been used to rep-

resent inﬂuence of human personality in Software En-

gineering.

Results show that this topic has been researched

over the past 14 years, including several recent publi-

cations, which indicates the importance of the theme

of human personality in Software Engineering. An

example is work presented in (Yilmaz and O’Connor,

Barroso, A., Silva, J., Soares, M. and Nascimento, R.

Inﬂuence of Human Personality in Software Engineering - A Systematic Literature Review.

DOI: 10.5220/0006292000530062

In Proceedings of the 19th International Conference on Enterprise Information Systems (ICEIS 2017) - Volume 3, pages 53-62

ISBN: 978-989-758-249-3

2012), carried out in industry and academia, with 63

developers, which showed through preliminary stud-

ies that there are more extroverts than introverts pro-

fessionals in software teams. Another example is the

work presented in (Ferreira, Vito and Natasha, N,

2014) which concluded that software testers have sig-

niﬁcantly higher conscientiousness factor than other

software development professionals. This paper is or-

ganized as follows: Section 2 is about the method

followed in the SLR, Section 3 discussed the results

achieved and in Section 4 we conclude the study and

indicate perspectives for future work.

2 PLAN AND EXECUTION OF

THE SYSTEMATIC

LITERATURE REVIEW

A SLR is conducted to catalog studies on professional

personality in software engineering in order to iden-

tify relevant primary studies related to this issue. The

review was carried out between June 18, 2016 and

July 20, 2016. Two phases of the SLR, (A) Plan-

ning and (B) Conduction are presented in this section.

First, two research questions were proposed as fol-

lows:

• RQ1: What psychological models are used to

identify traits of personality of software engineer-

ing professionals?

• RQ2: Activities performed by software engineer-

ing professionals are inﬂuenced by their personal-

ity?

These research questions are answered in Section

2.1 Search Strategy

This research process includes a search in sources ac-

cessed on the web, thus manual query was discarded.

Most common found terms were used in publications

on the topic “Human Personality Inﬂuence in Soft-

ware Engineering”.

Initially, we conducted the search considering a

period of ﬁve years, ﬁnding 110 works. Whereas a

SLR must obtain potentially relevant primary stud-

ies (Kitchenham, 2004), we consider 110 instances

a low number, then we decided to extend the period

for 14 years. Thus, the search was conducted be-

tween 2003 and 2016. This choice is because even

though some models have been published since 1980,

we have identiﬁed that relevant work about human

personality in Software Engineering began to occur

in 2003.

We chose to carry out the search in three scien-

tiﬁc repositories: IEEE, both “Conference Publica-

tion” and “Journals & Magazines” sections; ACM,

in sections “Proceeding”, “Newsletter”, “Magazine”

and “Journal”, and Elsevier (Subject area: Computer

Science, Human-Computer / Publications, type: jour-

nals). We also use Google Scholar as a search support

tool.

In order to establish the search strategy, we ini-

tially identiﬁed the main keywords “Personality” and

“Software Engineering”. We have also identiﬁed syn-

onyms for these keywords. Thus, the ﬁnal search

string was:

(Personality OR Motivation OR Behavior OR

Humans Aspects OR Satisfaction) AND (Software

Engineering OR Software Development OR

Programming OR Software Test OR Developer OR

Tester OR ICT Manager OR Project)

2.2 SLR Conduction

Three steps were considered in this SLR: (1) Selec-

tion process and evaluation quality of primary study,

(2) Inclusion and exclusion criteria, and (3) Informa-

tion extraction strategy.

After applying the search string, 391 papers were

found, distributed as follows: ACM (220), IEEE

(141), and Elsevier (30). Each item found in the

search process was analyzed by two researchers and

reviewed by two others. Two researchers were re-

sponsible for searching and performing a ﬁrst scan

and the other two to perform a second check and ini-

tiate inclusion and exclusion process. Within this ap-

proach, the most important roles of a SLR, as quoted

by Kitchenham (Kitchenham, 2004), were consid-

ered. Figure 1 illustrates the number of publications

distributed by year.

2.2.1 Inclusion and Exclusion Criteria

Selection criteria are used to evaluate each study re-

covered from the search sources. Thus, the inclusion

criteria (I) used to include relevant studies in our sys-

tematic review are:

• I1 - the primary study proposes or uses models to

identify human personality in Software Engineer-

ing;

• I2 - the primary study proposes that the human

personality inﬂuences activities of software engi-

neering professionals.

Alternatively, the exclusion criteria (E) used to ex-

clude studies that do not contribute to answering the

research questions are:

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

Figure 1: Number of publications by year.

