Survey Development

The Journal of Systems & Software 192 (2022) 111408
Contents lists available at ScienceDirect
The Journal of Systems & Software

journal homepage: www.elsevier.com/locate/jss
Perceptions of the human and social factors that influence the

productivity of software development teams in Colombia: A statistical
analysis✩
∗
Liliana Machuca-Villegas a , Gloria Piedad Gasca-Hurtado b , , Solbey Morillo Puente c ,
Luz Marcela Restrepo Tamayo b
a
Facultad de Ingeniería, Universidad del Valle, Colombia
b
Facultad de Ingenierías, Universidad de Medellín, Colombia
c
Facultad de Ciencias Sociales y Humanas, Universidad de Medellín, Colombia
article info a b s t r a c t
Article history: This research aims to know if software engineering professionals consider that social and human
Received 8 April 2021 factors (SHF) influence the productivity of a work team. A survey-based study was conducted among
Received in revised form 11 May 2022 112 members of software development teams. Empirical results show professionals agree with the SHF
Accepted 11 June 2022
in the context of software development influence in the productivity of work teams. It was identified
Available online 17 June 2022
that the 13 SHFs have a weak or moderate correlation with each other. Additionally, the results
Keywords: of the exploratory factorial analysis suggest categorizing the factors into those associated with the
Software development productivity individual, those associated with team interaction, and those related to capabilities and experience.
Human factors This categorization reduced the number of items in the original questionnaire while preserving the
Social factors variability explained in the latent variables, which will require a shorter response time. Our results
Statistical analysis broaden the understanding of the SHFs that influence software development team productivity and
open up new research opportunities. Measuring the perception of these factors can be used to identify
which SHFs should be prioritized in a strategy to improve productivity. In addition, this knowledge can
help software organizations appropriately manage their development teams and propose innovative
work approaches that have a positive impact on the success of their projects.
© 2022 Elsevier Inc. All rights reserved.
1. Introduction and Salanitri, 2019; Canedo and Santos, 2019). SHFs can be con-
sidered characteristics of an individual that identify them based
Activities related to software development require work teams on their behaviors from both a social and individual perspective
with capacities associated with collaboration and cooperation (Cunha De Oliveira, 2017).
(Hernández López, 2014; Kosa and Yilmaz, 2015; Olgun et al., In turn, SHFs are important cost elements of a software project
2017; Matturro et al., 2019). In turn, the execution of tasks de- (Adolph et al., 2011; Fernández-Sanz and Misra, 2011). Conse-
mands well-honed social interaction skills among team members quently, studies of their influence on productivity in software
in order to share information accurately, debate ideas, and make development are a topic of special interest to software organiza-
timely decisions. Given these skills, it is important that the work tions (de Barros Sampaio et al., 2010; Wagner and Deissenboeck,
team is well-integrated and cohesive so that it develops quality 2019). The study of personal characteristics and human behavior
products in accordance with client requirements. offers opportunities to improve productivity (Boehm, 1981).
From this perspective, social and human factors (SHFs) play Despite this, empirical investigations and research in software
an important role in software engineering and can influence the development have been oriented towards the design of tech-
productivity of the development team (Curtis et al., 1988; Ruiz niques, methods or processes, with low levels of articulation with
the social and human dimension of software development (Sharp,
2007; Pirzadeh, 2010; Adolph et al., 2011; Fernández-Sanz and
✩ Editor: Daniela Damian.
Misra, 2011; Donoso Barraza and Vega Zepeda, 2017; Capretz and
∗ Correspondence to: Universidad de Medellín, Facultad de Ingenierías,
Ahmed, 2018; Murphy-hill et al., 2019).
Medellín, Antioquia, Carrera 87 N◦ 30 – 67, 50026, Colombia.
E-mail addresses: liliana.machuca@correounivalle.edu.co
The observations mentioned above have motivated the study
(L. Machuca-Villegas), gpgasca@udemedellin.edu.co (G.P. Gasca-Hurtado), of the SHFs that influence the productivity of software develop-
smorillo@udem.edu.co (S.M. Puente), lmrestrepo@udem.edu.co (L.M.R. Tamayo). ment teams, with the aim of designing improvement strategies
https://doi.org/10.1016/j.jss.2022.111408
0164-1212/© 2022 Elsevier Inc. All rights reserved.
L. Machuca-Villegas, G.P. Gasca-Hurtado, S.M. Puente et al. The Journal of Systems & Software 192 (2022) 111408
such as new work methods, strengthening SHFs using approaches • We obtained a summarized instrument for measuring the
such as gamification. This motivation is part of a Ph.D. the- perception of the SHFs, which allows for a reduction in ap-
sis research that proposes a model based on gamification and plication time. This was achieved by optimizing the number
SHFs for the creation of such strategies (Machuca-Villegas and of items while maintaining the variance explained. In turn,
Gasca-Hurtado, 2019). In the initial phases of this study, 13 SHFs this instrument could be used as a diagnostic tool in soft-
were identified (Machuca-Villegas et al., 2020) and an instru- ware development companies by helping to identify how
ment for measuring perception of these factors was developed
the work team perceives these factors and making decisions
(Machuca-Villegas et al., 2021a).
based on the results of these perceptions.
Following this work, a survey-based study was carried out in
• The results of the perception on SHFs evidence that re-
order to determine which SHFs software developers from Colom-
spondents perceive these factors as influencing productivity,
bia consider influence the productivity of their work team. The
intention of this phase of research was to corroborate whether therefore, this helps to justify the importance of promot-
the SHFs identified in our previous studies (Machuca-Villegas ing human aspects in Software Engineering (Capretz and
et al., 2020) are also considered by professionals as factors that Ahmed, 2018). Thus, a warning signal should be generated
influence their productivity. The results are then intended to to organizations regarding these SHFs and propose initia-
inform a classification of SHFs. This classification can serve as a tives that favor their intervention. Likewise, this evidence
criterion for choosing and prioritizing the SHFs to be incorporated represents an opportunity for academia in the education of
in the design of a strategy to increase the productivity of the engineers.
software development process.
This survey-based study followed the six-step process sug- The following sections of the article present the definition of
gested by Kitchenham and Pfleeger (2008). In the first step, Set- productivity used in this research (Section 2); an overview of re-
ting Objectives, the purpose of the survey was defined, which was search related to this study (Section 3); the research methodology
to measure the perception of software development profession- (Section 4); the empirical results (Section 5) and their discussion
als about the SHFs that influence their productivity (Machuca- (Section 6). Finally, conclusions and suggested avenues for future
Villegas et al., 2021a). In the second step, Survey Design, the research are outlined (Section 7).
research design was established as cross-sectional observational
and it was decided to use web-based self-administered question-
naire (Machuca-Villegas et al., 2021a). In the third step, Survey In- 2. Productivity in software engineering
strument Development, the items to measure each SHF were es-
tablished, and the instrument was constructed (Machuca-Villegas Software productivity is commonly defined in this field and
et al., 2021a). In the fourth step Survey Instrument Evaluation, the other disciplines as the ratio between output and input within
validity and reliability of the instrument were assessed (Machuca-
the software development production process. The output is un-
Villegas et al., 2021a). The fifth step, Instrument Administration,
derstood as the quantity produced, and the inputs as the effort
and the sixth step, Data Analysis are outlined in this article.
required to create that output (Barros, 2010; Petersen, 2011;
Recognizing whether SHFs influence productivity will make
it possible to include them in new work approaches, as they Yilmaz, 2013; Gómez-Jakobsen, 2016; Cunha De Oliveira, 2017).
are considered important, thereby reducing failures in software In software engineering, productivity can be considered from
project management, reducing product cost and development different perspectives: at the development level, the user level,
time, and thus making a software company more competitive and the management level. At the development level, it regards
(Paiva et al., 2010; Cunha De Oliveira, 2017). Furthermore, under- the amount of code produced for the system (Yilmaz, 2013), or
standing and working on SHFs from the perspective of software includes aspects related to requirements, implementation, and
development team productivity is relevant for software engineer- validation (Petersen, 2011). At the user level, it refers to the
ing as one of the main aspirations of research in this field is to degree of functionality achieved for the system, represented by
raise productivity (de Barros Sampaio, 2010; Canedo and Santos, the value delivered to the user (Yilmaz, 2013; Petersen, 2011).
2019). Likewise, knowing these SHFs can help software devel- At the management level, productivity analysis is focused on
opment organizations ensure proper management of their work monetary aspects (Petersen, 2011).
teams, thereby achieving success in their projects and growing On the other hand, Wagner and Deissenboeck (2019) suggest
more competitive (Curtis et al., 1988; Wagner and Ruhe, 2008;
an integrated definition of productivity based on effectiveness
Sommerville, 2016; Ruiz and Salanitri, 2019; Franca et al., 2020).
and efficiency, which contemplates the purpose, functionality,
This research extends understanding about SHFs that influence
quality, and cost of the software. Efficiency is related to the effort
software development productivity, which can be used to im-
prove the team’s working conditions, support team management employed (input) and effectiveness to the value of the software
decision making, help to define productivity measures related for its users or customers (output).
to such SHFs (Machuca-Villegas et al., 2021), and support future Productivity in software development is a crucial subject in
research. the areas of Software Engineering (de Barros Sampaio, 2010) and
The following contributions are obtained from the results: software project management (Project Management Institute,
and IEEE Computer Society, 2013). However, there is still no
• We studied 13 SHFs in the context of software development
consensus on its definition among academics and software in-
productivity in contrast to other studies that address at most
dustry professionals (Wagner and Deissenboeck, 2019) and it is a
two. Because of our interest in measuring perception, it was
easy to involve more factors and have a complete set of difficult concept to define. Nevertheless, this research adopts the
them without being explicit on one in particular. definition of productivity as the ratio between output and input
• We propose a classification of the 13 SHFs into three cat- within the software development production (Petersen, 2011;
egories. Through this, it will help to design strategies by Yilmaz, 2013) This definition facilitated the identification process
category of factors and not individually. Therefore, it may of SHFs and productivity measures close to them. In this research,
result in a reduction of effort in the definition of the strategy respondents were allowed to handle their own concept taking
and costs. into consideration their work experience.
2
3. Related work productivity were identified and categorized under Team Dy-
namics, Company Dynamics, Team Collaboration, Access to Re-
Human factors are recognized as important aspects in the sources, Work Environment, Emotional Well-being, Proximity to
context of software engineering (Amrit et al., 2014). They have the COVID-19 virus, and Miscellaneous. The results indicate that
been studied from different application scenarios, including soft- team cooperation, activity level, well-defined goals, number of
ware development (Pirzadeh, 2010), quality management sys- tasks assigned, work feedback, and frequency of communication
tems (Sanchez-Gordón et al., 2016), in software development are the six most influential factors that positively impact produc-
teams (Matturro et al., 2015b), in the context of agile devel- tivity. However, it was found that the factors that have negatively
opment (Matturro et al., 2015a; Chagas et al., 2015), in the influenced productivity are related to accessibility to resources,
quality and productivity of software (Fernández-Sanz and Misra, the work environment and emotional well-being.
2011), in the eliciting of software requirements (Donoso Barraza In this same context, in previous studies, we have identified
and Vega Zepeda, 2017), in the software process improvement 57 SHFs that influence the productivity of software development
(Morales-Aguiar and Vega-Zepeda, 2018), in the hiring of soft- from a tertiary literature review (Machuca-Villegas and Gasca-
ware professionals (Matturro, 2013), and the productivity of soft- Hurtado, 2019), these factors have been studied from the point of
ware development (Machuca-Villegas and Gasca-Hurtado, 2019; view of organizational psychology and software engineering, as a
Machuca-Villegas et al., 2020), among others. result of this process, 13 SHFs listed in the first column of Table 1
Concerning the productivity of software development, studies were selected (Machuca-Villegas et al., 2020). For this analysis,
have been identified that present a classification of influencing the meaning of the SHFs, the relationship between them, and the
factors, some of them related to social and human factors. Oliveira complexity of their study within the context of this research were
et al. (2018a) show a tertiary literature review based on sec- taken into account.
ondary studies where they identified 35 factors that influence the Table 1 presents a comparison of the 13 SHFs with factors
productivity of software developers. The factors were classified identified in more recent studies already described above. For this
purpose, only factors related to social and human aspects have
into two categories, organizational and human. Among the lat-
been considered. It is necessary to clarify that the names of some
ter, motivation, cohesion, and communication in the team were
factors do not coincide with the names of the 13 SHFs, but their
highlighted.
definition or their affinity with SHFs has been taken into account.
A study by Fatema and Sakib (2018a,b) presents the devel-
All SHFs were mapped with at least one comparison study,
opment of a qualitative model of the productivity of agile de-
except Empathy and Interpersonal Relationships where no related
velopment teams using system dynamics. In this study, they
studies were presented. The correspondences found show that
identified 35 factors that influence productivity in agile teams and
the most prevalent SHFs are Communication, Collaboration, In-
established a model. Through the model, they sought to define the
novation, Motivation, and Capabilities and Experiences in software
interaction between the factors. They consider that a cause–effect
project management, with Collaboration being the factor mapped
relationship between the factors can help clarify and quantify
in all studies. The Commitment and Emotional Intelligence factors,
the factor’s influence so that quantitative models or formulations
by contrast, are mentioned in only one study in those reviewed.
can be constructed later. The results of this study show that the
On the other hand, the comparison studies cover only some
effectiveness of an agile team lies in the interrelationships of the
of the mapped SHFs, demonstrating that there is still a need
factors identified as communication, coordination, adaptability,
for empirical investigations that cover more social and human
feedback, leadership, and self-management.
aspects in software development teams.
In another study, Murphy-hill et al. (2019) present a set of After analyzing the aforementioned studies, it can be con-
48 factors that influence the productivity of software develop- firmed that SHFs constitute a topic of special interest for software
ers, categorized into 10 groups: practices, focus, experience, job, development companies and the scientific community at large.
work, capabilities, people, project, software, and context. Based However, these studies employ different approaches and classi-
on these factors, they consulted developers from three companies fications for the factors that influence productivity, presenting a
about their productivity. The top 10 productivity factors were generalized overview of the social and human factors. Therefore,
shown to be non-technical factors, the most prominent among this suggests a reorientation of the study of human factors and
them being: Job enthusiasm, Peer support for new ideas, and Useful their influence on productivity, in such a way that it leads to
feedback about job performance. positive results in activities related to software development. In
Similarly, in Canedo and Santos (2019) a systematic litera- this sense, we propose an empirical study on the perceptions that
ture review and a study about the factors that influence the software professionals have on the influence of SHFs on their own
productivity of software development are presented. The factors productivity. Through this study, software development teams
are grouped into four classes: Product, People, Organization, and will have a set of defined SHFs, as well as their classifications
Open-Source Projects. Under ‘People’ factors such as Experience, and the relationships that may exist between them, in a way
Skills and competences, Motivation, Team cohesion, Collaboration that means they can identify their importance in a work team
among team members, Availability of members for allocation in and strategies for improvements associated with them can be
the development team, Turnover, and Ease of communication are proposed.
included.
The study (Graziotin et al., 2018) presents results about the 4. Research method
influence of happiness and unhappiness on developers. The con-
sequences that these emotions generate from an internal and The method selected for this research was a survey-based
external perspective influence productivity, the quality of the study (Kitchenham and Pfleeger, 2008). This method aims to
code and personal well-being. Among the 10 consequences iden- gather information on the perceptions of software development
tified as outcomes of happiness are motivation, creativity, and teams about SHFs that influence productivity and compare this
collaboration. information against the different variables of the software devel-
A recent study by Ganguly et al. (2020), examined the influ- opers.
ence teleworking has had on the productivity of IT professionals The process suggested by Kitchenham (Kitchenham and
in times of COVID-19. From this scenario, 24 factors related to Pfleeger, 2008) mentions six phases to carry out the survey:
3
Table 1
Comparison of related studies.
Reference study Comparison studies
Social and human Using qualitative What predicts software Factors affecting What happens when Impact on the
factor classification of system dynamics in developers’ productivity? software development software developers are productivity of remotely
influence in the development of (Murphy-hill et al., 2019) productivity: an (un)happy (Graziotin working IT professionals
productivity in an agile teamwork empirical study (Dias et al., 2018) of
software development productivity model Canedo & Almeida Bangladesh during the
teams (Fatema and Sakib, Santos, 2019) coronavirus disease 2019
(Machuca-Villegas 2018a,b)
et al., 2020) (Ganguly et al., 2020)
Communication • Communication • ‘‘Knowledge flows • Ease of • Frequency of
adequately between the key communication communication
persons in our project’’
(Murphy-hill et al., 2019)
Empathy and
Interpersonal
Relationships
Collaboration • ‘‘Backup Behavior’’ • ‘‘People on my project are • Collaboration • Collaboration • Team cooperation
supportive of new ideas’’ among team • Team Collaboration
• ‘‘Mutual trust’’ (Murphy-hill et al., 2019) members
• ‘‘I feel positively about
• ‘‘Feedback’’ other people on my project’’
(Fatema and Sakib, (Murphy-hill et al., 2019)
2018a,b)
Team Cohesion • ‘‘Coordination’’ • Team Cohesion • The definition of team

