• No results found

Researching the conflicts between user experience, front-end and back-end in software development process

N/A
N/A
Protected

Academic year: 2021

Share "Researching the conflicts between user experience, front-end and back-end in software development process"

Copied!
28
0
0

Loading.... (view fulltext now)

Full text

(1)

Västerås, Sweden

Thesis for the Degree of Master of Science in Computer Science,

Software Engineering 15.0 credits

RESEARCHING THE CONFLICTS

BETWEEN USER EXPERIENCE,

FRONT-END AND BACK-END IN

SOFTWARE DEVELOPMENT PROCESS

Dejana Tomasevic

dtc20002@student.mdh.se

Tea Pavicevic

tpc20002@student.mdh.se

Examiner: Federico Ciccozzi

Mälardalen University, Västerås, Sweden

Supervisor: Alessio Bucaioni

Mälardalen University, Västerås, Sweden

(2)

every stage of the research project. Additionally, we would like to sincere thanks to research participants that they set aside time for taking part in this research, especially employees from

(3)

Abstract

User experience design is the process of improving the accessibility and use of a product during user’s interaction with it. This study investigates the issues that occur in the coordination of UX design and software development. Furthermore, it examines diverse types of conflicts between UX designers, front-end and back-end developers, factors contributing to these conflicts and their influence on the software development process. The method used in this study is a survey conducted in an online form with a target group of practitioners. The data show that task conflicts are the most common type of conflicts in teams, that gender of a person can influence its awareness of the project status and that age and geographical location do not affect the occurrence of the identified conflicts.

(4)

Table of Contents

1 Introduction 1 2 Background 2 2.1 User Experience . . . 2 2.2 Front-end development . . . 3 2.3 Back-end development . . . 3

2.4 Software development process . . . 3

3 Research method 4 3.1 Research questions . . . 5

3.2 Respondents selection . . . 6

3.3 Survey definition . . . 6

3.4 Data collection and analysis . . . 7

3.5 Threats to validity . . . 7

4 Results 8 4.1 What types of conflicts can arise between UX designer, front and back-end devel-oper? (RQ1) . . . 8

4.1.1 Do socio-cultural factors favor the rising of the identified conflicts? (RQ1.1) 9 4.1.2 Does the geographic distribution of the team members favors the identified conflicts? (RQ1.2) . . . 11

4.2 Do the identified conflicts affect the success of the project? (RQ2) . . . 11

5 Discussion 13

6 Related Work 17

7 Conclusion and Future Work 19

(5)

List of Figures

1 User Experience . . . 2

2 Research process . . . 4

3 Type of conflicts . . . 8

4 Difficulties in synchronization between UX design and back-end development . . . 9

5 Difficulties in synchronization between front-end and back-end development . . . . 9

6 Participants‘ age . . . 10

7 Participants‘ gender . . . 10

8 Projects status awareness . . . 10

9 Teamwork type . . . 11

10 Same timezone of the team . . . 11

11 Influence of conflicts on a project success . . . 11

12 Influence of conflicts on a project failure . . . 11

13 Participants specialization . . . 13

14 The main specialization for males . . . 13

15 The main specialization for females . . . 13

16 Work experience . . . 14

17 Duration of work with team . . . 14

18 The form of communication . . . 15

(6)

List of Tables

1 Literature review . . . 5 2 Experience in other specializations . . . 14

(7)

Acronyms

HCI Human Computer Interaction

IBM International Business Machines Corporation ISO The International Organization for Standardization RQ Research question

RQs Research questions UI User Interface UX User Experience

(8)

1

Introduction

Ian Sommerville defines software engineering as an engineering discipline, which deals with theo-ries, methods and tools for professional software development [1]. The ultimate goal of software engineering is to produce software applications (e.g., systems, services), which comply to a set of requirements. Requirements come from all the stakeholders involved in the development process such as users, customers, developers. User requirements not only affect the functionality of the software application under development, but also quality aspects related to the user-software ap-plication interaction [2]. The User Experience (UX) is a process whose sole objective is to design software applications, which can deliver a satisfactory user experience [2]. UX is a complex process, which uses different theories and methods from several disciplines, e.g., Human Computer Interac-tion (HCI) and psychology, focuses on different quality aspects such as usability, accessibility, and involves many stakeholders, e.g., designers and developers [2]. As Ries wrote, “no amount of design can anticipate the many complexities of bringing a product to life in the real world” [3]. Besides UX design, software development process needs to include two more crucial processes which are front-end and back-end development. All of these processes together are beneficial for project success [4].