• E1 - the primary study does not address human

personality in Software Engineering;

• E2 - the primary study does not propose or does

not use models for identifying human personality;

• E3 - the primary study does not present an abstract

or its full text is not available;

• E4 - the primary study is written in a different lan-

guage than English;

• E5 - the primary study consists of a compilation

of works, for instance, from a conference or work-

shop.

After the adoption of the inclusion criteria, 122

works were selected as follows: ACM (61), IEEE

(48) and Elsevier (13). The next step was to read

the abstracts of 122 papers selected in the previous

step. The researchers, in consensus, selected 53 pa-

pers as follows: ACM (29), IEEE (21), and Else-

vier (3). Thereafter, we applied the exclusion criteria,

which resulted in 21 papers, distributed as follows:

ACM (9), IEEE (9), and Elsevier (3). Table 1 illus-

trates the steps for ﬁnal paper selection.

Figure 3 shows the number of publications by

country. Note that the U.S. has the highest number

of publications in searched period, followed by New

Zealand with 3 and Pakistan with 2.

2.2.2 Information Extraction Strategy

For each selected study, the process of assessing

the quality of primary studies was executed, and

the following data were extracted: date of execu-

tion of study; description treatment of observed risks;

description of performed study; experimental study

project; threats to validity: internal, external, con-

Table 1: Selection Steps.

Steps Repositories T

AC IE EL

After search

string

(2003-2016)

220 141 30 391

After inclu-

sion criteria

61 48 13 122

After read-

ing the

abstracts

29 21 3 53

After exclu-

sion criteria

9 9 3 21

AC=ACM; IE=IEEE; EL=Elsevier; T=Total

Figure 2: Number of publications by country.

struction, conclusion; results of the study; lessons

learned; future perspectives; additional comments.

After analysis, the authors of this paper came to

consensus on what would be selected and deﬁned the

ﬁnal list of twenty one papers that are presented in

Table 2.

Inﬂuence of Human Personality in Software Engineering - A Systematic Literature Review

Table 2: Included Primary Studies.

ID Author Year Venue

01 (Capretz, 2003) 2003 International Journal of Human-Computer Studies

02 (Gorla and Lam, 2004) 2004 Communications of the ACM

03 (Darcy and Ma, 2005) 2005 Annual Hawaii International Conference on System Sci-

ences

04 (Peslak, 2006) 2006 ACM SIGMIS CPR Conference on Computer Personnel

Research:

05 (Chao and Atli, 2006) 2006 Agile Conference

06 (Aronson et al., 2006) 2006 Journal of Engineering and Technology Management

07 (Rutherfoord, 2006) 2006 Conference on Information Technology Education

08 (Mourmant and Gallivan, 2007) 2007 ACM SIGMIS CPR Conference on Computer Personnel

Research

09 (Gomez and Acuna, 2007) 2007 International Conference on Software Engineering &

Knowledge Engineering

10 (Feldt et al., 2008) 2008 International Workshop on Cooperative and Human As-

pects of Software Engineering

11 (Salleh et al., 2009) 2009 International Symposium on Empirical Software Engineer-

ing and Measurement

12 (Shoaib et al., 2009) 2009 International Multitopic Conference

13 (Salleh et al., 2010) 2010 ACM/IEEE International Conference on Software Engi-

neering

14 (Hannay et al., 2010) 2010 IEEE Transactions on Software Engineering

15 (Salleh et al., 2011) 2011 IEEE-CS Conference on Software Engineering Education

and Training

16 (Raza, A and Capretz, L F, 2012) 2012 Journal of Software Engineering

17 (Yilmaz and O’Connor, 2012) 2013 EUROMICRO Conference on Software Engineering and

Advanced Applications

18 (Ferreira, Vito and Natasha, N,

2014)

2014 International Conference on Computer Science & Educa-

tion

19 (Kanij et al., 2015) 2015 International Workshop on Cooperative and Human As-

pects of Software Engineering

20 (Gulati et al., 2016) 2016 ACM SIGSOFT Software Engineering

21 (Smith et al., 2016) 2016 Int. Workshop on Coop. and Human Aspects of Soft. Eng.

(CHASE)

3 ANALYSIS OF RESULTS

Before discussing the results, it is of utmost im-

portance to understand that personality encompasses

non-intellectual, psychological characteristics that are

informative about an individual, and that helps to

describe the differences between people. It is also

thought to be organized, relatively enduring, an inﬂu-

ence on the person’s interactions with others and in-

ﬂuences their adaptation to their social environment.