• ‘‘Goals’’ goals
• ‘‘Team orientation’’
(Fatema and Sakib,
2018a,b)
Innovation • ‘‘Task variety and • ‘‘People on my project are • Creativity

Innovation’’ (Fatema supportive of new ideas’’
and Sakib, 2018a,b) (Murphy-hill et al., 2019)
Autonomy • ‘‘Self-management’’ • ‘‘My job allows me to • Freedom of time
(Fatema and Sakib, make decisions about what management decisions
2018a,b) methods I use to complete
my work’’ (Murphy-hill
et al., 2019)
• ‘‘My job allows me to
make my own decisions
about managing my time’’
Leadership style • ‘‘Team Leadership’’ Self-confidence

(Fatema and Sakib, Being proud
2018a,b)
Commitment • Work engagement and
perseverance
Motivation • ‘‘Motivation’’ • Motivation • Motivation
(Fatema and Sakib,
2018a,b)
Work Satisfaction • ‘‘I am enthusiastic about • Enjoying the moment
my job’’ (Murphy-hill et al.,
2019)
Emotional Intelligence • ‘‘My project resolves
conflicts quickly’’
(continued on next page)
(1) definition of the objective; (2) survey design; (3) construction 4.1. Data collection
of the instrument; (4) evaluation of the validity and reliability of
the instrument; (5) administration of the instrument and (6) anal- To collect the data, self-administered questionnaires were
ysis of the results. The first four phases were previously executed, completed via the Google Forms platform during February and
which resulted in the instrument used for the administration March 2020. The survey was completed by 112 respondents
of this survey (Machuca-Villegas et al., 2021a) and the last two ensuring the required sample size was obtained for a confidence
phases are described in this article. level of 95% and a maximum permissible error of 12 points. The
4
Table 1 (continued).
Reference study Comparison studies
Social and human Using qualitative system What predicts software Factors affecting What happens when Impact on the
factor classification of dynamics in the developers’ productivity? software development software developers are productivity of remotely
influence in development of an agile (Murphy-hill et al., 2019) productivity: an (un)happy (Graziotin working IT professionals
productivity in teamwork productivity empirical study (Dias et al., 2018) of
software development model (Fatema and Canedo & Almeida Bangladesh during the
teams Sakib, 2018a,b) Santos, 2019) coronavirus disease 2019
(Machuca-Villegas
et al., 2020) (Ganguly et al., 2020)
Capabilities and • ‘‘People who write code • Skills and competences
experiences in for my software are highly
software development capable, efficient, thorough,
process communicative, and
cooperative’’ (Murphy-hill
et al., 2019)
• ‘‘My job requires me to
use a number of complex
or high-level skills’’
• ‘‘People who work on my
software’s requirements and
design are highly capable,
efficient, thorough,
communicative, and
et al., 2019)
• ‘‘I have extensive
experience with my
software’s platform
(software stack and
hardware stack)’’
experience with the tools
and programming languages
used in my software’’
experience developing other
software similar to the one
I’m working on’’
Capabilities and • ‘‘Team Management’’ • ‘‘People who manage my • Skills and competences
Experiences in (Fatema and Sakib, project are highly capable,
software project 2018a,b) efficient, thorough,
management communicative, and
et al., 2019)
participants were professionals from Colombia working in some reducing the possibility of respondents making an error in their
capacity in software development, who received an invitation to response and increases the precision at which a concept is mea-
answer the questionnaire through the Google platform, by email, sured (Kitchenham and Pfleeger, 2008). Based on the theoretical
and social networks. The database of graduate students provided definition of SHFs, each member of the research team proposed
by sponsoring universities of this research helped to invite the items to measure the respective SHF, and an item bank was
participants and guarantee the profile of the respondents. Besides, constructed. Subsequently, we held periodic meetings to review
the personal social networks of the researchers in this study were the items and considered the relationship between the item and
used to distribute the questionnaire and facilitated increasing
the factor, as well as their wording. This process made it possible
the sample. An Excel data file was generated from the Google
to refine the item bank and select those appropriate items for the
forms, which was imported from SPSS version 21 for Windows
instrument (Machuca-Villegas et al., 2021).
to process and analyze the results.
Besides, given that the concept of productivity in Software
Engineering has different approaches and the lack of unicity of
4.1.1. Data collection instrument
criteria of a unique definition (Sadowski and Zimmermann, 2019),
A 79-item questionnaire was developed to assess the per-
each respondent was left to use his concept of productivity in
ception of the 13 SHFs, which included statements framed in
both positive and negative valence, according to Aiken’s approach broad scope, thus allowing each one to review the concept in his
(Aiken, 1996). The full version of this instrument is published activity.
in (Machuca-Villegas et al., 2021a), the collected data are avail- The content validity and reliability of the instrument were
able online (Machuca-Villegas et al., 2021b). The questionnaire evaluated. There are three types of instrument validity, namely
items were created from the definitions of SHFs we established content validity, construct validity, and related criterion validity
(Machuca-Villegas et al., 2020). For each SHF, we drafted more (Martinez Arias, 1996). The instrument used in this research was
than five items as having multiple items improves reliability by validated in two ways. First, the content validity was evaluated,
5
for this, the written items were sent to five judges who were con- Table 2
sidered experts in the area, including software engineers, a psy- General characteristics of the software professional respondents.
chologist and a professional in linguistics. Each expert rated each Variable Categories N◦ %
item on a scale of one to three points, considering pertinence, Country Canada 1 .9
relevance, and wording. Based on the experts’ assessments, the Colombia 105 93.8
Content Validity Coefficient was calculated, which made it pos- Spain 1 .9
United States 1 .9
sible to eliminate some items and make some adjustments. This Mexico 3 2.7
validation was performed before applying the instrument. Sec- Paraguay 1 .9
ond, construct validity was calculated using the Exploratory Fac-
Sex Male 85 75.9
torial Exploratory. For its computation, the answers given by the Female 27 24.1
respondents to the instrument were taken.
Educational level Undergraduate 59 52.7
The Content Validity Coefficient obtained an average value of Specialization 21 18.8
0.86, which indicates that the instrument is valid, as it is above Master’s 31 27.7
the optimal value (0.8) suggested. To evaluate reliability, Cron- PhD doctoral studies 1 .9
bach’s Alpha coefficient based on item covariances was used. A Profession Software engineer 79 70.54
value of 0.958 was obtained, indicating that the instrument is re- Engineer in fields related 17 15.18
liable (Machuca-Villegas et al., 2021a). The results of Exploratory to software engineering
Factorial Exploratory are described in Section 5.2.1. Professional in another 16 14.29
area
4.1.2. Pilot sample
The instrument was piloted with a sample of 23 representative
professionals with experience in software development in Medel- 4.2.1. Respondent profiles
lín, Colombia. The people were invited to participate through The questionnaires were answered principally by Colombian
the Software Engineering academic program of the Universidad individuals who work in Colombia (93.8%) in this country carried
de Medellín. Through that, it was possible to estimate the time out the research, followed by respondents who live in Mexico
taken by the respondent to answer the questionnaire, as well with a 2.7% share. Individuals in countries such as Canada, Spain,
as to review the organization of the results generated by the the United States, and Paraguay completed a questionnaire. All of
platform and to identify opportunities for improvement. The test them are Colombians residing in those countries. A high preva-
results showed that the average time required to answer the
lence of males was observed (75.9%), reflecting the fact that,
questionnaire is 25 min. Aspects for improvement were also
for every woman who works in the software development field,
identified in terms of the form and presentation which were used
there are three men. Regarding the academic level, more than
to adjust the final version of the instrument. Among these were
half (52.7%) of those surveyed dedicated to software development
ensuring consistency in the language used across all items of
reported holding a university undergraduate degree. The rest
the instrument, listing the factors and questions, and adjusting
reported having at least one postgraduate qualification, either at
the presentation format of the questions to make them more
the level of specialization (18.8%) or a Master’s degree (27.7%)
understandable.
and one respondent indicated having doctoral studies. Table 2
4.1.3. Sampling presents the profile characteristics of the participants.
A non-probability quota sampling technique was used due to a The professions of these developers are diverse: seven out of
sampling frame was not available. The survey was accessible until 10 developers are software engineers (70.54%), but engineers in
the sample size required to estimate an average with a confidence areas related to software engineering (15.18%) and professionals
level of 95% and a precision of 12 points was reached. The formula in other areas (14.29%) were recorded.
employed was the following (Anderson et al., 2008): This sample consisted of developers between 21 and 53 years
of age, with an average age of 32.44 ± 7.384 years. The greatest
Zα2 ∗ S 2 length of service in their place of work was 35 years, with an
= 81 respondents
e2 average length of service of 6.54 ± 7.347 years. The minimum
Where Zα2 is the value of the Z coefficient for a 95% confidence and maximum values of professional experience corresponded
level (1.96) to the number of years of service in the company, with the
average length of professional experience in the area of software
S 2 is the value of the estimated variance (3048.996) development also being similar (8.62 ± 6.654 years). Twenty-five
e2 is the maximum permissible error percent of the respondents claimed to have more than 13.75 years
A non-response percentage of 30% was assumed and a total of of professional experience but only more than 10 years of service
112 fully completed questionnaires were collected. The variance in the company. In age, as well as in years of experience and
value was obtained from a pilot study. length of service, the distribution of the data is positively skewed,
indicating that most of the cases are below the mean, with this
4.2. Data analysis pattern of distribution being more marked in the variable of the
length of service in the company. Table 3 presents the descriptive
Data analysis was carried out at the two statistical levels: de- statistics associated with age, length of service, and years of
scriptive and inferential. At a descriptive level, the characteristics professional experience.
of the study sample are shown, and the responses given by the Regarding the size of software development teams, respon-
software professionals to each of the 13 SHF items are presented dents reported that team sizes ranged from one member to 40,
in percentage terms. For each factor, a score was obtained from with an average of 9.02 ± 7.292 members per team. Regarding
the sum of the responses and quantitative variables were created. the number of projects that these companies have developed, the
At an inferential level, the correlation between the factors is variation in the range is between zero and 80 projects, with an av-
shown using the Spearman–Brown correlation coefficient with a erage of 10.98 ± 18.690 projects. Fifty percent of those surveyed
significance level of 5%. Similarly, exploratory factor analysis was affirmed that their company had developed seven projects or less
performed using the 13 SHFs calculated as quantitative variables and only one in four respondents indicated that their company
to evaluate the validity of the instrument as well as to classify the had developed more than 10.75 projects. Table 4 presents the
SHFs. respective results.
6
Table 3
Descriptive statistics of age, seniority and experience of software developers.
Standard Percentiles
Variable Mean Range Skewness
deviation 25 50 75
Age (years) 32.44 7.384 21–53 27 31 37 0.816
Years of service in a company 6.54 7.347 0–35 1 3 10 1.494
Years of professional 8.62 6.654 0–35 3 7 13.75 1.130
experience in the area of
software development
Table 4
Descriptive statistics of the number of employees, number of team members and number of projects developed by the company
where the software developers work.
Standard Percentiles
Variable Mean Range Skewness
deviation 25 50 75
No. of members in the 9.02 7.292 1–40 4 7 10.75 1.968
software development team
No. of projects developed 10.98 18.690 0–80 2 4 10 2.879
5. Empirical results their respective activities should be explicit and clear for all team
members (82.1%). The ‘‘definition of a communication protocol
This section presents and analyzes the results of the survey between team members and external personnel’’ and ‘‘each task
from a descriptive and inferential perspective. For the descriptive should have a clearly identified responsible person’’ are perceived
analysis, percentage measures were used for categorical variables with a degree of disagreement that differs from the rest of the
whereas, for the inferential analysis, quantitative variables were items that comprise the communication factor, with more than
calculated to create 13 SHFs with the relationship between them 5% of respondents favoring the alternatives Strongly disagree and
determined using the Spearman–Brown correlation coefficient. Disagree. Table 5 presents the results in percentage of responses
Content validity and construct validity were verified to comply to the communication factor items.
with the instrument’s evaluation requisites. The content validity
was performed before administering the instrument, and in this 5.1.2. Commitment
case, it was carried out through expert judgment, and in Sec- Commitment is the level of individual responsibility that a
tion 4.1.1 the results were mentioned. The construct validity was person assumes to carry out their activities as part of a work
calculated from the results obtained from the application of the team. In this sense, it also refers to how the work team is col-
instrument to determine the underlying structure of the set of lectively responsible for the goals set within the project. The
items under each factor; in this research, it was determined by responses of software developers to the items under the factor
exploratory factor analysis (EFA). of commitment reveal that all respondents agree to some ex-
In the rotated components matrix of the EFA, when the factor tent that all team members should assume responsibility for the
loadings weight of the item is similar, it is convenient to analyze it results obtained, fulfill their duties and be able to admit their
to determine whether it belongs to a factor. Consequently, it may mistakes to improve the productivity of the software develop-
result in the elimination of an item since it cannot be established ment process. For the remainder of the items, low percentages
to which of the factors it belongs. are observed in the disagreement categories. However, it is worth
highlighting that more than 10% indicated a certain level of dis-
5.1. Descriptive analysis: Perceptions of the social and human factors agreement with the idea that, to improve the productivity of the
influencing software development productivity software development process, team members should fully and
punctually complete tasks assigned to them.
An analysis of the items was performed for each SHF, using Software developers do agree that following aspects improve
frequencies and percentages to visualize the level of agreement the productivity of the software development process: carrying
that software developers had towards the proposed items listed out the necessary tasks according to the objectives set; having
that influence the productivity of teams. The responses were a level of responsibility that enables them to take on tasks in