One of the main complexities related to developing a software product is the communication among the different stakeholders, i.e., designers and developers, involved in the software develop-ment process. In fact, a lack or an excess of communication may lead to problems such as omission, commission, incomplete communication, and misunderstanding [5]. For instance, it might happen that information is not properly communicated between designers and developers or that the in-formation is misunderstood [5]. For instance, a designer may define a conceptual solution that is optimal for the user, but not practically viable for the developers [4]. An ineffective commu-nication may also lead to a lack of trust and transparency between teams and to an ineffective leadership that can cause an unclear division of roles and work tasks [5], [6]. Other complexities are conflicts which can be classified as task and social conflicts according to Marisa K. Wilson. In her study she describes how the lack of understanding of the roles and responsibilities can influence the emergence of conflicts as well, but states that these types of conflicts can be diminished by communicating more and preferably in person [6]. There exists a consistent body of literature examining conflicts related to communication in teams. However, other socio-cultural factors that lead to conflicts between team members are far less researched.

In this thesis, we investigate the types and the sources of conflicts, which can arise between UX designers, front-end and back-end developers in a software development process. We examine if different socio-cultural factors (i.e., age, gender or origin) have an influence on the identified conflicts. Besides, we examine whether the geographical location and time zone of members in distributed development teams affect the conflicts and the overall work process. Eventually, we investigate if conflicts have a positive or negative influence on the software development process and the final result.

The research method we use for this study is an online survey conducted with voluntary par-ticipants [7]. The survey consists of 33 questions being a mix of closed and open-ended questions grouped in 5 sections. The target group of respondents is composed of 56 software engineers who have experience in UX design, front-end or back-end development.

We analyze the obtained data by performing chi-squared tests and vertical analyses. The results show that the most prevalent type of conflicts is task conflicts. Team members having different specializations experience different types of conflicts. While UX designers experience all types of conflicts, developers experience task conflicts, mostly. Geographical location does not affect the identified conflicts. Gender of team members plays a major role when it comes to members’ awareness of project status, which can favor the emergence of problems in teamwork. Eventually, we observe that improved communication and team training could contribute to reducing these conflicts.

The remainder of this thesis is structured as follows. The following section presents background for this thesis. It contains UX, front-end and back-end development as relevant concepts. Section3 presents detailed description of research method used in the thesis. Section4 contains answers to the RQs obtained from the survey. Additionally, Section5includes additional findings by analysing the data collected through the survey. Section7closes the study with a conclusion and future work.

(9)

2

Background

In this section, we review some of the relevant concepts used in the remainder of this thesis.

2.1

User Experience

User Experience (UX) refers to the way users communicate with a product. The International Organization for Standardization (ISO), defines UX as “the perception and reaction of a person as a result of the use or expected use of a product, system or service” [8]. In 1973, during a lecture at the University of Pennsylvania, Thomas J. Watson (the founder of International Business Machines Corporation (IBM)) said that a “good design is a good busines” [9]. The main goal of UX is to determine how the representation of the product would work in the best possible way possible to satisfy user needs and business goals (Figure1). UX is often uses knowledge from ergonomics and psychology.

Figure 1: User Experience

With the development of the Internet and its applications (online stores, interactive websites and software, etc.) there has been a necessity to transfer the UX into the digital world. In the web domain, the UX design is the process of improving the accessibility and use of a product during the user interaction. It refers to the communication between human and machine happening on all the segments of the product that the user interacts with [2]. Currently, UX design plays a crucial role in the creation of an ideal software product or service.

In order to meet the users expectations, UX designers need to plan for an easy and intuitive navigation of the product they use [10]. In his research, Mike Volpe found that 76% of 174 web users stated that the most important factor in a website is the easiness and expeditiousness to find what they look for [11]. Deng et al. came to similar conclusion and stated that the general requirement for UX is that users must always be able to achieve what they intended by using the product [12]. To this end, it is crucial that designers must perform an appropriate analysis of the problem and recognize potential design challenges in order to create an adequate solution which will be tested in practice.