Criteria by which people differ from each other are

called psychological traits. Traits are representative

factors to predict one’s behaviour patterns, feeling,

thinking and related activities (Kanij et al., 2015).

There are several conclusions that can be drawn

from these ﬁndings that seem worthwhile for further

discussion.

This SLR identiﬁed MBTI as the most popular

model to analyze software developer personalities.

However, studies differ on the use of experimen-

tal projects (Yilmaz and O’Connor, 2012). In addi-

tion, research on the software developer’s personal-

ity should distinguish more carefully between models

Big 5 and FFM, as they differ in certain characteris-

tics (McCrae and John, 1998) (Zillig et al., 2002).

However, there are limitations that have to be

taken into consideration. First, one should consider

that this revision may have lost some relevant work,

moreover, the quality of the studies used in this analy-

sis were not homogeneous. Another factor that should

be taken into consideration is that issues such as

methodologies used in the studies, sample size and

use of statistical methods were not taken into account

in the analysis. In case the work had a psycholog-

ical model applied to Software Engineering, it was

enough for this research. This behavior may limit the

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

conclusions drawn from the reported results.

3.1 RQ1: What Psychological Models

are Used to Identify Traits of

Personality of Software Engineering

Professionals?

In response to RQ1, the twenty-one selected articles

were read, and by consensus the researchers found

that during the period under study 3 psychological

models are considered essentials to assessing the per-

sonality of software engineering professionals. An-

other 3 models were used only once in the researched

period 5.

3.1.1 MBTI - Myers-Briggs Type Indicator

The MBTI model is the most applied one to iden-

tify software developers and is based on the model of

Carl Jung (Myers et al., 1985) (Myers et al., 1998).

MBTI divides human personality into three dimen-

sions: how do people relate to the world; the way

information is known and; the way information is pro-

cessed.

The MBTI is a personal characteristics identiﬁca-

tion tool, which enables identifying the characteris-

tics, strengths and aspects of development. Thus, the

model identiﬁes four pairs of preferences known as

dichotomies (Myers et al., 1985) (Myers et al., 1998),

brieﬂy explained as follows.

Dichotomy Attitude (E-I): (I) Introversion: tend

to be involved with ideas. Prefer to think before act-

ing. Need time to think and recover energy. In gen-

eral, they are not very sociable; (E) Extroversion: tend

to act. Enjoy performing various activities. Act ﬁrst

and think later. When inactive, their energy decreases.

They are generally sociable.

Dichotomy Functions (S-N and T-F): (S) Sensing:

they rely more on tangible things, concrete, sensory

information. They prefer details and facts. For them

the meaning is in data. They need a lot of informa-

tion; (N) Intuition: they prefer abstract and theoret-

ical information, which can be associated with other

information. They like to interpret data based on prior

knowledge. They work well with incomplete and de-

ductible information; (T) Thinking: they take deci-

sions based on logic and look for rational arguments;

(F) Feeling: they take decisions based on their feel-

ings rather than emotions.

Dichotomy Lifestyle (J-P): (J) Judging: they feel

calm when decisions are taken; (P) Perceiving: they

feel tranquil when leaves open decisions.

Combination of these 4 pairs can result in 16

different personalities. Personality type is made

up of four letters. First letter represents the atti-

tude dichotomy ((I) Introversion or (E) Extroversion),

the second and third letters represent functions di-

chotomy ((S) Sensory or (I) Intuitive - (T) Thinking

or (F) Feeling) and the fourth letter is the Lifestyle di-

chotomy ((J) Judging or (P) Perceiving) (Myers et al.,

1998), as depicted in Fig. 3.

Figure 3: MBTI Personality Types.

In theory, each of sixteen different kinds of per-

sonality measured by MBTI can be seen as a set

of patterns that indicates how the individual behaves

(Ferreira, Paula G and Silva, F, 2008).

3.1.2 BIG 5 Personality Dimensions

Another important model to evaluate personality is

the Big Five Personality Dimensions (Big 5) (Gold-

berg, 1990).

The Big Five model was originally created in

the 1970s by two independent research teams - Paul

Costa and Robert McCrae (at the National Institutes

of Health) and Warren Norman (at the University of

Michigan) / Lewis Goldberg (At the University of

Oregon) (Norman, 1967) - who have followed differ-

ent paths to achieve the same results, meaning that

most human personality traits can be reduced to ﬁve

large dimensions regardless of language or culture.

In order to identify the ﬁve dimensions, the

researchers conducted interviews with hundreds of

questions to thousands of people and then analyzed

data using a statistical procedure known as factorial

analysis, which is used to reduce a large amount of

information to a synthetic and relevant set (McCrae

and John, 1998) (Norman, 1967).