analyzed to determine whether they agree or disagree with the favor of the objectives of the work team; having clarity about
items proposed and the most prevalent among the complete set responsibility regarding the completion of tasks and being willing
of responses is highlighted. to help when required; Table 6 presents the results in percentage
of responses to the commitment factor items.
5.1.1. Communication
Communication refers to how one person relates to another. 5.1.3. Motivation
In software development teams, the forms of communication According to its definition, motivation moves a person towards
facilitate the proper flow of information among its members, action. The motivations of the individual can be uncountable and,
improving project results. When inquiring among software de- according to Herzberg (Herzberg et al., 1959) is classified as in-
velopers about aspects such as the clarity of instructions and trinsic and extrinsic motivations. The highest degree of disagree-
objectives, the avoidance of repeating processes, the improvement among those surveyed was registered around the statement
ment of tasks and activities, the correct resolution of conflicts, that the tasks assigned to team members should be perceived as
among others, there was a tendency by respondents to be in interesting and challenging to improve the productivity of the
total agreement with the items related to this factor, particularly software development process (13.4%). The other items under
where it is stated that to improve the productivity of the software motivation were perceived with high percentages of agreement,
development process, communication among team members is either strongly or partially. Thus, all perceived that it is encour-
an important support (87.5%) and that the project objectives and aging for team members to receive rewards for the activities
7
Table 5
Responses by software developers to the items under the factor Communication (in %).
Communication Level of agreement (%)
to improve productivity in the software development process. . . Strongly disagree Disagree Agree Strongly agree
The communication among team members is an important – – 12.5 87.5
support.
The project objectives and their respective activities should be – – 17.9 82.1
explicit and clear for all team members.
Each task should have a clearly identified responsible person. 1.8 3.6 33.9 60.7
Team members should maintain fluid communication. 0.9 – 24.1 75.0
Team members should be informed in a timely manner about the 0.9 – 35.7 63.4
progress of goals and achievement of objectives.
It is necessary to define a communication protocol between team 0.9 4.5 44.6 50.0
members and external personnel.
Table 6
Responses by software developers to the items under the factor Commitment (in %).
Commitment Level of agreement (%)
To improve productivity in the software development process. . . Strongly disagree Disagree Agree Strongly agree
The team should carry out the tasks necessary for the success of – 0.9 38.4 60.7
the project in accordance with the objectives set.
Team members should have a level of responsibility that enables – – 34.8 65.2
them to take on tasks in favor of the objectives of the team.
the team should be clear about its responsibility to fulfill the 1.8 – 32.1 66.1
tasks set and be willing to help when required.
All team members should take responsibility for the results – – 27.7 72.3
obtained, fulfill their duties, and be able to admit their mistakes.
Team members should fully and punctually fulfill their assigned 1.8 11.6 41.1 45.5
tasks.
Table 7
Responses by software developers to the items under the factor Motivation (in %).
Motivation Level of agreement (%)
Positive attitudes are essential and a product of achieving – 6.3 38.4 55.4
objectives.
Tasks assigned to team members should be perceived as 0.9 12.5 49.1 37.5
interesting and challenging.
Team members should feel that the tasks they perform are – 8.9 45.5 45.5
valuable in achieving the objectives.
It is encouraging for team members to receive rewards for their – – 28.6 71.4
activities.
Good furniture, computer equipment and optimal working – 2.7 29.5 67.9
conditions should be provided.
they carried out, 93.8% considered positive attitudes essential that the tasks assigned to each team member should correspond
and the result of achieving objectives, and nine out of 10 of to a large degree with what each member wants to do as a means
the respondents agreed that team members should feel that the to improve the productivity of the software development process
tasks they perform are valuable in achieving the objectives. In (41.1%). By contrast, it was found that almost all respondents
summary, the items under the motivation factor are perceived by agreed in their perception that team members should feel satis-
most developers as elements that should be taken into account fied with the possibility of acquiring additional knowledge about
if the productivity of the software development process is to software development, as well as agreeing that tasks should be
be improved. Table 7 presents the results in the percentage of assigned according to team members’ profiles to improve pro-
responses to the motivation factor items. ductivity (96.5% each). In summary, based on the perceptions of
the respondents, it can be stated that work satisfaction tends to
5.1.4. Work satisfaction be expressed through the distribution of tasks and equitable dis-
Work satisfaction is determined by the discrepancy between tribution of work, the acquisition of knowledge, and satisfaction
what one wants and what one has in a job. It is derived from with the activities each individual performs. Table 8 presents the
the content of one’s work, or the challenging and stimulating responses in percentage to the work satisfaction factor items.
activities of the position. Among the aspects that enable job satis-
faction to be measured are those related to expectations, personal 5.1.5. Emotional intelligence
and professional growth, recognition, opportunities, salary, work Table 9 shows the responses given to the items under Emo-
environment, and relationships with colleagues and superiors. tional Intelligence, understanding this factor to be the ability of
The responses to the items under the work satisfaction factor an individual to appropriately identify and process emotions and
differ from the trend observed in the previous factors, with four not be dominated by them; that is, an individual who can be in
out of 10 respondents stating that they disagreed to some extent control of their behavior. The results reveal a tendency to agree
8
Table 8
Responses by software developers to the items under the factor Work Satisfaction (in %).
Work Satisfaction Level of agreement (%)
The contribution of each team member should be recognized. – – 36.6 62.5
The activities carried out by team members should contribute to – 7.1 44.6 48.2
their personal and professional growth.
Team members should feel satisfied that the distribution of work 0.9 7.1 49.1 42.9
is equitable.
Team members should feel satisfied with the activities they – 5.4 51.8 42.9
perform.
The tasks assigned to each of the team members should 1.8 39.3 44.6 14.3
correspond to a large degree with what each one wants to do.
Tasks should be assigned according to the profile of each team 0.9 3.6 50.9 44.6
member.
Team members should be satisfied with the possibility of 0.9 2.7 42.9 53.6
acquiring additional knowledge about software development.
Table 9
Responses by software developers to the items under the factor Emotional Intelligence (in %).
Emotional Intelligence Level of agreement (%)
Each of the team members should be able to adapt to the – 4.5 50.9 44.6
changes inherent in the project.
Each of the team members should express their disagreement in – 1.8 38.4 59.8
a timely manner and to the appropriate person.
Team members are required to carry out their activities, even 0.9 14.3 50.0 34.8
when they get difficult.
It is necessary that team members know how to handle their 1.8 2.7 47.3 48.2
emotions appropriately.
It is important that each of the team members listen to criticism – 0.9 48.2 50.9
and respond appropriately in a reasoned manner.
The work team should have the ability to resolve conflicts – 0.9 39.3 59.8
appropriately.
Team members should know how to recognize the emotional 0.9 17.9 49.1 32.1
states of their colleagues and act empathically.
Team members should build relationships based on trust and – 3.6 29.5 67.0
respect.
with the statements listed, particularly when responding to the total agreement regarding collaborative work, as well as the will-
statement of the importance for each team member to listen ingness they should show to help and support their colleagues,
to criticism and respond reasonably, and that the work team such as with actions that would help to improve the productiv-
should have the ability to resolve conflicts appropriately (99.1%). ity of the software development process. Similarly, respondents
A similar level of agreement was recorded for the statement that agreed to some extent on items such as trust and knowledge
each of the team members should express their disagreement in sharing. Overall, there is a general perception in the responses
a timely manner and to the appropriate person (98.2%). These that collaboration, that is, the feeling of having the backing and
responses have in common the search for a resolution to conflicts support of others, which leads to joint action and implies a spirit
or problems based on speaking and listening. of solidarity, are important to take into account in the software
However, almost two in 10 respondents, (18.8%), were in development process if productivity is to be improved. Table 10
disagreement, either strongly or partially, with the statement that presents the responses in percentage to the collaboration factor
it is important for productivity that team members know how items.
to recognize the emotional states of their colleagues and be em-
pathetic. Similarly, there is disagreement around the statement 5.1.7. Team cohesion
that team members should continue with their work activities, The factor Team Cohesion was conformed of six items based
even if they get difficult (15%). These results stand out from the on the premise that feeling part of a team, through autonomous
rest of the responses, as they differ from the observed trend of and motivated participation, is important for its cohesion. Team
almost total agreement with the statements about adaptation cohesion refers to the sense that all efforts revolve around a
to change, conflict resolution, listening skills, and the timely common goal. Of the six items, four follow a trend of agreement
expression of problems, trust, and mutual respect — in short, on the part of the respondents, highlighting that more than 90%
statements that refer to the quality of the relationships between reflected some degree of agreement with the statements that, to
team members, based on the recognition of the emotional state improve productivity, each team member should put his or her
of their colleagues. best skills at the service of the project objectives (98.2%) and
that they should also have a feeling of identification with the
5.1.6. Collaboration team by participating in an autonomous and motivated manner
The responses to the items under the factor Collaboration (95.5%). Similarly, a high percentage agreed that activities should
show a clear trend in the perceptions of the respondents towards be executed in a timely manner and with the participation of
9
Table 10
Responses by software developers to the items under the factor Collaboration (in %).
Collaboration Level of agreement (%)
Team members should work collaboratively to achieve project – – 30.4 69.6
goals.
There should be trust between team members for the – 1.8 44.6 53.6
performance of their duties and the protection of common
interests.
Team members should be willing to help and support each other. – – 37.5 62.5
It is necessary for each team member to share their knowledge, – 2.7 34.8 62.5
information and experience with their peers.
Table 11
Responses by software developers to the items under the factor Team Cohesion (in %).
Team Cohesion Level of agreement (%)
It is necessary that team members work at a similar pace. 4.5 37.5 42.0 16,1
Members should feel identified with the team by participating in 1.8 2.7 48.2 47.3
an autonomous and motivated manner.
Each team member should put his or her best skills at the – 1.8 44.6 53.6
service of the project’s objectives.
Each team member should enjoy carrying out tasks with his or 0.9 11.6 53.6 33.9
her teammates.
Activities should be executed in a timely manner and with the 0.9 7.1 53.6 38.4
participation of all responsible parties.
Each member should know what each team member is doing. 2.7 17.9 47.3 32.1
all those responsible (92%). It is also noteworthy that almost Similarly, they recognize the importance of receiving training in
nine out of ten respondents agreed that productivity improves if topics aimed at improving interpersonal relationships to manage
members enjoy accomplishing tasks with their colleagues (87.5%). emotions effectively and to work as a team (91.1%). One in four
These results reveal the importance that members of software respondents did not agree that participation in social activities
development teams give to integration, and the distribution of inside and outside of the work environment improved productiv-
responsibilities among members for decision making, which is ity (23.2%). Table 12 presents the responses in percentage to the
likely to reduce errors in the execution of their work. items under the empathy and interpersonal relationships factor.
The highest levels of disagreement corresponded to the propo-
sitions that all members should work at a similar rhythm or 5.1.9. Leadership
pace (42.2%) and should know what each of their colleagues is The factor Leadership is understood as the ability of some
doing (20.6%). Such results suggest that software developers don’t people to influence the work team in order to achieve goals and
strongly associate working speed nor working collectively with
objectives and it was summarized in the seven items presented in
productivity. Table 11 presents the responses in percentage to the
Table 13. The results show that one in four respondents disagreed
team cohesion factor items.
with the statement that any team member has the qualities to
lead the activities within a project, thereby acknowledging that
5.1.8. Empathy and interpersonal relationships
a leader should have specific characteristics that the literature
Empathy is a mental state in which an individual identifies
highlights which makes him/her stand out within the group,
with another, being able to put themselves in their position and
such as self-confidence and confidence in their own convictions,
feel what the other person feels. One of the items under Em-
decision-making abilities, and the courage to take significant
pathy and Interpersonal Relationships was posed with negative
valence according to Aiken’s approach (Aiken, 1996), meaning risks. There was general agreement on two characteristics of a
the behavior of this item should be interpreted inversely. This is leader which influence the productivity of the software devel-
the item that states that members of a development team can opment process: the first was that they should promote positive
have little or no relationship with each other as long as they do attitudes and generate trust among team members, and the sec-
their work, a stance that prioritizes work tasks over interpersonal ond that they should coordinate and guide the team’s activities
relationships. The results show that respondents were inclined towards the project’s objectives and goals.
to choose the options of disagreement (52.7%) indicating that In addition, most of the respondents (97.3%) agreed that a
for them personal interrelationships are indeed important. The leader should be concerned about both the results of the project
remaining eight items are characterized by a high prevalence and human relations, thus highlighting that leader should have
of agreement, with almost all respondents (99.2%) considering the ability to lead the tasks and objectives that he/she proposes
that everyone should try to complete the tasks set, provide a and develop strong relationships with others. Similarly, 94.6%
pleasant work environment and respect the working agreements consider that a leader should be fair in their treatment towards
established by the team, as well as ensure the existence of good members of the software development team and in the demands,