User Interface (UI) focuses on how the product will look and function. UI is the ´´space” where users interact and communicate with the product or with each other. This interaction creates an experience for the user. Creating a user interface requires to know the user, to adapt the use of the product, and to create a successful aesthetic while respecting some standards. According to Donald A. Norman, there are several steps in the UI creation: researching, goal defining, designing, prototyping, testing and finally implementing [13]. To conclude, UX and UI are highly dependent to each other and they are both to create a successful UX [14].

(10)

2.2

Front-end development

The web is based on the client-server communication model. When users access websites using their web browsers, what they see as text, images and video is a representation of the content hosted on a remote web server. This means that their browsers represent the clients [15]. The front-end development deals with the presentation of the web content to the users via their web browser. What is important to note is that, although front-end development deals with the visible and interactive side of a website, it does not focus on how the appearance of the content. Front-end developers do not design the actual layout of the website: this is the task of the designers. Front-end developers task is to turn designer ideas into a functional website [16].

2.3

Back-end development

The back-end development of a website deals with communicating with the front-end side by sending and receiving data, organising and storing data, security and all the other aspects that user can not see. Back-end developers are able to create a dynamic website, the content of which can change depending on what is its purpose. Unlike a static website, dynamic websites do not need a database because its content mostly remains the same [16].

2.4

Software development process

Software systems and services are ubiquitous in our lives. The requirements on these systems and services are ever growing. Nowadays, their correct functionality is an essential requirement together with innovative application interfaces that are easy to use. This leads to the necessity to develop this systems and services in a collaborative manner, involving UX designers from the beginning of the development process [4].

However, the interplay of UX designers and developers may generate issues related to not being able to find common solutions meeting both designers and developers needs, perspectives and possibilities [17]. In some cases, developers are not able to fulfil the requirements set by designers due to lack of knowledge or complexity of the design [17] In turns, this negatively affect the quality of the final products [18].

In this context, one of the main issue with developing current software systems is to balance the design and the development so that the final result is the best possible outcome for the end user [17]. Several works show that one successful way of achieving such a balance is to improve the quality of communication among designers and developers [18]. The communication flow between designers and developers should be consistent and undisturbed. Both designers and developers should show special openness and willingness to accept different opinions. Unfortunately, there is a consistent body of literature that argues that this is not the case for most of the projects [18].

Other works investigate different factors that can contribute to the occurrence of complications in synchronizing UX design and development. These factors include lack of transparency and trust between team members, ineffective leadership, lack of motivation, different personality traits, etc [17].

(11)

3

Research method

The research in this thesis has been designed and carried out following the guidelines for online surveys in software engineering [7]. Figure 2 depicts the research process we have followed. It consists of nine steps being: research the literature, define research questions, respondents selection, initial pilot survey, final survey, data collection and analysis, results and report.

Figure 2: Research process

In the research literature step, we have collected the relevant literature related to conflicts among user experience designers, front-end and back-end developers during the software devel-opment process. We have used the findings from the literature for defining the thesis research questions and the questions composing the survey. The output of this step has been the set of relevant works listed in Table1.

In the define the research questions step, we have detailed the research questions that have driven this thesis. We have defined the research questions starting from the literature elicited in the previous step. We have used the research questions for defining the pool of respondents, the research method, the choice of data and the basis for interpretation. In the respondents selection step, we have defined the target group for our research. That is, software engineers with experience in UX design, back- or front-end development. We have defined the survey using a two-steps process, which comprises of the initial pilot survey and the final survey steps. In the initial pilot survey step, we have designed a first draft of the survey using Google Forms1and have

tested it on a control group of six respondents. The aim of this step has been to evaluate the survey and uncover any possible issue or limitation. The output of this step has been a set of possible improvements on the survey. In the final survey step, we have refined the initial draft using the feedback from the previous step and submit it to the pool of respondents. In the data collection and analysis step, we have gathered the answers from the respondents and analysed them for providing answers to the research questions. We have provided both quantitative and qualitative

(12)

analysis using comparative and statistical analyses of the collected data. In the results step, we have used the synthesised data for answering the research questions. In the last step, report, we have described and presented our observations and conclusions in this report. The surveys and the collected data are available athttps://github.com/pavte/master_thesis . This research has

been conducted according to ethical guidelines. In particular, we have not collected or stored any personal information of the participants. All the collected data are anonymised. We have made an effort to inform the participants about the scope and the objective of this research. This research has not been funded or sponsored by any third-party institution and we have not pressured or coerced respondents into taking part in the survey.