The Big 5 consists of ﬁve personality traits that

are universal to the human population: (i) Insurgency:

Inﬂuence of Human Personality in Software Engineering - A Systematic Literature Review

refers to the orientation of an individual in relation to

others. Individuals with insurgency traits tend to be

talkative, bold, assertive and sociable; (ii) Agreeable-

ness: refers to the sympathy and social interaction of

a person. They are nice guys, get along well with oth-

ers, they are reliable and useful; (iii) Conscientious-

ness: refers to the organization. Conscientious indi-

viduals are suitable for hard working; they are orga-

nized and able to complete tasks in the proposed time;

(iv) Neuroticism: Refers to stress, anxiety, fear, and

the volatility of a person. Individuals with this trait

tend to not let emotion interfere with their work; (v)

Openness to experience: Refers to imagination, cu-

riosity and wit of an individual. Individuals with this

trait tend to be curious, open-minded and arts con-

noisseur (Goldberg, 1990).

Those features are understood as a complete de-

scription of the persona, are stable over a period of ten

years and may vary among cultures (Goldberg, 1990).

3.1.3 FFM - Five Factor Model

FFM is a variation of the Big 5 model. If on one hand

the descriptions of the ﬁve characteristics are similar

to those of Big 5, on the other hand these two models

differ in terms of the theoretical basis, of causality and

measurement. While Big 5 assumes that personality

traits are important for social interaction, FFM pro-

vides a comprehensive model of causes and contexts.

FFM assumes, in a bio-social context, the inﬂuence

of genetic and environmental causality. Big 5 model

comprises a circular measurement model in contrast

to the FFM which is based on hierarchical measure-

ments (McCrae and John, 1998). The ﬁve factors

FFM are: (i) Extroversion: refers to a person involved

with the outside world. Extroverts feel comfortable in

social relations, are enthusiastic, friendly and active;

(ii) Agreeableness: refers to co-operation ability of an

individual; (iii) Conscientiousness: refers to how in-

dividuals manage, regulate and direct their impulses;

(iv) Neuroticism: refers to how an individual expe-

riences negative feelings. Those who have low neu-

roticism are emotionally stable, calm, conﬁdent and

secure; (v) Openness to experience: refers to an indi-

vidual’s imaginative and creative traits (McCrae and

John, 1998)(Salleh et al., 2009).

These three models were used in the period sur-

veyed to assess personality of software developers,

software testers, students and teachers by means of

surveys and experiments. A relationship between the

selected works and the theoretical models is presented

in Table 3. For the sake of data comprehension, we

used the IDs deﬁned in Table 2.

3.1.4 Additional Models

Other less relevant models, such as Job Diagnostic

Survey (JDS) were found that measure how work im-

pacts people’s lives in an organization or society in

general (Morris and Venkatesh, 2010).

The Personal Style Inventory (PSI) consists of

8 traits, including 5 Big 5 models, plus Assertive-

ness, Image Management, Optimism, Tough Minded-

ness, Work Drive and Customer Service Orientation

(Lounsbury et al., 2003). This study found that these

characteristics signiﬁcantly affect job satisfaction.

Temperament Classiﬁcation Keirsey (KTS) is a

self-rated personality test designed to help people un-

derstand themselves and others. It is closely associ-

ated with the Myers-Briggs Indicator (MBTI) type.

However, the difference is that MBTI is concerned

about the thinking and feelings of people while KTS

is concerned about their obvious behaviors. In the

classiﬁcation of personality types, the MBTI focuses

on contrasting extraversion and introversion, while

KTS emphasizes the sensation / intuition perspective

(Salleh et al., 2014).

3.2 RQ2: Are Activities Performed by

Software Engineering Professionals

Inﬂuenced by Their Personality?

On RQ2, identiﬁed through reading selected articles,

some points are raised by authors and leverage the dis-

cussion on the topic.

According to works presented in (Mourmant and

Gallivan, 2007), (Ferreira, Vito and Natasha, N,

2014) and (Kanij et al., 2015), constant technologi-

cal change, new methodological paradigms, outsourc-

ing services and distribution of work teams, chal-

lenges to understand personalities of software devel-

opers, Open source development, evolution of tasks

complexity, globalization, new business models, and

constantly changing activities have also their share of

contribution.

Some researchers have found that software devel-

opers with independent tasks that require a certain de-

gree of creativity tend to be introverts (Capretz, 2003)

(Darcy and Ma, 2005), while developers who perform

tasks that require collaboration and leadership tend to

be extroverts (Salleh et al., 2009).