personal relationships (93.8%). they make and that the members should feel that their leader can
Empathy was also highly favored as an aspect that contributes offer solutions to problems within the project. All the answers
to the productivity of the software development process with underline the importance given to leadership as a factor that
98.2% of respondents agreeing that team members should put helps to improve the productivity of the software development
themselves in the place of others and help them when needed. process.
10
Table 12
Responses by software developers to the items under the factor Empathy and Interpersonal Relationships (in %).
Empathy and Interpersonal Relationships Level of agreement (%)
It is beneficial for its members to participate in social activities, 2.7 20.5 50.9 25.9
both inside and outside of the work environment.
Members may have little or no relationship with each other, as 8.9 43.8 30.4 17.0
long as they do their work.
Each member of the team should recognize that failure to – 0.9 43.8 55.4
perform their tasks can affect the performance of their
teammates and the team as a whole.
It is necessary that team members receive training in 2.7 6.3 49.1 42.0
interpersonal relations, effective management of emotions,
teamwork and quality.
There should be good personal relationships among members. 0.9 5.4 53.6 40.2
Team members should be able to take the place of the other – 1.8 54.5 43.8
when you need help and collaborate to meet your need.
Each team member should participate in the activities carried out – 13.4 62.5 24.1
in his or her area of work.
Team members should provide a pleasant working environment. – 0.9 42.9 56.3
Team members should respect the collectively established – 0.9 42.9 56.3
working agreements.
Table 13
Responses by software developers to the items under the factor Leadership (in %).
Leadership Level of agreement (%)
Any team member can have the qualities required to lead the 2.7 22.3 51.8 23.2
activities within the project.
Each team member should feel that they can offer solutions to 0.9 4.5 48.2 46.4
problems within the project.
The leader should promote positive attitudes and build trust – – 37.5 62.5
among members of the software development team.
Work-related decisions should be made through group 0.9 8.0 50.0 41.1
discussions and not unilaterally.
The leader should coordinate and guide the team’s activities – – 39.3 60.7
towards the project’s objectives and goals.
A leader should be fair in their treatment of and demands made 0.9 4.5 33.0 61.6
on members of the software development team.
The leader should be concerned about both project results and – 2.7 35.7 61.6
interpersonal relationships.
5.1.10. Innovation of something new, different and of a certain value, from the
Following the recommendations of (Aiken, 1996), one of the experiences and knowledge that the creator possesses. This def-
items under the factor Innovation was constructed with negative inition is reflected in the perceptions expressed by respondents,
valence, forming the statement, ‘‘the use of solutions that have as presented in Table 14.
not been satisfactorily tested should be avoided’’ to which 34%
disagreed. The behavior of this item should be interpreted in the 5.1.11. Autonomy
inverse sense, that is, it indicates that six in nine respondents Five items constitute the Autonomy factor, which is defined
consider that untested solutions cannot be used to increase the as the power to make decisions in the work environment in-
productivity of the software development process. These results dependently of those in management. Among the respondents
suggest, therefore, that respondents do not consider this idea an (Table 15), there was a high prevalence of responses at the
aspect of innovation. The rest of the items under the innovation extremes reflecting agreement with the propositions included in
factor showed a tendency towards a high degree of agreement the items, ranging from 99.2% in agreement that members can
with 97.4% supporting the idea that members should be able organize themselves to establish and meet their objectives, and
to take on new challenges and develop different skills. Another that they should trust their abilities to perform the tasks they are
high percentage, (96.4%), agree that to innovate and improve responsible for, followed by 89.2% who agreed that members can
productivity in the development process, customer suggestions make decisions about methods, techniques, and strategies, among
or complaints should be taken into account, and also that the others, to perform the tasks. This is related to the level of freedom
leader should encourage members to put their own ideas into that the employee and the work team have to make decisions
practice and find new ways of dealing with problems. In addition, related to a project and the way they work on it. Autonomy
the incorporation of innovation in projects as part of a company’s also encompasses the freedom to decide what to work on, which
policies is understood by the respondents (95.6%) as support for projects to investigate, which ideas to develop, how to solve
the improvement of the software development process and that problems, and how to adapt the work to the project without
the members should be receptive to new ideas, supporting them. interference from agents outside the team. The sample shows that
It is worth highlighting here that, according to theoretical princi- 85.7% of respondents agree, to some extent, that team members
ples, innovation is related to creativity that leads to the creation can take the initiative in making corrective measures. In addition,
11
Table 14
Responses by software developers to the items under the factor Innovation (in %).
Innovation Level of agreement (%)
And in order to innovate, customer suggestions, complaints 0.9 2.7 46.4 50.0
and/or claims should be taken into account.
Company policies should encourage the incorporation of – 4.5 42.0 53.6
innovation in projects.
The use of solutions which have not been satisfactorily tested 5.4 28.6 44.6 21.4
should be avoided.
Its members should have the opportunity to take on new – 2.7 42.0 55.4
challenges and develop diverse skills.
Software development team members should be supportive and 0.9 3.6 43.8 51.8
receptive to new ideas.
Leaders should encourage members to put their own ideas into – 3.6 50.0 46.4
practice and find new ways of resolving problems.
Table 15
Responses by software developers to the items under the factor Autonomy (in %).
Autonomy Level of agreement (%)
Team members should be empowered to make decisions 1.8 17.0 54.5 26.8
regarding the project and their way of working within it.
Team members should be able organize themselves to establish – 0.9 56.3 42.9
and meet their objectives.
Team members should be able to make decisions about methods, – 10.7 57,1 32,1
techniques and strategies, among others, to perform their tasks.
Team members should have confidence in their abilities to – 0.9 50.0 49.1
perform the tasks for which they are responsible.
Team members should be able take corrective action on their – 14.3 52.7 33.0
own initiative.
81.3% agree that members should be allowed to make decisions should adequately manage agreed time frames (96.3%) and that
regarding the project and the way they work on it. they should have the ability to plan, execute or control project
activities (84.8%).
5.1.12. Capabilities and experience in the software development pro-
cess 5.2. Inferential analysis — classification of perceptions of social and
The factor Capabilities and Experience in the software devel- human factors (SHFs) influencing software development productivity
opment process refers to the knowledge of and experience in the
analysis, design, development, and testing of a software product For the inferential analysis of the data, an Exploratory Factor
according to the role exercised by each member of the work Analysis of items (EFA) was first performed, which is one of the
team. It implies familiarity with the application domain, with most widely used techniques to develop and validate surveys,
software platforms, and with programming tools and languages. because according to Lloret-Segura et al. (2014) it is the tech-
Seven items were used to measure this factor. nique par excellence used to explore a set of latent variables or
Except for one (0.9%), the respondents agreed that the mem- common factors that explain the responses to the items of a test.
bers should stay up to date in the use of tools and practices
In this case, the 13 SHFs were first calculated from the sum of
to perform their tasks. Table 16 also shows that a high per-
the responses to the items under each factor, thus generating
centage (97.3%) consider that, in order to improve the software
13 quantitative variables that were included in the model to
development process, logical reasoning and systemic thinking
determine the components that emerged and that served, on the
are important skills that team members should have, as well as
one hand, for the validation of the construct and, on the other
the ability to implement efficient solutions that meet project re-
hand, to classify the SHFs.
quirements and also have knowledge or experience in the use of
software development methodologies (92.9%). It was notable that
almost three in 10 respondents (28.7%) disagree that knowledge 5.2.1. Exploratory factor analysis of SHFs
of the subject or experience in similar contexts contributes to the The sum of each SHF is calculated from the items that con-
improvement of the software development process. stitute it using a Likert scale ranging from 1 point (Strongly
Disagree) to 4 points (Strongly Agree). Table 18 presents the
5.1.13. Capabilities and experience in software project management descriptive statistics of the 13 SHFs analyzed. In general, the
The factor Capabilities and Experience in software project theoretical fluctuation and the empirical tend to coincide at the
management is the only factor in which responses for all items maximum point, but diverge at the minimum point, except the
were recorded across all four levels of agreement, as can be seen Commitment factor in which the perception scores did agree with
in Table 17. There is a tendency towards an agreement with the the empirical values. Similarly, the distribution of the data in all
items although some disagreement was recorded. For example, factors is negatively skewed, indicating that most of the data are
one in four (28.5%) did not agree that the members of software above the group mean. All averages were in the upper third of
development teams should have knowledge or experience in the the scale.
use of project management tools and techniques or in the use In order to determine whether the 13 factors could be classi-
of metrics to monitor the project (23.3%) to improve produc- fied, an EFA was performed. The factors are quantitative variables
tivity. On the other hand, the majority agreed that members created from the sum of the responses given by respondents to
12
Table 16
Responses by software developers to the items under the factor Capabilities and Experience in the software development process (in %).
Capabilities and Experience in the software development process Level of agreement (%)
To improve the productivity of the software development process, team members should... Strongly disagree Disagree Agree Strongly agree
Keep up to date with the best tools and practices for the execution of tasks. 0.9 46.4 52.7
Have knowledge of the subject or have worked in similar contexts. 0.9 27.7 50.0 21.4
Have knowledge or experience in the use of the programming tools and languages 9.8 48.2 42.0
required for the project.
Have knowledge or experience in the analysis, design, construction or implementation of 0.9 10.7 46.4 42.0
software.
Have logical reasoning and systemic thinking skills. 0.9 1.8 44.6 52.7
Have the ability to implement efficient solutions that meet project requirements. 2.7 47.3 50.0
Have knowledge or experience in the use of software development methodologies. 7.1 52.7 40.2
Table 17
Responses by software developers to the items under the factor Capabilities and Experience in software project management (in %).
Capabilities and Experience in software project management. Level of agreement (%)
Team members should adequately manage the time frames 0.9 1.8 51.8 45.5
agreed.
Team members should have the ability to plan, execute or 0.9 14.3 48.2 36.6
control project activities.
Team members should have knowledge or experience in the use 3.6 25.9 46.4 24.1
of project management tools and techniques to improve
productivity.
Team members should have knowledge or experience in the use 4.5 18.8 46.4 30.4
of metrics that allow the project to be monitored.
Table 18
Summary of the SHF descriptive statistics.
Factor Theoretical Mean Standard Empirical range Skewness
range deviation
Communication 6–24 22.02 2.07 17–24 −.866
Commitment 12–20 17.90 2.06 12–20 −.688
Motivation 5–20 17.46 1.99 11–20 −.562
Work Satisfaction 7–35 23.34 2.75 17–28 −.070
Emotional Intelligence 8–32 27.44 3.12 20–32 −.184
Collaboration 4–16 14.44 1.54 10–16 −.663
Team cohesion 6–24 19.21 2.88 11–24 −.276
Empathy and Interpersonal Relationships 9–36 29.23 3.30 19–36 −.121
Leadership 7–28 24.05 2.71 15–28 −.657
Autonomy 5–20 16.37 2.17 13–20 .405
Innovation 6–24 19.55 2.15 14–24 −.340
The software development 6–24 20.02 2.47 14–24 −.153
Capabilities and experience in. . .
process
Software project 4–16 12.56 2.31 6–16 −.287
management
the items that make up each factor. For a preliminary analysis, Table 19
the complete set of 13 factors was subjected to an EFA, using the Communalities.
principal component extraction method and with Varimax nor- SHF Initial Extraction
malization as the rotation method. Similarly, a reliability analysis Communication 1.000 .730
of the 13 SHF scales was carried out, resulting in high internal Commitment 1.000 .789
Motivation 1.000 .574
consistency (Alpha = 0.918), indicating that they are reliable and Job Satisfaction 1.000 .574
all factors are necessary to form the construct ‘Perceptions of Emotional Intelligence 1.000 .631
software development team members about the SHFs that affect Collaboration 1.000 .718
their productivity’. In the factor analysis, all the SHFs entered had Team Cohesion 1.000 .682
loading values greater than 0.4, resulting in none being discarded Empathy and Interpersonal Relationships 1.000 .732
Leadership 1.000 .679
(Table 19). Autonomy 1.000 .657
Factor analysis assumptions were verified: (a) univariate nor- Capabilities and experience in the software 1.000 .796
mality (p>0.05; (b) collinearity, since almost all correlations be- development process
tween SHFs were above 0.3 and at least two of the 13 factors Capabilities and experience in software project 1.000 .733
have a moderate correlation with all the others, i.e., there is management
Innovation 1.000 .608
moderate collinearity (Zamora and Esnaola, 2015) (Table 20),
and (c) multicollinearity (det = 0.001) there is a high level of
collinearity in the set of variables included in the matrix without
becoming linearly dependent to the point where it is not equal of 0.895 and a significance level of 0.000 in Bartlett’s test of
to zero. sphericity (Table 21). After ensuring the adequacy of the factor
The EFA solution obtained was satisfactory as the Kaiser– analysis, the factors were extracted using Principal Component
Meyer–Olkin (KMO) measure of sample adequacy reached a value Analysis (PCA).
13
Table 20
Correlations between the SHFs that influence productivity.
SHF 1 2 3 4 5 6 7 8 9 10 11 12 13
1 1.000 .635 .347 .443 .511 .529 .449 .477 .351 .301 .399 .393 .432
2 1.000 .370 .506 .416 .558 .498 .382 .436 .251 .475 .354 .393
3 1.000 .579 .480 .370 .548 .623 .504 .479 .327 .372 .482
4 1,000 .527 .529 .610 .546 .551 .459 .424 .387 .459
5 1.000 .566 .660 .626 .558 .494 .444 .384 .593
6 1.000 .515 .608 .559 .360 .325 .252 .562
7 1.000 .635 .650 .481 .351 .487 .578
8 1.000 .672 .459 .309 .401 .559
9 1.000 .497 .377 .473 .621
10 1.000 .472 .462 .495
11 1.000 .586 .319
12 1.000 .397
13 1.000
1. Communication. 2. Commitment. 3. Motivation. 4. Work satisfaction. 5. Emotional Intelligence. 6. Collaboration. 7. Team Cohesion. 8. Empathy and Interpersonal
Relationships. 9. Leadership. 10. Autonomy. 11. Capabilities and Experience in software development process. 12. Capabilities and Experience in software project
management. 13. Innovation (creativity).
Table 21
KMO and Bartlett test results for the EFA.
Kaiser–Meyer–Olkin measure of sample adequacy .895
Chi-squared (approximate) 800.758
Bartlett’s test of sphericity Gl 78
Sig. .000
Of the 13 factors that make up the Perceptions Scale, only two