ID Authors Title Year

1 J. Huffman Hayes Do you like piña coladas? How

improved communication can improve software quality

2003 2 M. K. Wilson Examining the conflict between user

ex-perience and software development in industry

2019 3 J. Nieters, A. P. Uday and M. Shankar Great user experiences require great

front-end development 2012

4 S. T. Acuña, M. N. Gómez and D. L.

Juan Empirical study of how personality,team processes and task characteristics relate to satisfaction and software qual-ity

2008

5 B. Jerome and R. Kazman Surveying the solitudes: An investiga-tion into the relainvestiga-tionships between hu-man computer interaction and software engineering in practice

2005

6 E. Ries How Today’s Entrepreneurs Use

Con-tinuous Innovation to Create Radically Successful Businesses

2011 7 R. Abreu and R. Premraj How developer communication

fre-quency relates to bug introducing 2009 8 R. R. Patrashkova-Volzdoska, S. A.

Mc-Comb, S. G. Green and W. D. Compton Examining a curvilinear relationshipbetween communication frequency and team performance in cross-functional project teams

2003

9 T. Javed, M. Maqsood and Q. S.

Dur-rani A Survey to Examine the Effect ofTeam Communication on Job Satisfac-tion in Software Industry

2004

Table 1: Literature review

3.1

Research questions

We have used the selected studies for driving the design of the research questions. In this thesis, we are going to seek for answers to the following Research question (RQ).

RQ1 What types of conflicts can arise between UX designer, front and back-end developer? The main goal of this RQ is to provide a catalogue of conflicts arising among user experience designers, front-end and back-end developers.

RQ1.1 Do socio-cultural factors (e.g., gender, age or origin) favor the rising of the identified conflicts?

Personality traits can have a pivotal role in promoting the rise of conflicts in teams [5] [19]. However, socio-cultural traits are less assessed in the literature.

(13)

RQ1.2 Does the geographic distribution of the team members favors the identified conflicts? A study by Partashkova-Volzdoska et al. [20] examined the relationship between cross-functional team performance and communication frequency. In the research, they in-cluded both located and co-located teams. Their results showed that face-to-face com-munication is directly connected to goal achievement and that email is the most used communication method. With this research question, we investigate if the team’s geo-graphical location and time zone affect the conflicts and the overall software development process.

RQ2 Do the identified conflicts affect the success of the project?

The effects of conflicts on the software development process and on the software system under development are controversial and debated. With this research question, we investigate if and how the identified conflicts affect, either positively or negatively, the software development process as well as the software system under development.

3.2

Respondents selection

The survey has been conducted with voluntary participants. The participants had to have proven experience in user experience design, front-end or back-end development. To this end, we have selected the respondents using LinkedIn2 as it provides insights into the experience of each

in-dividual. Besides, we have selected further respondents using our academic supervisor contact network. In total, we have contacted 84 respondents via email and LinkedIn from which 56 have filled out the survey. The pilot survey has been conducted with six respondents with the intention to evaluate whether the questionnaire was clearly and comprehensibly designed. We have selected respondents for the pilot testing who are employees from two software development companies, Amplitudo3and WIND Ltd4.

3.3

Survey definition

The survey has been conducted online using a free survey tool being Google Forms. Google Forms is a user-friendly tool which provides insights into the survey data. We have constructed the survey to collect anonymous responses in order to obtain the data entered by each participant without compromising security of their information.

We have performed the pilot survey with the intention of improving the quality of the final sur-vey. We have had six participants in the pilot testing of which four were back-end developers, one was a front-end developer and one was a user experience designer. All of them were employees of Amplitudo and WIND Ltd. One suggested improvement was to mark the questions as mandatory: originally the questions were designed in a way that they could have been unanswered. Another improvement was adding open-ended questions where respondents could answer with comments, stories or experiences. Additionally, one participant has suggested to specify whether or not the questions were referring to a single project or targeting the whole professional life of the respon-dents.