On the other hand, there are inconsistencies be-

tween the results as reported in (Feldt et al., 2008) to

identify that there is shortage validity in personality-

software associations, and (Gorla and Lam, 2004)

when describing that lack guidance for the IT depart-

ment regarding the selection of personnel. In other

words, the personality often interferes on the software

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

Table 3: Overview of Studies.

MODEL ID METHODS/PARTICIPANTS LOCAL

MBTI 01 Survey / 100 developers and students U.S.A.

02 Survey with 92 students China

04 Experiment with 55 students U.S.A.

07 Experiment with 22 students U.S.A.

08 Experiment with 1.471 students France

12 Experiment with 71 students Pakistan

16 Survey with 52 developers and 18 instructors Pakistan

17 Experiment with 63 developers Ireland

18 Survey (not portrayed) South Africa

BIG 5 03 Experiment with 29 students U.S.A.

05 Experiment with 60 developers and 68 students U.S.A.

09 Experiment with 105 students Spain

10 Survey with 47 developers Sweden

14 Experiment with 196 developers United Kingdom

15 Experiment with 137 students New Zealand

19 Survey with 200 developers Australia

20 Survey with 66 students India

21 Survey with 797 professionals U.S.A.

FFM 06 Survey with 143 developers U.S.A.

11 Experiment with 54 students New Zealand

13 Experiment with 453 students New Zealand

development simply because the developer was not

analyzed properly at the time of hiring. Some authors

say that while personality affects job satisfaction, its

inﬂuence on software development still remains un-

clear (Mourmant and Gallivan, 2007) (Salleh et al.,

2011).

The MBTI model was applied to individuals

(Capretz, 2003), to groups (Peslak, 2006), to students

(Rutherfoord, 2006), to professional software devel-

opers’ (Gorla and Lam, 2004) and software quality

(Barroso et al., 2016) . Most of the analyzed studies

compared the effect of developers personality in rela-

tion to job satisfaction, individual forms of work and

teamwork.

The effect of personality traits of students as-

sessed with MBTI, as well as their effectiveness at ex-

ploratory testing is discussed in (Shoaib et al., 2009).

The authors concluded that extrovert personality traits

are positively correlated with effective exploratory

testing. Exploratory testing is a specialized testing

technique. Thus, based on the ﬁndings of this study,

while it can be predicted that extraverts can be good

exploratory testers, whether extraverts will be good

testers in general remains an open question.

There is also research performed on analysing

the personality traits of software engineers in gen-

eral. Authors of paper (Hannay et al., 2010) analysed

ﬁve research studies conducted from 1985 to 2010

using MBTI to determine the personality types of

software engineers. Combined results indicate think-

ing and judging type assessed with MBTI were over-

represented among software engineers. A systematic

literature review on personality research with soft-

ware engineers published in (Feldt et al., 2008) found

the majority of personality research examined pair

programming and team effectiveness, and MBTI was

used in most of the research.

Two studies compared personality types between

industry and academy (Raza, A and Capretz, L F,

2012) (Yilmaz and O’Connor, 2012) and concluded

that the most common types of personalities are INTJ,

ISTJ, ESTP and ESTJ (Fig. 2).

Studies such as (Gorla and Lam, 2004) and

(Mourmant and Gallivan, 2007) identiﬁed the ISTJ

type as the most common among software develop-

ers. The Big 5 model, which has gained prominence

in software engineering research in recent years, the

model has been applied to individuals and teams

(Gomez and Acuna, 2007). Researchers use the Big

5 model to analyze cooperation between software de-

velopers and to examine pair programming (Chao and

Atli, 2006) (Hannay et al., 2010) (Salleh et al., 2011).

These studies, however, showed contradictory inﬂu-

ence of personality in relation to performance, while

study (Salleh et al., 2011) claim that certain personal-

ity traits, such as satisfaction, signiﬁcantly affect the

developer’s performance. In addition, studies (Chao

and Atli, 2006) and (Hannay et al., 2010) found no

correlation that, statistically, provided evidence of the

inﬂuence.

Study (Smith et al., 2016) performed a survey

about the beliefs, practices, and personalities of soft-

Inﬂuence of Human Personality in Software Engineering - A Systematic Literature Review

ware engineers in a large software company, with 797

professionals. They found no signiﬁcant personality

differences between developers and testers. Managers

tend to be more aware and extroverted.