are not well represented, namely Motivation and Work Satisfac-
tion (0.574). The other proportion of variance values explained
by the set of common factors (communalities) are above 0.6
(Table 19).
The 13 SHFs are presented in Table 22 and it is observed
that there are three components to explain the variability of the
responses of software development team members whose initial
eigenvalues, following the Kaiser criterion, are greater than 1
(Méndez Martínez and Rondón Sepúlveda, 2012) and are there-
fore to be retained and they accounted for 68.48% of the total
explained variance. Likewise, these components were clearly in-
terpretable and the items or factors that saturated them reached Fig. 1. Component plot in rotated space of the social and human factors that
associated unifactor indices for the reduction of acceptable factors influence productivity in the software development process.
and contribute to one of the three components extracted.
The first component, ‘Factors associated with the person’, ex-
plains 34.368% of the variance and incorporated the factors Moti- and both factors refer to the knowledge and skills that mem-
vation, Work Satisfaction, Emotional Intelligence, Team Cohesion, bers of software development teams should have to improve
Empathy and Interpersonal Relationships, Leadership, Autonomy, productivity.
and Innovation (Table 23). All these factors saturate into a single Fig. 1 shows the distribution of the factors and the composi-
component because they constitute a differentiated group of vari- tion of the components in the plane.
ables in the correlation matrix, as these factors appear to reflect Thus, the factors associated with the person are grouped to-
the emotional qualities that the members of the software devel- gether in the lower right-hand corner of the upper part of the
opment teams should have to contribute to the improvement in plan. In the upper-right part of the same plane are the factors
productivity of the process. associated with interaction: Communication, Commitment, and
The second component, labeled ‘Factors associated with inter- Collaboration. And the capabilities and experience in both soft-
ware development and project management were in the lower
action’, explains 18.411% of the variance. It includes three of the
center of the left side.
13 factors (Communication, Commitment, and Collaboration) and
Based on the results of the EFA, three further EFAs were
denotes how a team member is linked to another (Vallejo-Nágera,
carried out with the items of the factors that made up the three
2002), the feeling of having the backing and support of others
aforementioned components. The purpose of performing these
(Tomasello, 2010), and the responsibility that each member is
EFAs on each component was:
willing to assume in their tasks within their work team, as well
as being responsible for the goals set within the project. 1. To identify items that belong to that component or di-
The third component, ‘Capabilities and Experience’, explains mension and exclude items that are not part of that latent
15.702% of the variance and includes two of the 13 factors (Ca- variable.
pabilities and experience in the software development process, 2. To reduce the length of the original questionnaire by ex-
and Capabilities and experience in software project management) cluding some items. In this way, it is possible to measure
14
Table 22
Total explained variance by component.
Component Initial eigenvalues Sum of squared saturations Sum of squared saturations
Factor of extraction of rotation
Total % of variance Cumulative % Total % of variance Cumulative % Total % of variance Cumulative %
1 6.734 51.802 51.802 6.734 51.802 51.802 4.468 34.368 34.368
2 1.098 8.449 60.250 1.098 8.449 60.250 2.393 18.411 52.778
3 1.070 8.230 68.480 1.070 8.230 68.480 2.041 15.702 68.480
4 .670 5.151 73.631
5 .563 4.327 77.959
6 .526 4.047 82.006
7 .463 3.565 85.570
8 .439 3.376 88.946
9 .394 3.030 91.976
10 .351 2.702 94.678
11 .285 2.196 96.874
12 .217 1.671 98.545
13 .189 1.455 100.000
Results obtained by extraction method: Principal component analysis.
Table 23
Rotated component matrix.
the same variables maintaining the same variability with a 5.2.3. Exploratory factor analysis for the items under the factors:
smaller number of items. Consequently, the response time Motivation, work satisfaction, emotional intelligence, team cohesion,
of the questionnaire is also reduced. empathy and interpersonal relationships, leadership, autonomy and
innovation
The following sections present these results. The results of the EFA carried out with the 53 items included
under the seven factors of the component ‘Factors associated with
5.2.2. Exploratory factor analysis for the communication, commit-
the person’ showed that there are 15 factors or components, as
ment and collaboration items
shown in Table 26. In this table, the components group items
For the component named ‘Factors associated with interac-
from different factors.
tion’, which included the factors Communication, Commitment,
Based on the results of the EFA, the items of each factor are
and Collaboration, and comprised of six, five, and four items re-
filtered and those corresponding to each factor are presented,
spectively, the EFA also determined three components, which are
taking as a criterion the items that are most prevalent in each
presented in Table 24. It can be seen from the rotated component
component. Thus, four items comprise the Innovation factor (Ta-
results that the Communication factor, initially composed of six
items, now has four items that belong to this factor and that item ble 27), three items comprise each of the factors of Emotional
two, ‘‘The project objectives and their respective activities should Intelligence, Work Satisfaction, Team Cohesion, Empathy, and
be explicit and clear to all team members, in order to improve Motivation, while under Leadership and Autonomy there are two
the productivity of the software development process’’, due to its items each.
behavior, is part of the Collaboration factor. On the other hand,
item five that stated, ‘‘To improve the productivity of the software 5.2.4. Exploratory factor analysis for the items under the factors
development process, team members should be informed in a capabilities and experience in software development and project
timely manner about the progress of goals and achievement of management
objectives’’, according to the results of the EFA belongs under An EFA was carried out with the 11 items under the factors
the Commitment factor. The remaining items, according to the Capabilities and Experience in software development (7 items)
EFA, did correspond to the factors under which they were initially and in software project management (4 items) employing the
placed. principal component extraction method with Varimax normaliza-
The results in Table 24 allow for a refinement of the items tion as the rotation method. In the factor analysis, all the SHFs
under the three factors, with each factor now being conformed of entered had loading values greater than 0.4, and therefore none
the items presented in Table 25. With this new configuration, the was discarded (Table 28).
Communication and Collaboration factors are composed of four Table 29 shows that the Kaiser–Meyer–Olkin (KMO) sample
items, while the Commitment factor contains five items. The two adequacy measure reached a value of 0.895 and a significance
items of the Communication factor that the EFA placed under the level of 0.000 in Bartlett’s test of sphericity, indicating that the so-
factors Commitment and Collaboration were eliminated. lution obtained from the EFA was satisfactory. Once the adequacy
15
Table 24
Rotated component matrix for the Communication, Commitment and Collaboration constructs.
Table 25
Distribution of the Communication, Commitment and Collaboration factor items.
Factor Items
Communication among team members is an important support.
Each task should have a clearly identified responsible person.
Communication
Team members should maintain a fluid communication.
It is necessary to define a communication protocol between team members and external personnel.
The team should carry out the tasks necessary for the success of the project in accordance with the objectives set.
Team members should have a level of responsibility that enables them to take on tasks in favor of the objectives of the team.
Commitment The team should be clear about its responsibility to fulfill the tasks set and be willing to help when required.
All team members should take responsibility for the results obtained, fulfill their duties, and be able to admit their mistakes.
Team members should fully and punctually fulfill their assigned tasks.
Team members should work collaboratively to achieve project goals.
There should be trust between team members for the performance of their duties and the protection of common interests.
Collaboration
Team members should be willing to help and support each other.
It is necessary for each team member to share their knowledge, information and experience with their peers.
Table 26
Distribution of the items for the SHF Innovation, Leadership, Emotional Intelligence, Work Satisfaction, Team Cohesion, Autonomy, Empathy and
Interpersonal Relationships, and Motivation in the components of the rotated component matrix.
Component Number of Items
items
1. 10 4 items under Innovation (1,3,5,6), 3 items under Leadership (3,6,7) 1 item under Empathy (9)
1 item under Cohesion (2) and 1item under Satisfaction (1)
2. 4 2 items under Leadership (2,5), 1 item under Empathy (8) and 1 item under Innovation (2)
3. 5 3 items under Intelligence (2,6,8), 1 item under Motivation (4) and 1 item under Cohesion (3)
4. 4 2 items under Intelligence (1,3), 1 item under Satisfaction (7) and 1 item under Empathy (3)
5. 4 3 items under Satisfaction (2,3,4) and 1 item under Empathy (5)
6. 4 3 items under Cohesion (1,4,5) and 1 item under Intelligence (7)
7. 2 2 items under Autonomy (3,5)
8. 5 3 items under Empathy (4,6,7), 1 item under Motivation (1) and 1 item under Cohesion (6)
9. 3 2 items under Intelligence (4,5) and 1 item under Autonomy (4)
10. 2 2 items under Autonomy (1,2)
11. 3 3 items under Motivation (2,3,5)
12. 3 2 items under Satisfaction (5,6) and 1 item under Empathy (2)
13. 1 1 item under Innovation
14. 1 1 item under Empathy (1)
15. 2 2 items under Leadership (1,4)
of the factor analysis was assured, the factors were extracted above 0.6, as shown in Table 28. The item with its proportion
using Principal Component Analysis (PCA). of variance or communality (.534) below the accepted value is
Ten of the 11 items that formed part of the two factors are that which states that team members should ‘‘have knowledge
well represented, while the values of the proportion of variance or experience in the use of the programming tools and languages
explained by the set of common factors (communalities) are required for the project’’.
16
Table 27
The items for the factors Innovation, Leadership, Emotional Intelligence, Work Satisfaction, Team Cohesion, Autonomy, Empathy and
Interpersonal Relationships, and Motivation.
Component Items
In order to innovate, customer suggestions, complaints and/or claims should be taken into account.
Innovation The use of solutions which have not been satisfactorily tested should be avoided.
(4 items) Software development team members should be supportive and receptive to new ideas.
Leaders should encourage members to put their own ideas into practice and find new ways of resolving
problems.
Leadership Each team member should feel that they can offer solutions to problems within the project.
(2 items) The leader should coordinate and guide the team’s activities towards the project’s objectives and goals.
Emotional Each of the team members should express their disagreement in a timely manner and to the appropriate
intelligence person.
(3 ítems) The work team should have the ability to resolve conflicts appropriately.
Team members should build relationships based on trust and respect.
The activities carried out by team members should contribute to their personal and professional growth.
Satisfaction
Team members should feel satisfied that the distribution of work is equitable.
(3 items)
Team members should feel satisfied with the activities they perform.
It is necessary that team members work at a similar pace.
Cohesion
Each team member should enjoy carrying out tasks with his or her teammates.
(3 items)
Activities should be executed in a timely manner and with the participation of all responsible parties.
Empathy and It is necessary that team members receive training in interpersonal relations, effective management of
Interpersonal emotions, teamwork and quality.
Relationships Team members should be able to put themselves in another person’s shoes and provide support where
(3 items) necessary.
Each team member should participate in the activities carried out in his or her area of work.
Autonomy Team members should be empowered to make decisions regarding the project and their way of working
(2 items) within it.
Members can organize themselves to establish and meet their objectives.
Tasks assigned to team members should be perceived as interesting and challenging.
Motivation
Team members should feel that the tasks they perform are valuable in achieving the objectives.
(3 items)
Good furniture, computer equipment and optimal working conditions should be provided.
Table 28
Commonalities of the factors Capabilities and Experience in the development and in the management of software projects.
To improve the productivity of the software development process, team members should... Initial Extraction
Keep up to date with the best tools and practices for the execution of tasks. 1.000 .651
Have knowledge of the subject or have worked in similar contexts. 1.000 .647
Have knowledge or experience in the use of the programming tools and languages required for the project. 1.000 .534
Have knowledge or experience in the analysis, design, construction or implementation of software. 1.000 .687
Have logical reasoning and systematic thinking skills. 1.000 .642
Havethe ability to implement efficient solutions that meet project requirements. 1.000 .697
Have knowledge or experience in the use of software development methodologies. 1.000 .609
Team members should adequately manage the time frames agreed. 1.000 .665
Team members should have the ability to plan, execute or control project activities. 1.000 .616
Team members should have knowledge or experience in the use of project management tools and techniques 1.000 .792
that improve productivity.
Team members should have knowledge or experience in the use of metrics that allow the project to be 1.000 .740
monitored.
Extraction Method: Principal Component Analysis.
Table 29 The first component kept the original name of the factor,
KMO y Bartlett test results for the EFA.
namely, ‘Capabilities and experience in the software development
Kaiser–Meyer–Olkin measure of sample adequacy .813
process’, which explains the 25.644% of the variance and incorpo-
Chi-squared (approximate) 480.727
Bartlett’s test of rated four of the seven items (Table 31). All these factors saturate
Gl 55
sphericity in a single component because they constitute a differentiated
Sig. .000
group of variables in the correlation matrix. This factor seems to
reflect the cognitive aspects and expertise that the members of
the software development teams should have and that in their
Table 30 presents the 11 SHF items which shows that there
opinion contribute to improving the productivity of the process.
are three components to explain the variability of responses of
The second component, grouping two of the four items of the
software development team members whose initial eigenval-
ues, following the Kaiser criterion, are greater than 1 (Méndez factor ‘Capabilities and experience in software project manage-
Martínez and Rondón Sepúlveda, 2012) and therefore to be re- ment’ and two items from the factor ‘Capabilities and experience
tained and that they represented 66.194% of the total explained in the software development process’, was named ‘Autonomous
variance. Similarly, these components were clearly interpretable management skills’. This component explains 21.234% of the total
and the items or factors that saturated them reached associ- variance. The common denominator of these items is the capacity
ated unifactor indices for the reduction of acceptable factors and that team members should have to manage their own time and
contribute to one of the three extracted components. carry out the activities they are responsible for.
17
Table 30
Total explained variance by component.
Initial eigenvalues Sum of squared saturations of extraction Sum of squared saturations of rotation
Component
Total % of variance Cumulative % Total % of variance Cumulative % Total % of variance Cumulative %
1 4.828 43.894 43.894 4.828 43.894 43.894 2.821 25.644 25.644
2 1.412 12.835 56.729 1.412 12.835 56.729 2.336 21.234 46.878
3 1.041 9.465 66.194 1.041 9.465 66.194 2.125 19.316 66.194
4 .730 6.639 72.833
5 .666 6.055 78.888
6 .591 5.376 84.264
7 .491 4.462 88.726
8 .384 3.491 92.217
9 .352 3.199 95.416
10 .320 2.913 98.329
11 .184 1.671 100.000
Extraction method: Principal component analysis.
Table 31
Table 32
Factor To improve the productivity of the software development process, team members should...
Have knowledge or experience in the use of the programming tools and languages required
Capabilities and experience in for the project.
software project development Have knowledge or experience in the analysis, design, construction or implementation of
software.
Have logical reasoning and systematic thinking skills.
Have knowledge or experience in the use of software development methodologies.
Autonomous management Adequately manage the time frames agreed.
skills (project management) Have the ability to plan, execute or control project activities.
Project management skills Have knowledge or experience in the use of project management tools and techniques that
using tools (project improve productivity.
management) Have knowledge or experience in the use of metrics that allow the project to be monitored.
The third component, ‘Project management skills using tools’, 6. Discussion