The final questionnaire has contained of 33 questions of which one is strictly an open-ended question. We have also included close-ended questions that have been structured in a form of open-ended questions with additional text fields. We have mostly used close-ended questions because of their simplicity, rate of response and their unlikeliness to give undisclosed answers. We have structured the final questionnaire in five sections. We have started the survey by informing the participants about the goal of the survey and how we were going to use its results. The first section of the questionnaire contains questions regarding the general information of the participants. The second section contains questions related to the work role and years of experience of the participants. The third section contains questions regarding the overall experience of working in a team. We have focused mainly on communication quality in teams and the acknowledgement and recognition of ideas and suggestions of each team member. The fourth section contains questions

2https://www.linkedin.com/ 3https://amplitudo.me/ 4http://wind.co.me/en/

(14)

referring to the conflicts that can occur in a team, such as task, relationship and value conflicts. To avoid misunderstandings, we have complemented the questions with the definitions of these conflicts.

The last section of the questionnaire is an optional section where participants can leave their thoughts, stories or comments regarding the topics addressed in the survey.

3.4

Data collection and analysis

We have received 56 responses to the survey. Out of the 56 respondents, 26 have identified as front-end developers, 19 as back-end developers and 11 as UX designers. We have extracted the data using the pre-built Google Forms services. In particular, we have exported the data into a Microsoft Excel table, from which various analyzes can be performed. Open-ended questions, that have been structured in a form of additional text fields, have been clarified as qualitative data. The last question has collected 10 additional answers and they have been analyzed individually. The obtained data has been qualified as attributive-qualitative variables. We have conducted comparative analyses using chi-squared tests with cross-classification tables. The chi-squared test is a practical test, which can be useful in analysing whether some of the observed frequencies deviate from the frequencies that we would expect under a certain hypothesis [21], [22]. For statistical analysis, we have used vertical analysis in order to represent percentage insight of the results.

3.5

Threats to validity

The main threat to external validity is the representation of the population. In particular, one concrete risk for this research is not having an adequate number of respondents, which could prevent the generalization of the results. To reach a large sample, which would help us in generalising our results, we have reached potential participants through several channels including LinkedIn and Reddit. The final number of respondents was 56.

Another concrete risk is the subject selection as the survey was targeting three categories being front and back-end developers, and user experience designers. To minimise this risk, we have tried to have an approximate number of respondents belonging to each category. Additionally, we have included respondents that have experience in two of listed specializations. We have had 26 of the respondents specialized in front-end development, 19 in back-end development and 11 in user experience design.

Concerning threats to internal validity, one risk is misinterpretation of questions. For reducing this risk, we have provided a detailed description along with instructions for each question of the survey that could be misinterpreted. Regrading the threats to conclusion validity, another risk was the overall quality of the survey. For minimizing this threat, we have conducted a literature review and based survey on the obtained data. The main reason was to minimize the author’s biases. Besides, we have defined the survey using a two-step process, which involved a control group of respondents. In the pilot survey, we have introduced two questions that explicitly addressed the clarity and completeness of the survey. This has helped us in further improving the quality of the survey and reducing threats to internal validity. Finally, we have included open-ended questions in the final survey for allowing respondents to better describe their thoughts.

Additionally, one threat to construct validity is relation between the theory and the observa-tions. To minimize this threat, we have included the participants that have had proven experience in user experience design, front-end or back-end development.

(15)

4

Results

In this section, we provide answers to the RQs by analysing the data collected through the survey. The results were analyzed used vertical analysis and chi-squared tests with cross-classification tables. Vertical analysis have been performed using the Google Forms services. We have analyzed the results through an Excel spreadsheet that Google Forms automatically exports. Chi-squared tests are used for finding correlations between obtained data, which in our case are attributive-qualitative variables observed as random events. The first step of the analysis is to set up two hypotheses that contradict each other. The first hypothesis assumes that the connection between the events does not exist. Next, we have to observe two types of frequencies (observed and expected frequency) in order to get an estimate value of the extent of the discrepancy between them. The lower the estimate, the more likely it is that the first hypothesis is correct and vice versa. This part of the analysis uses the results from the sample. It is known from probability theory that the first hypothesis represents a random variable that has a distribution with a certain degree of freedom. This distribution for a given level of significance has a critical value. We have used 99,9% as a degree of freedom because it gives the most credible results. [22] If the estimate is lower than critical value, first hypothesis is correct, i.e. the correlation between events A and B does not exist. If the estimate is higher than critical value, second hypothesis is correct, i.e. the correlation between events A and B exists [21], [22].

The results from the questionnaire that are not related to any RQ are discussed in Section5.