Authors of paper (Gulati et al., 2016) studied the

relationship between students performance in Soft-

ware Engineering and their personality. They have

also found no positive evidences. FFM model has

been applied in Software Engineering research, for in-

dividuals and teams (Salleh et al., 2009) (Salleh et al.,

2010). Topics ranged from the development of new

products as reported in (Aronson et al., 2006) and

to identify the dependencies between programming

methods, as reported in (Salleh et al., 2009). A set of

experiences related to students were executed in dif-

ferent dimensions of model proposed in (Salleh et al.,

2009) (Salleh et al., 2010) and (Salleh et al., 2011). In

this case, neither the conscientiousness nor the neu-

roticism affects academic students.

As it was observed, studies using the FFM model

were performed in an academic environment, and

found no signiﬁcant effects of personality on the

performance of performed activities (Darcy and Ma,

2005) (Salleh et al., 2009) (Salleh et al., 2010) (Salleh

et al., 2011), while industry studies identiﬁed signiﬁ-

cant inﬂuences of personality.

4 THREATS TO VALIDITY

This section describes concerns that must be im-

proved in future replications of this study. To organize

this section, the threats to validity were classiﬁed us-

ing the Internal, External and Construct (Wohlin et al.,

2012).

Construct Validity:

The main constructs in this re-

view are the two basic concepts “Personality” and

“Software engineering”.

For the ﬁrst concept, we use term “Personality or

Motivation or Behavior or Humans Aspects or Satis-

faction” to make sure that all selected papers are re-

lated to term searched.

For the second concept, the terms “Software Engi-

neering, Software Development, Programming, Soft-

ware Test, Developer, Tester, ICT Manager, Project”

are used to ensure high coverage of potentially rele-

vant papers on the inﬂuence to software engineering

activities from database search.

A complementary manual search was discard due

to the fact there are not conferences and journals

speciﬁcally focused on the joint use of these concepts.

This threat is partially mitigated by including the gen-

eral intervention term “personality” along with “Soft-

ware engineering” in the terms for the search in three

reputable databases.

Internal Validity: Some subjective decisions may

have occurred during selection and data extraction.

Some papers did not provide a clear description or

proper objectives and results, making difﬁcult the ob-

jective application of the inclusion and exclusion cri-

teria. To minimize selection and extraction system-

atic mistakes, the processes was performed by four

reviewers, and any conﬂicts were discussed and re-

solved by all the authors.

External Validity: The search process described in

Section 2.1 was deﬁned after several trial searches and

validated with the consensus of all authors. We tested

the coverage and representativeness of retrieved stud-

ies, including automatic database search and referral

scanning.

5 CONCLUSION AND FUTURE

WORKS

This paper helps to synthesize available evidence in

the literature on the study of the inﬂuence of hu-

man personality in software engineering and can help

academia and industry in their daily work. We have

found that there are at least three theoretical psy-

chological models to identify human personality that,

over the past ﬁfteen years, have been applied to stud-

ies that analyze human personality of students, devel-

opers and instructors of Software Engineering.

Currently, based on results provided by this SLR

study, a controlled experiment has been performed

with students in order to assess the inﬂuence of hu-

man personality in software developed by them. The

idea is to identify if the human personality inﬂuences

quality of software. The MBTI and Big Five models

found in this SLR have been used in order to assess

the psychological proﬁle, and to assess the quality of

software products using object-oriented metrics.

Upcoming works can search other repositories,

and other keywords could be added to the search

string. Moreover, it is worth highlighting that SLR

conduction is not a trivial task because of the amount

of papers that must be read. Besides that, relevant pri-

mary studies written in languages other than English

were disregarded in this research.

ACKNOWLEDGEMENTS

The authors would like to thank the Brazilian re-

search agency CNPq (grant 445500/2014-0) for ﬁnan-

cial support.

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

REFERENCES

Aronson, Z. H., Reilly, R. R., and Lynn, G. S. (2006). The

Impact of Leader Personality on New Product Devel-

opment Teamwork and Performance: The Moderating

Role of Uncertainty. Journal of Engineering and Tech-

nology Management, 23(3):221–247.

Barroso, A. S., Madureira, J. S., Melo, F. S., Souza, T. D. S.,

Soares, M. S., and do Nascimento, R. P. C. (2016). An

evaluation of inﬂuence of human personality in soft-

ware development: An experience report. In 8th Euro

American Conference on Telematics and Information

Systems (EATIS), pages 1–6.

Brooks, Jr., F. P. (1987). No Silver Bullet Essence and Acci-

dents of Software Engineering. Computer, 20(4):10–

19.

Capretz, L. F. (2003). Personality Types in Software En-

gineering. International Journal of Human-Computer

Studies, 58(2):207–214.