explains 19.316% of the variance and includes two of the four
items of the factor ‘Capabilities and experience in software project According to the results of this study, all SHFs are perceived
management’ and one item from the factor ‘Capabilities and ex- by the respondents as factors that influence the productivity of
perience in software development’. Underlying these items is the software development team members. These results are framed
idea that project management is performed using management in the Colombian context and confirm the previous findings of
tools or metrics. this research (Machuca-Villegas et al., 2020), meaning that the
In summary, based on the EFA, the items were filtered, ex- SHFs we identified in the literature are considered influential fac-
cluding those that initially belonged to a factor and were then tors in productivity, and the percentage of responses to the items
grouped under another component. The items which comprise for each SHF maintained a trend in the Agree and Strongly agree
these components are listed in Table 32. categories. Therefore, human factors play a key role in influencing
In summary, an improved version of the instrument to mea- productivity (Cunha De Oliveira, 2017) and in the success of IT
sure members’ perceptions of the SHF that improve the produc- projects (Iriarte and Bayona, 2020). Also, it was identified that
tivity of software development teams (Machuca-Villegas et al., SHFs correlate with each other and can be classified into three
2021a) is now conformed of the 44 items presented in Tables 25, broader components. In this section, we discuss the results of the
27 and 32. descriptive and inferential analysis.
18
6.1. Descriptive analysis 2020) also indicate a strong relationship between satisfaction and
agile development, due to collaborative practices such as self-
The descriptive analysis favored the interpretation of software organized teams and collective code ownership. Consequently,
development team members’ perception of the SHFs from their this may indicate a link with SHFs of autonomy and collaboration.
responses to the SHF items. The most significant findings were The results of the SHF emotional intelligence show that there
compared with other studies where some identified SHF in that is favorability regarding the statements that included adaptation
research were referenced (directly or indirectly) to productivity. to change, conflict resolution, listening skills, trust, and mutual
The findings are presented below. respect. Therefore, from this SHF, the emotional states of the
Communication among team members is considered by Colom- team members are considered influential for good interpersonal
bian software professionals as an essential support to improve relationships and therefore for their productivity. This result was
productivity in software development teams and perceived that manifested within a Colombian environment thus, it is neces-
project objectives and activities should be explicit and clear. sary to perform additional studies in other cultures to verify the
These results are in line with the research by Ganguly et al. influence of emotional states. Around this factor, the emotions
(2020) which identified that the frequency of communication be- of software developers could also be analyzed (Sánchez-Gordón
tween team members from Bangladesh, well-defined objectives, and Colomo-Palacios, 2019), as well as levels of happiness or
cooperation, and work feedback are considered factors that posi- unhappiness and their influence on productivity (Graziotin et al.,
tively influence productivity, during remote working in COVID-19 2018).
times. Communication is also trending in agile development con- In the SHF collaboration, there is a tendency for respondents
texts and global teams (Hidayati et al., 2020), along with the to agree that collaborative work and the characteristics inherent
factors of collaboration and trust, which are perceived as impor- to collaboration (e.g. trust, help, and support among team mem-
tant and positively influencing factors in agile projects (Chagas bers) are important in the productivity of their work team. In
et al., 2015). agreement with the comparative studies analyzed in the section
In the SHF commitment, 86.6% of the respondents agreed that on related work, collaboration stands out as the SHF mapped
‘‘team members should fully and punctually fulfill their assigned out in all studies, which identifies it as an important factor for
tasks’’. Likewise, regarding punctuality, for software managers teamwork and influential for productivity. Collaboration drives
and project leaders from Brazil, the developer’s productivity is productivity and facilitates innovative problem solving (Project
associated with the delivery of tasks on time and without any Management Institute & IEEE Computer Society, 2013). On the
repetition or reprocessing; they also consider important com- other hand, Franca et al. (2020) found that Brazilian software en-
mitment to the project (Cunha De Oliveira, 2017). On the other gineers can display collaborative behavior when they are highly
hand, the happiness of the developer positively influences their motivated and there is communication and participation among
commitment and perseverance with the task at hand (Graziotin team members. This may relate to our results, given that the SHFs
et al., 2018). of collaboration and communication were grouped in the second
Regarding the SHF motivation, it is important to note that the component of the exploratory factor analysis and presented a
item which obtained the highest level of disagreement was that moderate correlation of 0.529.
which stated that the tasks assigned to team members should Regarding the SHF team cohesion, the results indicate that it
be interesting and challenging. These results contrast with opin- is important to consider the level of integration among team
ions on what factors motivate software engineers when faced members and the distribution of their responsibilities to achieve
with challenging tasks (Beecham et al., 2008; López-Fernández their objectives and for the work to be completed. This facili-
and Yagüe Panadero, 2011). According to the results of Cunha tates teamwork and decision making, so it is possible to have
De Oliveira (2017) framed in the Brazilian context, some devel- a more cohesive and productive team. This factor also stands
opers need challenges to stay motivated. The results above may out for being a predominant factor in high-performance teams
indicate that it is still unclear what motivates a software engi- (Dutra et al., 2015) and for its positive influence on productivity
neer (Beecham et al., 2008; Hernández López, 2014) regardless (Canedo and Santos, 2019). Therefore, it is important to consider
of the cultural context and that motivation is internal to the cohesion among team members in order to maintain integration
individual and different for each person. Rewards, however, are and improve performance. Also, it was found that respondents do
well received by software development teams. These are used to not closely associate productivity with the pace of work nor with
motivate people and therefore having a balance between extrinsic collective working. This could mean, according to the perception
rewards and intrinsic rewards is recommended, since extrinsic of the Colombian respondents, that each team member can work
motivators have been shown to undermine intrinsic motivators at different rhythms and productivity would not be affected, thus
(Werbach & Hunter, 2012). promoting autonomy. However, in other cultural contexts alter-
In the Colombian setting, for the SHF Work satisfaction, 41.1% natives to Colombia, different results could be obtained, which
of the respondents disagreed with the item ‘‘the tasks assigned to constitutes a limitation of the research identified in Section 6.3.
each of the team members should correspond to a large degree Empathy and interpersonal relationships is another influential
with what each one wants to do’’. This is one of the items that SHF in the productivity of software development teams. Respon-
presented a higher level of disagreement when compared to the dents value interpersonal relationships over work itself. In ad-
items under other SHFs. When examining those items with the dition, they consider it necessary to support each other and
greatest tendency towards agreement and strong agreement, the put themselves in other team members’ shoes in order to have
need to recognize the contribution of each team member stands good personal relationships. This SHF, therefore, is considered an
out. Recognition is considered an antecedent of satisfaction for important factor given that in software projects there is a need
software engineers (Franca et al., 2020), and this suggests that to interact among team members (Paiva et al., 2010) even when
recognizing the work of team members can be a way to ensure they are global (Hidayati et al., 2020). The success of projects
their satisfaction is not affected. also depends on how professionals carry out their tasks and
Likewise, Work satisfaction is a factor studied in its relationship how they interact with their work team (Capretz and Ahmed,
with productivity (Storey et al., 2019), and in turn, in conjunction 2018). Similarly, studies in Brazil by Paiva et al. (2010) show that
with motivation for the purpose of helping software organizations interpersonal relationships have a high positive influence on team
be more productive (Franca et al., 2020). Studies by (Kropp et al., productivity. This corroborates the perception of 76.8% of the
19
respondents who consider that participating in social activities From these results, software development companies will be able
inside and outside the work environment improves their pro- to propose improvement and training strategies so that their
ductivity. Given this finding, team leaders and software project work teams strengthen these types of SHFs and can see favorable
managers should consider such activities to raise productivity. results in team members’ performance and job satisfaction. This
Although this recommendation is associated with the Colombian is also a topic to be considered in the fields of academia and
culture, we believe it could be explored in other backdrops. research as part of the training of software professionals (Capretz
The results of the SHF leadership highlight the characteristics and Ahmed, 2018).
of a software development team leader considered influential
for productivity, including self-confidence, promoting positive 6.2. Inferential analysis
attitudes and generating trust among team members, and co-
ordinating and guiding the team’s activities towards the ob-
The inferential analysis enabled research into the correlation
jectives and goals of the project. Therefore, it is an essential
between the SHFs, leading to the establishment of three groups of
factor for team management, and it aligns with the findings of
SHFs, to which their classification was proposed. The instrument
the study conducted to software engineering professionals in
used in this study was also validated and, based on this validation
Uruguay (Matturro et al., 2015b), which mentions that leadership
is one of the soft skills that leaders and members of a devel- process, a new version of the instrument was created, reducing
opment team should have, as well as any software professional the number of items from 79 to 44. We clarify that these findings
(Matturro et al., 2019). are the result of a study conducted in a Latin American country
The assimilation of new ideas and innovative solutions in the (Colombia) and may be influenced by the cultural and work char-
development of software projects is influential in improving pro- acteristics of the respondents. However, they can be considered
ductivity. Thus, the SHF innovation is perceived by respondents as a reference for the development of studies in similar contexts.
as an important factor in the productivity of the work team. The following section outlines the key findings.
Similarly, studies by Matturro et al. (2019) consider this SHF
as a relevant skill for software engineering practice, as well as 6.2.1. Correlations between SHFs that influence productivity
creativity, which is also central to innovation and is positively The results show that there is a relationship between the
influenced by happiness (Graziotin et al., 2018). 13 SHFs that influence productivity. Although the correlations
Respondents’ perceptions of the SHF autonomy show a high are weak or moderate, from these correlations it is possible to
tendency towards agreement with all the statements under this identify that certain SHFs tend to be more closely interrelated.
factor. This may mean that the autonomy of team members to The SHFs that obtained a correlation higher than 0.6 are:
make decisions about their work and responsibilities influences
their productivity. This is an important factor for high perfor- • Communication and Commitment with 0.635
mance teams (Dutra et al., 2015) and agile teams (Stray et al., • Motivation and Empathy and interpersonal relationships with
2018). Moreover, it is an SHF related to a work team’s self- 0.623.
management (Fatema and Sakib, 2018b) and self-organization • Work satisfaction and Team cohesion with 0.610.
(Iqbal et al., 2019). The above supports the importance of auton- • Emotional intelligence and Team cohesion with 0.660
omy in productivity and highlights its usefulness in different team • Emotional intelligence and Empathy and interpersonal rela-
styles. tionships with 0.626
Regarding the items under the SHF Capabilities and experi- • Collaboration and Empathy and interpersonal relationships
ence in the software development process, respondents perceived with 0.608
staying updated in the use of tools and practices for the soft- • Team cohesion and Empathy and interpersonal relationships
ware development process as well as having logical and systemic with 0.635
thinking skills as being important factors for their productivity. • Team cohesion and Leadership with 0.650
However, respondents also perceived that it is not so necessary • Empathy and interpersonal relationships and Leadership with
to have previous experience or knowledge of working in similar 0.672
contexts. • Leadership and Innovation with 0.621
Capabilities and experience in software project management is
These findings contribute to the understanding of the relation-
another SHF that was considered by respondents as influential in
the productivity of the software development team. The results ships that can occur between SHFs, in such a way that they en-
show that it is important to manage the agreed time frames courage intervention with a set of SHFs together in the designing
for tasks and the planning, execution, and control of a project of improvement strategies for the management and productivity
as a key aspect of productivity for the team. These results are of a software development team. In other words, the inclusion of
related to the study by Murphy-hill et al. (2019) conducted with one of the factors in an improvement strategy would implicate
developers from three U.S. companies. It is reported that auton- the inclusion of the factor with which it is most highly related,
omy in time management is positively related to productivity. provided the correlation is positive, i.e. increasing the scores of
Similarly, in Ganguly et al. (2020) an increase in decision making one of the factors increases the scores of the factor with which it
about time management during online working was perceived is associated. Therefore, in the design of any type of improvement
as something that positively influenced productivity. However, strategy, fewer factors need to be included and this can contribute
28.5% of respondents considered that knowledge or experience in to a simplification of the strategy.
the use of project management tools and techniques is not such Other studies such as (Hsu and Mujtaba, 2007) found that
a necessity to achieve a productivity improvement. team transformational leadership is strongly related to team em-
Finally, although the results are within the Colombian back- powerment and team trust. In our case, these factors refer to the
ground, the descriptive analysis reveals the importance of SHFs in SHF leadership, autonomy and collaboration, respectively. Hsu and
the software development process and their influence on produc- Mujtaba (2007) indicate that team empowerment and team trust
tivity. This is aligned with the need to study and promote these are not related to Work satisfaction in development teams and
types of skills in software development teams as they contribute that there is a weak relationship between team commitment and
to successful outcomes in projects (Capretz and Ahmed, 2018). Work satisfaction.
20
6.2.2. Proposed SHF classification based on exploratory factor anal- 6.3. Limitations and threats to the validity of the study
ysis (EFA)
The results of the EFA suggest three components or groups Several limitations or threats to the validity of the results of