4.1

What types of conflicts can arise between UX designer, front and

back-end developer? (RQ1)

In the literature review, we have found that the most common conflicts that can arise between UX designer, front and back-end developer are task conflicts. However, this might be because other types of conflicts are less assessed in the literature. According to literature review, conflicts that are examined were divided into task and social conflicts. In our study, we wanted to divide social conflicts in more precise categories of conflicts. We have have represented task conflicts as differ-ences based on work details and goals. Corresponding to RQs, we have came up with relationship and value conflicts. We have identified relationship conflicts as differences in personality, style and taste and value conflicts as fundamental differences in identities and values, which can include differences in politics, religion, ethics, norms, etc. We have given the participants the option to select multiple answers, hence all the options they relate to. Figure 3 shows the results for the mentioned types of conflicts.

Figure 3: Type of conflicts

The most common type of conflicts that has been experienced by participants is task conflict with 75.4% of the answers. This confirms the results of the studies from the literature review. Value and relationship conflicts have a significantly smaller number of responses. Relationship conflicts have 12.3% answers, while value conflicts have not been considered as the solely cause

(16)

of conflict. Other results are classified in: 7% of answers for both task and relationship conflicts, 3.5% for all types of conflicts and 1,8% for task and value conflicts.

Using chi-square tests, we have examined whether a specialization of a person has an influence on identified conflicts. We have set up two hypotheses: the first is that the specialization of a person does not have an influence on the identified conflicts, while the second is that the specialization has an influence on the identified conflicts. We have got that the estimate value is 14.19 and the critical value is 13.82. As the estimate value is higher than critical value, it is accepted that a specialization of a person can have an influence on the identified conflicts with a very high probability - a confidence level of 0.999. UX designers differ from the other two specializations in experiencing the type of conflicts since the responses vary. Only 54% of UX designers, believes that task conflicts affect the work of the team. On the contrary, developers predominantly believe that task conflicts are the most common ones (77% front-end developers and 89% Back-end developers). The data collected in this thesis seem to indicate that the specialization of a team member affect the type of experienced conflicts with developer experiencing mainly task conflicts and UX designer equally experiencing task and other type of conflicts.

Regarding task conflicts between UX design and front-end development, we have focused on synchronization between UX design and back-end development. Figure4 shows that 57.1% of the respondents experienced difficulties in synchronization between UX design and front-end develop-ers. Difficulties in synchronization between front-end and back-end development are experienced by 75% of respondents and figure5shows the results.

Figure 4: Difficulties in synchronization

be-tween UX design and back-end development Figure 5: Difficulties in synchronization be-tween front-end and back-end development

4.1.1 Do socio-cultural factors favor the rising of the identified conflicts? (RQ1.1) In this thesis, we have examined whether age or gender of team members favor the rising of the identified conflicts. Figure 6 represents the age ranges of respondents who participated in the survey. The ranges vary, with 55.4% of respondents that are between 25 and 34 years old, 37.5% of respondents that are between 20 and 24 years old, 5.4% of respondents that are between 35 and 44 years old and only one respondent who is between 45-54 years old. No one of the participants is under 20 or over 55 years. Concerning the gender of the participants, 67.9% of them identify as a male, 30.4% as a female and one participant does not identify as male nor as female. Figure7 represents the results.

Using chi-square tests, we have examined the influence of team members’ age on the identified conflicts. We have set up two hypothesis. The first hypothesis is that there is no correlation between the age of a team member and the type of conflict, while the second hypothesis is contradictory. The estimate value for this examination is 0,791 and the critical value is 10,83. As the estimate value is lower than critical value, it is accepted that the age of a team member does not favor the identified conflict with a confidence level of 0.999. Approximately 81% team members aged 20 to 24 and 73.52% aged 25 to 44 years answered that task conflict is the main conflict that occurs in the team. With a large percentage, which is significantly higher than 50%, we have concluded that task conflicts are the main conflicts that occur in the team for all age ranges.

(17)

Figure 6: Participants‘ age Figure 7: Participants‘ gender

team. The first hypothesis is that there is no correlation between the gender of a team member and the type of conflict, while the second hypothesis is contradictory. The estimate value for this examination is 1,52 and the critical value is 10,83. As the estimate value is lower than critical value, it is accepted that the gender of a team member does not favor the identified conflict with a confidence level of 0.999. With a large percentage, which is significantly higher than 50%, we have concluded that task conflicts are the main conflicts that occur in the team for all genders.