Chao, J. and Atli, G. (2006). Critical Personality Traits in

Successful Pair Programming. In Agile Conference,

pages 88–93.

Darcy, D. P. and Ma, M. (2005). Exploring Individual

Characteristics and Programming Performance: Im-

plications for Programmer Selection. In Proc. of the

38th Annual Hawaii Int. Conf. on System Sciences

(HICSS), pages 314a–314a.

Feldt, R., Torkar, R., Angelis, L., and Samuelsson, M.

(2008). Towards Individualized Software Engineer-

ing: Empirical Studies Should Collect Psychometrics.

In Proc. of the Int. Work. on Coop. and Human As-

pects of Software Eng., pages 49–52.

Ferreira, Paula G and Silva, F (2008). Fatores Humanos

que Inﬂuenciam a Utilizacao de Processos de Soft-

ware. VII Simposio Brasileiro de Qualidade de Soft-

ware(SBQS), pages 123–138.

Ferreira, Vito and Natasha, N (2014). The Correlation Be-

tween Personality Type and Individual Performance

on an ICT Project. In 9th International Conference on

Computer Science & Education (ICCSE), pages 425–

430.

Goldberg, L. R. (1990). An Alternative Description of Per-

sonality: The Big-Five Factor Structure. Journal of

Personality and Social Psychology, 59(6):1216.

Gomez, M. and Acuna, S. T. (2007). Study of the Rela-

tionships Between Personality, Satisfaction and Prod-

uct Quality in Software Development Teams. In Proc.

of the 19th Int. Conf. on Software Engineering and

Knowledge Engineering(SEKE), pages 292–296.

Gorla, N. and Lam, Y. W. (2004). Who Should Work With

Whom?: Building Effective Software Project Teams.

Communications of the ACM, 47(6):79–82.

Gulati, J., Bhardwaj, P., Suri, B., and Lather, A. S. (2016).

A Study of Relationship Between Performance, Tem-

perament and Personality of a Software Programmer.

In ACM SIGSOFT Software Engineering Notes, pages

1–5.

Hannay, J. E., Arisholm, E., Engvik, H., and Sjøberg,

D. I. (2010). Effects of Personality on Pair Program-

ming. IEEE Transactions on Software Engineering,

36(1):61–80.

Kanij, T., Merkel, R., and Grundy, J. (2015). An Empirical

Investigation of Personality Traits of Software Testers.

In 8th International Workshop on Cooperative and

Human Aspects of Software Engineering (CHASE),

pages 1–7.

Kitchenham, B. A. (2004). Procedures for Undertaking Sys-

tematic Reviews, Joint Technical Report. Computer

Science Department, pages 1–26.

Lounsbury, J. W., Loveland, J. M., Sundstrom, E. D., Gib-

son, L. W., Drost, A. W., and Hamrick, F. L. (2003).

An Investigation of Personality Traits in Relation to

Career Satisfaction. Journal of Career Assessment,

11(3):287–307.

McCrae, R. R. and John, O. P. (1998). An Introduction to

The Five-Factor Model and Its Applications. Person-

ality: Critical Concepts in Psychology, 60:295.

Meyer, A. N., Fritz, T., Murphy, G. C., and Zimmermann,

T. (2014). Software Developers’ Perceptions of Pro-

ductivity. In Proceedings of the 22Nd ACM SIGSOFT

International Symposium on Foundations of Software

Engineering, FSE 2014, pages 19–29.

Morris, M. G. and Venkatesh, V. (2010). Job Character-

istics and Job Satisfaction: Understanding the Role

of Enterprise Resource Planning System Implemen-

tation. Mis Quarterly, pages 143–161.

Mourmant, G. and Gallivan, M. (2007). How Personal-

ity Type Inﬂuences Decision Paths in The Unfolding

Model of Voluntary Job Turnover: An Application to

IS Professionals. In Proc. of the ACM SIGMIS CPR

Conf. on Comp. Pers. Res.: the Global Inf. Tech. Work-

force, pages 134–143.

Myers, I. B., McCaulley, M. H., and Most, R. (1985). Man-

ual, A Guide To The Development And Use of The

Myers-Briggs Type Indicator. Consulting Psycholo-

gists Press.

Myers, I. B., McCaulley, M. H., Quenk, N. L., and Hammer,

A. L. (1998). MBTI Manual: A Guide To The Devel-

opment and Use Of The Myers-Briggs Type Indicator,

volume 3. Consulting Psychologists Press Palo Alto,

CA.