of SHFs that were identified from the common factors or latent this research should be considered.
variables necessary to explain the common variance of the set of One of the threats is associated with the perceptions of the
items of each factor (Lloret-Segura et al., 2014, p. 1152). Based respondents consulted. Firstly, these perceptions are subject to
on the components that emerged from the EFA, the following an individual’s state of mind and their specific situation con-
classification of the 13 SHFs is proposed: cerning their work team and the company where they work. To
minimize this threat, people from different companies and cities
1. Under the category, ‘Factors associated with the person’, were surveyed. Secondly, the perceptions reflect the opinions of
are the SHFs Motivation, Work Satisfaction, Emotional In- members of development teams in a Latin American context. A
telligence, Team Cohesion, Empathy and Interpersonal Re- recommendation therefore would be to broaden this study to
lationships, Leadership, Autonomy, and Innovation. include respondents from countries with different cultural con-
2. Under the category, ‘Factors associated with team inter- ditions to have a more generalized result and to observe whether
action’, are the SHFs Communication, Commitment, and the perceptions are similar.
Collaboration. Respondents were left to use the concept of productivity in
3. Under the category, ‘Capabilities and Experience’, are the broad scope, due to the lack of agreement about a productivity in
SHFs Capabilities and Experience in the software develop- Software Engineering definition. This is a threat to the research
ment process and Capabilities and Experience in software because of the difficulty of identifying the concept of productiv-
project management. ity adopted by each respondent. Nevertheless, the results show
Although classification is proposed for SHFs, these ‘‘factors are a degree of agreement on the perception of each SHFs being
studied from a global and integrative perspective of the social considered as influential in productivity. Based on this threat,
and the individual, enabling an understanding of SHFs within the we recommend adding to the instrument the definition of the
individual-group interaction. They are also analyzed in a non- concept of productivity on which the study is based.
hierarchical way to focus on how they can favor the results of Another limitation is related to the study sample. Although
a software development team’’ (Machuca-Villegas et al., 2020). the number of respondents is representative for the scope of this
Therefore, this classification proposal can be seen as a starting research and the findings described were reached, expanding the
point for developing improvement strategies that stimulate these sample size is necessary to achieve a greater generalization of
types of factors in software development teams. In a similar way the results. To this end, widening the survey reach using the new
to what happens with the correlations between the SHFs, this instrument is suggested.
three-group classification can contribute to simplifying the design The items used to define each of the 13 SHFs are likely to have
of these strategies since multiple SHFs can be worked on via the a bias in their characterization. However, to minimize this bias,
broader categories. the instrument was subjected to content validation by experts,
On the other hand, according to Oliveira et al. (2018b), there not only in the subject but also in linguistics and psychology.
are different classification structures of factors that influence The instrument included statements with positive and nega-
productivity. They identified, from the secondary studies ana- tive valences, following the approach of Aiken (1996). In this way,
lyzed, common categories such as factors at the Product, Project, it was possible to break the auto-pilot response pattern for some
and People levels. In turn, as part of their research results, they of the SHF questions to help the respondents read each statement
propose a classification of factors categorized into organizational closely.
and human. This classification is based on the similarities of the
factors. In examining the category of human factors, a relation- 6.4. Relevancy and implication of the research results
ship with our classification proposal can be noted, especially
with those factors associated with capabilities and experiences, Due to the way in which the subjects participating in the re-
as similarly classified by Wagner and Ruhe (2008). However, search were selected, the inferences obtained cannot be extended
with the other two categories there are no such corresponding beyond the sample used, that is, software development teams in
similarities, although related factors are included. Colombia. To establish whether other teams in other countries
Other studies (Canedo and Santos, 2019; Murphy-hill et al., register the same perceptions, it is necessary to carry out research
2019) also include human or people factors as influencing soft- that uses another sampling framework.
ware development productivity, however, these studies lack any Finally, this research contributes to the fulfillment of the sus-
classification to group the factors. tainable development goals related to economic growth, industry,
innovation, and infrastructure, by promoting improvements in
6.2.3. Proposal of an instrument to assess SHF perceptions the productivity of software development teams and by providing
With the validity of the construct of the instrument used in management tools to the software industry, which is responsible
this research, the SHF definitions are consolidated and a newly for enhancing the technological capabilities of industrial sectors.
refined and the validated instrument is proposed to evaluate Therefore, implications for software teams and for research are
the perception of these type of factors. As an adjusted version described below.
of the original proposal, it requires less time to be applied. In 6.4.1. Implications for software teams
addition, it is a useful resource for software development com- This work provides an instrument that can be applied in soft-
panies to identify needs related to SHFs within their work teams ware development teams to identify which SHFs are perceived
and then propose improvement initiatives. The instrument also as influencing productivity and may affect their performance.
seeks to highlight the importance of SHFs as key elements for Additionally, the analysis of the results allows the design and
success in software development projects and the well-being of implementation of management strategies, which can be based
development teams. on gamification to promote team performance.
This instrument adds to related research (Fatema and Sakib, In turn, the results of this study provide theoretical and em-
2018b; Iqbal et al., 2019; Canedo and Santos, 2019; Murphy- pirical support for the SHFs. This support can be used by team
hill et al., 2019; Ganguly et al., 2020) to study the factors that leaders and managers to recognize which SHFs are predomi-
influence productivity in software development. nant in productivity and should be addressed as a priority in
21
their management processes to improve team performance and with the perceptions of a team members can guide the man-
well-being. agement of the team itself, improve its work and therefore its
On the other hand, with the fourth industrial revolution, the productivity.
organizational structure and processes of industries have been The findings serve as recommendations so that organizations
modified, thus creating an urgent need to adopt technology and can propose new work approaches to strengthen SHFs and orient
information systems to the new needs (Kornyshova and Barrios, the work of their software development teams and management
2020). From this context, we consider that this research con- under their identified needs concerning SHFs. For instance, un-
tributes to the software industry since technical and soft skills derstanding these SHFs could facilitate the design of strategies
are required, including thinking, social and personal skills (Maisiri to promote collaboration in knowledge transfer (Galeano-Ospino
et al., 2019), aligned with the 13 identified SHFs. et al., 2020), training the team, motivating the team through
6.4.2. Implications for researchers rewards (Gasca-Hurtado et al., 2020), creating teams according to
By identifying the 13 SHFs, we can contribute to new research the profile of their members, as well as fostering a collaborative
opportunities. For instance, the measurement of the 13 SHFs and work environment and good interpersonal relationships.
their level of influence on the productivity of software develop- In addition, we recommend that software teams take into ac-
ment teams. For this measurement, it is necessary to design and count the classification of SHFs suggested by the results, through
implement measurement tools. Some alternative tools to psycho- which they can design improvement solutions from an integral
metric tests (based on self-report) to measure SHFs are interactive perspective, which in turn facilitates their joint work. This could
styles (Muñoz et al., 2018) and linguistic analysis (Licorish et al., reduce time and costs in proposing strategies by intervening on
2009; Licorish and MacDonell, 2014, 2015). We consider it possi- SHFs by category rather than individually.
ble to establish a measurement process by implementing any of
Likewise, we believe software development teams can use
the above tools based on these 13 SHFs.
this new instrument resulting from the EFA as a diagnostic tool
The 13 SHFs provide a basis for the design of models based
to identify which SHFs need to be intervened in productivity
on system dynamics that seek to simulate the behavior of soft-
improvement strategies.
ware development teams, intending to analyze their behavior
We consider the results of this study can also contribute to
and make decisions aimed at increasing their productivity. These
the academic and research contexts of Software Engineering.
models are a support tool for software project management
From the academic context, they serve as support for curricu-
(Franco et al., 2018).
lar modifications and knowledge body guides in the curricula
Emotion is a topic that is interesting in the software indus-
associated with this area to include this type of SHF as part
try. We believe it would be useful to consider the relationship
of the training process (Capretz and Ahmed, 2018). From the
between emotions and SHFs in team performance and well-
being. The findings of this research can contribute to the study research context, they constitute a basis for the development of
of developers’ emotions (Sánchez-Gordón and Colomo-Palacios, new research lines associated with team management, software
2019). project management, productivity improvement, among others.
As a potential future area of study, we seek to expand the
7. Conclusions and future research study of each of the SHFs analyzed in this research, performing
a more exhaustive analysis of their level of influence and impor-
SHFs are key factors for success in the software development tance about to productivity. Using the new instrument to make an
process. In this paper, we present the results of a survey-based intervention based on those SHFs that are of interest to enhance
study in which we consulted members of software development a software development team’s productivity is also proposed. A
teams about their perceptions of 13 SHFs that influence produc- pre- and post-test could be performed so that any possible effects
tivity. A descriptive and inferential analysis was performed on the of an intervention on productivity could be identified. In addition,
subjects’ responses. The results confirm the importance of SHFs it could be useful to include in the instrument more demographic
in the software development process and productivity and point information that would allow stratified analysis.
to the value of a greater interest in their study, knowledge and We are also designing gamification-based strategies where
training so that their influence and management may have fa- interventions can be made around an SHF to analyze any positive
vorable consequences for the software industry and the scientific influence it has on the productivity of the software development
community at large. team, as well as the measurement of the SHF in relation to such
The results of this study provide empirical evidence about the productivity.
importance of SHFs in software development productivity, while
technical factors have been more studied in this area. Addition- CRediT authorship contribution statement
ally, these findings expand understanding about the SHFs that in-
fluence software development productivity and are in agreement Liliana Machuca-Villegas: Conceptualization, Methodology,
with the results of Cunha De Oliveira (2017) which emphasize
Investigation, Writing – original draft, Visualization, Funding
human factors over organizational factors.
acquisition. Gloria Piedad Gasca-Hurtado: Conceptualization,
The existing correlation between the 13 SHFs facilitates the
Methodology, Investigation, Writing – review & editing, Super-
analysis and the identification of relationships among them for
vision, Project administration, Funding acquisition. Solbey Mo-
their intervention through improvement strategies, and the def-
rillo Puente: Formal analysis, Data curation, Writing – review
inition of productivity measures (Machuca-Villegas et al., 2021).
& editing. Luz Marcela Restrepo Tamayo: Funding acquisition,
With these interrelationships, team leaders will be able to pro-
Writing – review & editing, Resources, Data curation.
pose the design of such strategies more easily. Given that the
correlations between SHFs are positive, increasing the scores of
one of the factors would mean an increase in the scores of the Declaration of competing interest
factors with which it is associated.
We consider that the empirical evidence of this work con- The authors declare that they have no known competing
tributes as a point of reference and study for software companies financial interests or personal relationships that could have
to strengthen SHFs in their work teams. The details associated appeared to influence the work reported in this paper.
22
Acknowledgments Ganguly, K.K., Tahsin, N., Fuad, M.M.N., Ahammed, T., Asad, M., Huq, S.F., et
al., 2020. Impact on the productivity of remotely working IT professionals
of Bangladesh during the coronavirus disease 2019. Retrieved from http:
This research was funded to Universidad de Medellín under ID
//arxiv.org/abs/2008.11636.
project 1109. Gasca-Hurtado, G.P., Gómez-Álvarez, M.C., Muñoz, M., Betancur, S.H., 2020.
Gamified strategy oriented to decrease SPI change resistance: A case study.
References In: European Conference on Software Process Improvement. pp. 313–325.
Gómez-Jakobsen, F., 2016. Medidas de productividad en los proyectos de desar-
rollo de software: una aproximación por puestos de trabajo. http://dx.doi.
Adolph, S., Hall, W., Kruchten, P., 2011. Using grounded theory to study the
org/10.1174/021435502753511268.
experience of software development. Empir. Softw. Eng. 16 (4), 487–513.
Graziotin, D., Fagerholm, F., Wang, X., Abrahamsson, P., 2018. What happens
http://dx.doi.org/10.1007/s10664-010-9152-6.
when software developers are (un)happy. J. Syst. Softw. 140, 32–47. http:
Aiken, L., 1996. Test PsicolÓ. Prentice Hall Iberoamericana, México D.F..
//dx.doi.org/10.1016/j.jss.2018.02.041.
Amrit, C., Daneva, M., Damian, D., 2014. Human factors in software development:
Hernández López, A., 2014. Medidas de productividad en los proyectos de
On its underlying theories and the value of learning from related disciplines.
desarrollo de software: una aproximació.
a guest editorial introduction to the special issue. Inf. Softw. Technol. 56 (12),
Herzberg, F., Mausner, B., Bloch Snyderman, B., 1959. Motivation to Work. John
1537–1542. http://dx.doi.org/10.1016/j.infsof.2014.07.006.
Wiley & Sons, Nueva York.
Anderson, D.R., Sweeney, D.J., Williams, T.A., 2008. EstadÍ, 10a. edici Cengage
Hidayati, A., Budiardjo, E.K., Purwandari, B., 2020. Hard and soft skills for scrum
Learning Editores, S.A.
global software development teams. In: Proceedings of the 3rd International
Barros, E.A. de, 2010. CatÁ. Universidade Federal de Pernambuco. Conference on Software Engineering and Information Management. pp.
Beecham, S., Baddoo, N., Hall, T., Robinson, H., Sharp, H., 2008. Motivation in 110–114.
software engineering: A systematic literature review. Inf. Softw. Technol. 50 Hsu, S.-Y.C., Mujtaba, B.G., 2007. Team transformational leadership, trust, sat-
(9–10), 860–878. http://dx.doi.org/10.1016/j.infsof.2007.09.004. isfaction, and commitment: The testing of a structural equation model
Boehm, B., 1981. Software Engineering Economics. Prentice Hall PTR, Upper in software development teams. Rev. Bus. Inf. Syst. (RBIS) 11 (3), 17–28.
Saddle River, NJ, United States. http://dx.doi.org/10.19030/rbis.v11i3.4412.