With the obtained data, we also have examined if the gender can have an influence on receiving proper recognition and acknowledgement in the team. We have assumed that the awareness of the projects status affects the conflicts within the team. Firstly, using vertical analysis we have analyzed the close-ended question related to this assumption. The question investigated if partic-ipants have been in a situation that they were not aware of the status of the project. Figure 8 shows the results. All of the respondents, that have been in a situation that were not aware of the projects‘ status, have experienced conflicts in the team.

Figure 8: Projects status awareness

The first hypothesis is that there is no correlation between the gender of a team member and the awareness of the projects‘ status, while the second hypothesis is contradictory. The estimate value for this examination is 27,58 and the critical value is 10,83. As the estimate value is higher than critical value, it is accepted that the gender of a team member can have an influence on receiving proper recognition and acknowledgement in the team with a confidence level of 0.999. Approximately 71% of female team members out of the total number of females from the sample believe that they were not aware of the project status. Thus, this analysis shows that the gender of team members affects the perception of the problem with projects‘ status awareness. Similarly, non-awareness of the project status and gender of team members can favor identified conflicts.

(18)

4.1.2 Does the geographic distribution of the team members favors the identified conflicts? (RQ1.2)

Figure 9 shows that 57.1% of participants have worked locally and 42.9% remotely. Figure 10 shows that 86.8% of respondents who have worked remotely have been located in the same time zone as their teams, while the remaining 13.2% of respondents have not.

Figure 9: Teamwork type Figure 10: Same timezone of the team Only one respondent that has worked remotely experienced relationship conflicts, while the rest have experienced task conflicts. Using chi-square test, we have concluded that geographic distribution of the team members does not favor the identified conflicts. The estimate value for this examination is 1.27 and the critical value is 10,83. As the estimate value is lower than critical value, it is accepted that geographic distribution of the team members does not favor the identified conflicts with a confidence level of 0.999. Task conflicts are the most typical type of conflicts between teams regardless of geographic distribution. Since Partashkova-Volzdoska et al. [20] concluded that email is the most used communication method between co-located teams, we have investigated what are the new technology channels for communication between them. We discuss this in Section5.

4.2

Do the identified conflicts affect the success of the project? (RQ2)

With this question, we have investigated the impact of the identified conflicts on the success of the project.

Figure11displays the influence of the mentioned conflicts on the outcome of the project. 77.8% of the respondents have answered that the conflicts have had a positive impact, while 22.2% of them have stated that the conflicts have had a negative impact on the success of a project. According to 89.1% of respondents, conflicts have not caused a project failure, whereas 10.9% of respondents stated that the conflicts they experienced have lead to a failure (Figure12). Only one respondent has experienced relationship conflict that led to failure, while the remaining have experienced task conflicts.

Figure 11: Influence of conflicts on a project

(19)

Using chi-squared tests, we have confirmed that identified conflicts can not directly led to project failure. The first hypothesis is that there is no correlation between the identified conflict and project failure, while the second hypothesis is contradictory. The estimate value for this examination is 0.165 and the critical value is 10,83. As the estimate value is significantly lower than critical value, it is accepted that identified conflicts can not lead directly to project failure with a confidence level of 0.999. Similarly, the results have shown that identified conflicts can not rigorously have positive or negative influence on a project success. These results could be motivation for further examinations and researches on this topic.

(20)

5

Discussion

In this section, we present additional findings obtained from the survey.

Figure 13 shows the specialisation of the the respondents. According to the data: 46.4% of participants are back-end developers, 33.9% are front-end developers and 19.6% are UX designers.

Figure 13: Participants specialization

Gender does not affect the rise of conflicts in the team, but it affect the receiving of proper information regrading projects status. We have investigated whether there is a correlation between the gender of the team members and their main specialization using chi-squared tests. Figures14 and15show the most common specialisations for men and women, respectively. From the charts,

Figure 14: The main specialization for males Figure 15: The main specialization for fe-males it is evident that gender has an influence on the specialization. The predominant specialization among females is UX designer with 52.9% of participants. Front-end developer (53.8%) and back-end developer (43.6%) are predominant specializations among males. For chi-squared tests, we have set up two hypotheses. The first hypothesis assumes that the gender of a person can not have an impact its main specialization, while the second hypothesis is the opposite. The estimate value for this examination is 20.15 and the critical value is 13,82. As the estimate value is significantly higher than critical value, the second hypothesis is accepted with a confidence level of 0.999. Hence, we can state that the main specialization of a person depends on their gender.