Norman, W. T. (1967). 2800 Personality Trait Descriptors -

Normative Operating Characteristics for a University

Population. Institute of Education Sciences.

Peslak, A. R. (2006). The Impact of Personality on Informa-

tion Technology Team Projects. In Proc. of the ACM

SIGMIS CPR Conf. on Comp. Pers. Res.: Forty Four

Years of Comp. Pers. Res.: Achievem., Challenges &

Future, pages 273–279.

Pressman, R. S. and Maxim, B. (2014). Software engineer-

ing: a practitioner’s approach. McGraw-Hill Sci-

ence/Engineering/Math.

Raza, A and Capretz, L F (2012). Do Personality Pro-

ﬁles Differ in The Pakistani Software Industry and

Academia. Int. Journal of Soft. Eng., pages 60–66.

Richardson, I., Casey, V., Mccaffery, F., Burton, J., and

Beecham, S. (2012). A Process Framework for Global

Software Engineering Teams. Information and Soft-

ware Technology, 54(11):1175–1191.

Inﬂuence of Human Personality in Software Engineering - A Systematic Literature Review

Rutherfoord, R. H. (2006). Using Personality Inventories

to Form Teams for Class Projects. In Proc. of the 7th

Conf. on Inf. Tech. Education, pages 9–14.

Salleh, N., Mendes, E., and Gru, J. (2011). The Effects

of Openness to Experience on Pair Programming in a

Higher Education Context. In 24th IEEE-CS Confer-

ence on Software Engineering Education and Training

(CSEE&T), pages 149–158.

Salleh, N., Mendes, E., and Grundy, J. (2014). Investigat-

ing the Effects of Personality Traits on Pair Program-

ming in a Higher Education Setting through a Fam-

ily of Experiments. Empirical Software Engineering,

19(3):714–752.

Salleh, N., Mendes, E., Grundy, J., and Burch, G. S. J.

(2009). An Empirical Study of The Effects of Per-

sonality in Pair Programming Using The Five-Factor

Model. In Proc. of the 3rd Int. Symp. on Emp. Soft-

ware Engineering and Measurement, pages 214–225.

Salleh, N., Mendes, E., Grundy, J., and Burch, G. S. J.

(2010). An Empirical Study of The Effects of Con-

scientiousness in Pair Programming Using The Five-

Factor Personality Model. In Proc. of the 32nd

ACM/IEEE International Conference on Software En-

gineering, pages 577–586.

Shoaib, L., Nadeem, A., and Akbar, A. (2009). An Empiri-

cal Evaluation of The Inﬂuence of Human Personality

on Exploratory Software Testing. In Proc. of the IEEE

13th Int. Multitopic Conf., pages 1–6.

Smith, E. K., Bird, C., and Zimmermann, T. (2016). Be-

liefs, Practices, and Personalities of Software Engi-

neers: A Survey in a Large Software Company. In

Proceedings of the 9th International Workshop on Co-

operative and Human Aspects of Software Engineer-

ing (CHASE), pages 15–18.

Spinellis, D. and Androutsellis-Theotokis, S. (2014). Soft-

ware Development Tooling: Information, Opinion,

Guidelines, and Tools. IEEE Software, 31(6):21–23.

Sudhakar, G. P., Farooq, A., and Patnaik, S. (2012). Mea-

suring Productivity of Software Development Teams.

Serbian Journal of Management, 7(1):65–75.

Wohlin, C., Runeson, P., Höst, M., Ohlsson, M. C., Reg-

nell, B., and Wesslén, A. (2012). Experimentation in

software engineering. Springer Science & Business

Media.

Yilmaz, M. and O’Connor, R. V. (2012). Towards The

Understanding and Classiﬁcation of The Personality

Traits of Software Development Practitioners: Situ-

ational Context Cards Approach. In 38th EUROMI-

CRO Conference on Software Engineering and Ad-

vanced Applications (SEAA), pages 400–405.

Yilmaz, M., O’Connor, R. V., and Clarke, P. (2016). Effec-

tive Social Productivity Measurements during Soft-

ware Development - An Empirical Study. Interna-

tional Journal of Software Engineering and Knowl-

edge Engineering, 26(3):457–490.

Zillig, L. M. P., Hemenover, S. H., and Dienstbier, R. A.

(2002). What do we Assess When we Assess a Big

5 Trait? A Content Analysis of the Affective, Behav-

ioral, and Cognitive Processes Represented in Big 5

Personality Inventories. Personality and Social Psy-

chology Bulletin, 28(6):847–858.

ICEIS 2017 - 19th International Conference on Enterprise Information Systems