Canedo, E.D., Santos, G.A., 2019. Factors affecting software development pro- Iqbal, J., Omar, M., Yasin, A., 2019. An empirical analysis of the effect of
ductivity: An empirical study. In: ACM International Conference Proceeding agile teams on software productivity. In: 2nd International Conference on
Series. 307–316, http://dx.doi.org/10.1145/3350768.3352491. Computing, Mathematics and Engineering Technologies. ICoMET, IEEE, pp.
Capretz, L.F., Ahmed, F., 2018. A call to promote soft skills in software engineer- 1–8.
ing. Psychol. Cogn. Sci. - Open J. 4 (1), e1–e3. http://dx.doi.org/10.17140/ Iriarte, C., Bayona, S., 2020. It projects success factors: A literature review.
pcsoj-4-e011. Int. J. Inf. Syst. Project Manag. 8 (2), 49–78. http://dx.doi.org/10.12821/
Chagas, A., Santos, M., Santana, C., Vasconcelos, A., 2015. The impact of human ijispm080203.
factors on agile projects. In: Proceedings - 2015 Agile Conference, Agile, 2015. Kitchenham, B.A., Pfleeger, S.L., 2008. Personal opinion surveys. In: Guide to
pp. 87–91. http://dx.doi.org/10.1109/Agile.2015.11. Advanced Empirical Software Engineering. pp. 63–92. http://dx.doi.org/10.
Cunha De Oliveira, E.C., 2017. Fatores de Influência Na Produtividade Dos 1007/978-1-84800-044-5_3.
Desenvolvedores de Organizaciones de Software. Universidade Federal Do Kornyshova, E., Barrios, J., 2020. Industry 4.0 impact propagation on enterprise
Amazonas., Retrieved from https://tede.ufam.edu.br/bitstream/tede/6137/5/ architecture models. Procedia Comput. Sci. 176, 2497–2506.
Tese_Edson C. C. Oliveira.pdf. Kosa, M., Yilmaz, M., 2015. Designing games for improving the software, 1,
Curtis, B., Krasner, H., Iscoe, N., 1988. A field study of the software design process 303–310. http://dx.doi.org/10.1007/978-3-319-24647-5.
for large systems. Commun. ACM 31 (11), 1268–1287. http://dx.doi.org/10. Kropp, M., Meier, A., Anslow, C., Biddle, R., 2020. Satisfaction and its correlates
1145/50087.50089. in agile software development. J. Syst. Softw. 164, http://dx.doi.org/10.1016/
de Barros Sampaio, S.C., 2010. Uma meta estraté. Universidade Federal de j.jss.2020.110544.
Pernambuco. Licorish, S.A., MacDonell, S.G., 2014. Personality profiles of global software de-
de Barros Sampaio, S.C., Barros, E.A., De Aquino, G.S., Carlos E. Silva, M.J., velopers. In: Proceedings of the 18th International Conference on Evaluation
De Lemos Meira, S.R., 2010. A review of productivity factors and strategies and Assessment in Software Engineering. pp. 1–10.
on software development. In: Proceedings - 5th International Conference on Licorish, S.A., MacDonell, S.G., 2015. Communication and personality profiles of
Software Engineering Advances. ICSEA, 2010, pp. 196–204. http://dx.doi.org/ global software developers. Inf. Softw. Technol. 64, 113–131. http://dx.doi.
10.1109/ICSEA.2010.37. org/10.1016/j.infsof.2015.02.004.
Donoso Barraza, Í., Vega Zepeda, V., 2017. Factores sociales y humanos que Licorish, S., Philpott, A., MacDonell, S.G., 2009. Supporting agile team composi-
afectan el proceso de educción de requerimientos: Una revisión sistemática. tion: A prototype tool for identifying personality (in) compatibilities. In: 2009
RISTI - Revista Iberica de Sistemas E Tecnologias de Informacao (24), 69–83. ICSE Workshop on Cooperative and Human Aspects on Software Engineering.
http://dx.doi.org/10.17013/risti.n.69-83. pp. 66–73.
Dutra, A.C.S., Prikladnicki, R., França, C., 2015. What do we know about high Lloret-Segura, S., Ferreres-Traver, A., Hernández-Baeza, A., Tomás-Marco, I., 2014.
performance teams in software engineering? Results from a systematic El análisis factorial exploratorio de los ítems: una guía práctica, revisada y
literature review. In: 41st Euromicro Conference on Software Engineering actualizada. Anales de PsicologÍ 30 (3), 1151–1169.
and Advanced Applications. http://dx.doi.org/10.1109/SEAA.2015.24. López-Fernández, D., Yagüe Panadero, A., 2011. Factor humano en el desarrollo
Fatema, I., Sakib, K., 2018a. Factors influencing productivity of agile soft- de software: Motivando a un ingeniero. In: XVI Jornadas de IngenierÍa Del
ware development teamwork: A qualitative system dynamics approach. Software Y Bases de Datos. JISBD2011, pp. 1–6.
In: Proceedings - Asia-Pacific Software Engineering Conference. APSEC, Machuca-Villegas, L., Gasca-Hurtado, G.P., 2019. Towards a social and human
2017-Decem, pp. 737–742. http://dx.doi.org/10.1109/APSEC.2017.95. factor classification related to productivity in software development teams.
Fatema, I., Sakib, K., 2018b. Using qualitative system dynamics in the devel- In: Mejia, J., Muñoz, M., Rocha, Á., Calvo-Manzano, J.A. (Eds.), Trends and
opment of an agile teamwork productivity model using qualitative system Applications in Software Engineering. International Conference on Software
dynamics in the development of an agile teamwork productivity model. Int. Process Improvement. CIMPS 2019, In: Advances in Intelligent Systems and
J. Adv. Softw. 11, 170–185. Computing, Springer, Cham, pp. 36–50. http://dx.doi.org/10.1007/978-3-030-
Fernández-Sanz, L., Misra, S., 2011. Influence of human factors in software 33547-2_4.
quality and productivity. In: Lecture Notes in Computer Science (Includ- Machuca-Villegas, L., Gasca-Hurtado, G.P., Morillo Puente, S., Re-
ing Subseries Lecture Notes in Artificial Intelligence and Lecture Notes strepo Tamayo, L.M., 2021a. An instrument for measuring perception
in Bioinformatics), vol. 6786, pp. 257–269. http://dx.doi.org/10.1007/978-3- about social and human factors that influence software development
642-21934-4_22, LNCS(PART 5). productivity. J. UCS 27 (2), 111–134. http://dx.doi.org/10.3897/jucs.65102.
Franca, C., Da Silva, F.Q.B., Sharp, H., 2020. Motivation and satisfaction of Machuca-Villegas, L., Gasca-hurtado, G.P., Muñoz, M., 2021. Measures re-
software engineers. IEEE Trans. Softw. Eng. 46 (2), 118–140. http://dx.doi. lated to social and human factors that influence productivity in software
org/10.1109/TSE.2018.2842201. development teams. Int. J. Inf. Syst. Project Manag. in press.
Franco, E.F., Hirama, K., Carvalho, M.M., 2018. Applying system dynamics ap- Machuca-Villegas, L., Gasca-Hurtado, G.P., Puente, S.Morillo., Re-
proach in software and information system projects: A mapping study. Inf. strepo Tamayo, L.M., 2021b. Data from the study of perceptions of social
Softw. Technol. 93, 58–73. http://dx.doi.org/10.1016/j.infsof.2017.08.013. and human factors that influence the productivity of software development
Galeano-Ospino, S., Machuca-Villegas, L., Gasca-Hurtado, G.P., 2020. Knowledge teams in Colombia. Mendeley Data http://dx.doi.org/10.17632/sbhbrbykw8.1.
transfer in software development teams using gamification: A systematic. Machuca-Villegas, L., Gasca-Hurtado, G.P., Restrepo Tamayo, L.M., Mo-
In: Mejia, J., Muñoz, M., Rocha, Á., Quiñonez, Y. (Eds.), New Perspectives in rillo Puente, S., 2020. Social and human factor classification of influence in
Software Engineering. CIMPS 2020, In: Advances in Intelligent Systems and productivity in software development teams. In: Yilmaz, M., Niemann, J.,
Computing, pp. 115–130. Clarke, P., Messnarz, R. (Eds.), Systems, Software and Services Process
23
Improvement. EuroSPI 2020. Communications in Computer and Information Sharp, H., 2007. Social and human aspects of software engineering. In:
Science. Springer, Cham., Düsseldorf, Germany, pp. 717–729. http://dx.doi. Selby, R.W, Basili, V.R., Rombach, D., Schneider, K., Kitchenham, B., Pfahl, D.
org/10.1007/978-3-030-56441-4_54. (Eds.), Empirical Software Engineering Issues. Critical Assessment and Future
Maisiri, W., Darwish, H., Van Dyk, L., 2019. An investigation of industry 4.0 skills Directions. In: Lecture Notes in Computer Science, Vol. 4336, Springer, Berlin,
requirements. South Afr. J. Ind. Eng. 30 (3), 90–105. Heidelberg, p. 40. http://dx.doi.org/10.1007/978-3-540-71301-2_13.
Martinez Arias, R., 1996. Psicometrí. Madrid: Síntesis. Sommerville, I., 2016. Software Engineering, tenth ed. Pearson.
Matturro, G., 2013. Soft skills in software engineering: A study of its demand Storey, M.A., Zimmermann, T., Bird, C., Czerwonka, J., Murphy, B.,
by software companies in Uruguay. In: 2013 6th International Workshop Kalliamvakou, E., 2019. Towards a theory of software developer job
on Cooperative and Human Aspects of Software Engineering, CHASE 2013 - satisfaction and perceived productivity. IEEE Trans. Softw. Eng. 1–18.
Proceedings. pp. 133–136. http://dx.doi.org/10.1109/CHASE.2013.6614749. http://dx.doi.org/10.1109/TSE.2019.2944354.
Matturro, G., Fontán, C., Raschetti, F., 2015a. Soft skills in scrum teams. a survey Stray, V., Moe, N.B., Hoda, R., 2018. Autonomous agile teams: Challenges and
of the most valued to have by product owners and scrum masters. In: future directions for research. In: ACM International Conference Proceeding
Proceedings of the 27th International Conference on Software Engineering Series, Part F, vol. 1477, pp. 1–5. http://dx.doi.org/10.1145/3234152.3234182.
and Knowledge Engineering. 2015(August), pp. 42–45. http://dx.doi.org/10. Tomasello, M., 2010. Por QuÉ Cooperamos?, Vol. 2030, Primera ed. Katz Editores.
18293/seke2015-026. Vallejo-Nágera, J., 2002. GuÍ. Ediciones Temas de Hoy.
Matturro, G., Raschetti, F., Fontán, C., 2015b. Soft skills in software development Wagner, S., Deissenboeck, F., 2019. Defining productivity in software engineering.
teams a survey of the points of view of team leaders and team members. In: Sadowski, C., Zimmermann, T. (Eds.), Rethinking Productivity in Software
http://dx.doi.org/10.1109/CHASE.2015.30. Engineering. pp. 29–38. http://dx.doi.org/10.1007/978-1-4842-4221-6.
Matturro, G., Raschetti, F., Fontán, C., 2019. A systematic mapping study on soft Wagner, S., Ruhe, M., 2008. A systematic review of productivity factors in
skills in software engineering. JUCS 25 (1), 16–41. software development. In: Proceedings of the 2nd International Software
Méndez Martínez, C., Rondón Sepúlveda, M., 2012. Introducción al análisis Productivity Analysis and Cost Estimation. SPACE 2008, pp. 1–6.
factorial exploratorio. Revista Colombiana de PsiquiatrÍ 41 (1), 197–207. Yilmaz, M., 2013. A software process engineering approach to understand-
http://dx.doi.org/10.1016/S0034-7450(14)60077-9. ing software productivity and team personality characteristics: An em-
Morales-Aguiar, N., Vega-Zepeda, V., 2018. Factores humanos y la mejora de pirical investigation. http://dx.doi.org/10.1145/291712.291759, Dublin City
procesos de software. Propuesta inicial de un catálogo que guíe su gestión. University.
RISTI - Revista IbÉ (29), 30–42. http://dx.doi.org/10.17013/risti.29.30-42. Zamora, R., Esnaola, J., 2015. AnáLisis Factorial Y anáLisis de Componentes
Muñoz, M., Peña, A., Mejia, J., Rangel, N., Torres, C., Hernández, L., 2018. Building Principales. Universidad de Chile, Santiago de Chile.
high effectives teams using a virtual environment. In: European Conference
on Software Process Improvement. pp. 554–564.
Murphy-hill, E., Jaspan, C., Sadowski, C., Shepherd, D., Phillips, M., Winter, C.,
Liliana Machuca-Villegas is a System Engineer from Universidad Francisco
et al., 2019. What predicts software developers ’ productivity? IEEE Trans.
de Paula Santander, Colombia. She received the M.Sc. in Engineering from
Softw. Eng. 1–13.
Universidad del Valle, Colombia. Currently she is PhD from Universidad de
Olgun, S., Yilmaz, M., Clarke, P.M., O’Connor, R.V., 2017. A systematic investi-
Medellín, Colombia.
gation into the use of game elements in the context of software business
She is assistant professor at Escuela de Ingeniería de Sistemas y Computación
landscapes: a systematic literature review. In: International Conference on
at the Universidad del Valle, Colombia. Her main research areas include Project
Software Process Improvement and Capability Determination. pp. 384–398.
Management, Gamification, and Software Engineering.
Oliveira, E., Conte, T., Cristo, M., Valentim, N., 2018a. Influence factors in software
productivity - a tertiary literature review. In: Proceedings of the 30th Inter-
national Conference on Software Engineering and Knowledge Engineering, Gloria Piedad Gasca-Hurtado is titular professor-researcher in the Engineer-
Vol. 2018. pp. 68–103. http://dx.doi.org/10.18293/seke2018-149. ing Faculty of Universidad de Medellín. Her Ph.D. was taken in Universidad
Oliveira, E., Conte, T., Cristo, M., Valentim, N., 2018b. Influence factors in software Politécnica de Madrid, Spain in Languages, Informatics Systems and Software
productivity - a tertiary literature review. Int. J. Softw. Eng. Knowl. Eng. 28 Engineering Department in Informatics Faculty. Her research areas include
(11–12), 1795–1810. http://dx.doi.org/10.1142/S0218194018400296. Information technology and communications (TIC) software process improve-
Paiva, E., Barbosa, Jr., D.R.L., Albuquerque, A., 2010. Factors that influence the ment and optimization, multi-model environment for software development,
productivity of software developers in a developer view. Innov. Comput. Sci. software development and agile methodologies applied to small enterprises
Softw. Eng. http://dx.doi.org/10.1007/978-90-481-9112-3. (SME’s), security informatics, among others. She serves as Director of Software
Petersen, K., 2011. Measuring and predicting software productivity: A systematic Engineering Academic Program.
map and review. Inf. Softw. Technol. 53 (4), 317–343. http://dx.doi.org/10.
1016/j.infsof.2010.12.001. Solbey Morillo Puente holds BA in Early Childhood Education (1991), Magister
Pirzadeh, L., 2010. Human Factors in Software Development : A Systematic Scienteae in Education and Doctor in Education from the University of Los
Literature Review (Master of Science Thesis in Computer Science and Andes (Universidad de Los Andes ULA-Mérida, Venezuela). I am member of
Engineering). p. 368. the research group "Education, Society and Peace", of the Faculty of Social and
Project Management Institute, and IEEE Computer Society, 2013. Software Human Sciences of the University of Medellín (UdeM). I work as a fulltime
®
Extension To the PMBOK Guide, fifth ed. Project Management Institute, professor and researcher at UdeM. I have been the tutor of several undergraduate
Inc. degree projects in Criminology and more than 30 postgraduate theses in the
Ruiz, M., Salanitri, D., 2019. Understanding how and when human factors are master’s degrees in ICT and Master’s in Education at UdeM.
used in the software process: A text-mining based literature review. In: Publications in specialized magazines and book chapters related to juvenile
Lecture Notes in Computer Science (Including Subseries Lecture Notes in delinquency, educational management and social and human factors that influ-
Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11915 LNCS, ence productivity. Area of specialization statistical data analysis and quantitative
pp. 694–708. http://dx.doi.org/10.1007/978-3-030-35333-9_54. research methodology.
Sadowski, C., Zimmermann, T., 2019. Rethinking productivity in sof tware engi-
neering. Apress open, http://dx.doi.org/10.1007/978-1-4842-4221-6. ISBN-13
Luz Marcela Restrepo-Tamayo is a professor of industrial engineering at the
(electronic): 978-1-4842-4221-6.
Universidad de Medellín and Ph.D. student in engineering at the same university.
Sánchez-Gordón, M., Colomo-Palacios, R., 2019. Taking the emotional pulse of
She is working in the research field of statistical tools to manage software
software engineering — A systematic literature review of empirical studies.
projects. Before becoming a Ph.D. student, she worked in statistical applications
Inf. Softw. Technol. 115 (July), 23–43. http://dx.doi.org/10.1016/j.infsof.2019.
in the industry, particularly in companies of mass consumer products and
08.002.
appliances. She also has a bachelor’s degree in industrial engineering and a
Sanchez-Gordón, M.-L., Colomo-Palacios, R., Herranz, E., 2016. Gamification and
master’s degree in Science - Statistics. She has published some peer-reviewed
human factors in quality management systems: Mapping from octalysis
articles in journals, conference, and workshop proceedings.
framework to ISO 10018. In: 23rd European Conference on Systems, Software
and Services Process Improvement, Vol. 1. EuroSPI 2016, pp. 234–241. http:
//dx.doi.org/10.1007/978-3-319-44817-6.
24

Survey Development

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Survey Development

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Survey Development

Uploaded by

Copyright:

Available Formats

The Journal of Systems & Software 192 (2022) 111408

Contents lists available at ScienceDirect

The Journal of Systems & Software

Perceptions of the human and social factors that influence the

Team Cohesion • ‘‘Coordination’’ • Team Cohesion • The definition of team

Innovation • ‘‘Task variety and • ‘‘People on my project are • Creativity

Leadership style • ‘‘Team Leadership’’ Self-confidence

(continued on next page)

Of the 13 factors that make up the Perceptions Scale, only two

Results obtained by extraction method: Principal component analysis.

Extraction Method: Principal Component Analysis.

Extraction method: Principal component analysis.

The third component, ‘Project management skills using tools’, 6. Discussion

You might also like