60.71% of the participant have had experiences in other specializations. In particular, 7 de-velopers have had experience in UX design (Table2), 3 UX designers and 13 back-end developers have had experience in front-end development (Table 2) while 16 front-end developers have had experience in back-end development (Table2).

As future work for her paper, Marisa K. Wilson wanted to examine whether the years of experience of a could affect the presence of conflicts [6]. Figure 16 shows the results of vertical analysis of participants work experience in our study.

(21)

UX design Front-end development Back-end development

UX designer x 3 0

Front-end developer 6 x 16

Back-end developer 1 13 x

Table 2: Experience in other specializations

Figure 16: Work experience

Using chi squared-tests, we have examined whether team members with different years of work experience in software development experience different type of conflicts. The first hypothesis is that work experience does not favor identified conflicts and the second hypothesis is contradictory. The estimate value for this examination is 1.46 and the critical value is 16,27. As the estimate value is notably lower than critical value, the first hypothesis is accepted. The conclusion is that work experience does not affect the identified conflicts with a confidence level of 0.999. Additionally, using chi-squared tests we have investigated whether team members that have worked less with the team experience more conflicts than team members that have worked longer. Figure17shows the results from the questionnaire for how long the participants worked together with a team. The first hypothesis is that no matter how long the team worked together, all teams experience the same type of conflict, while the second hypothesis is contradictory. The estimate value for this examination is 12.69 and the critical value is 13,83. As the estimate value is lower than critical value, the first hypothesis is accepted with a confidence level of 0.999. The conclusion is that all teams have experienced task conflict the most.

Figure 17: Duration of work with team

Communication can be the key for both conflicts and success in a team. Forms of communica-tion can vary depending on the team. Partashkova-Volzdoska et al. performed a study on local and distributed teams and their results showed that face-to-face communication is the most meticulous

(22)

method to communicate in order to achieve a given goal [20]. When it comes to distributed teams, they concluded that email is the most used communication method. In our thesis, we have discov-ered new technology channels for communication by analyzing the results using vertical analysis. First, we have set up an open-ended question where we asked the main communication means used within a team. The question had multiple given options and an additional field for other responses. Participants were able to choose all the options that apply to them. Figure18shows the results.

Figure 18: The form of communication

The most significant result is that 5.56% of participants communicate only through meetings in person. The remaining participants have combined in person meetings with online meetings, email or Skype. All of the participants that are in teams spread across different time zones communicate via email or online meetings. Other channels that are used are Slack, WhatsApp, Discord and Rocket chat. We believe that the COVID-19 pandemic might have contributed in increasing the use of digital communication channel.

Figure19shows the main factors that may contribute to conflict avoidance. The best solution for avoiding conflicts is better communication among team members (50.9% of participants). Other possible solutions are training (38.6% of participants), more members (8.8% of participants) and more time (1.79% of participants).

Figure 19: Contribution to teamwork during software development process

As mentioned, in the last part of the questionnaire, the respondents have had the opportunity to give their comments, stories and thoughts about the addressed subject. One respondent, who has describe itself as a designer, has described a situation of disagreement due to developers inability to implement the prototype as it was originally planned and it has emphasized the importance of putting users in the first place during the development process. The respondent has also mentioned that personality type and individuality play a major role in conflicts occurrence.

The respondents have stressed the importance of learning how to communicate in a profes-sional manner when working in teams. They have emphasized the fact that different suggestions should always be taken into consideration and need to be discussed, so that no team member feels neglected or less important. According to one front-end developer, the conflicts can also arise

(23)

as a consequence of lacking of communication caused by poor leadership and undefined responsi-bilities. UX designers have stressed the significance of learning development terminology, which could improve communication between team members. Eventually, the respondents agreed that conflicts are always an issue in development process, but that they can also lead to positive effects. However, in this case, their effect can be determined only afterwards.

(24)

6

Related Work

In this section, we analyse some of the research in the field of software engineering, which focuses on conflicts between user experience designers and developers.

Marisa K. Wilson conducted a study that examines the conflicts between user experience and software development in industry [6]. Wilson stated that the conflicts in the software development process are more common when user experience designers and developers work together on the same project. She observed that the probability of conflicts is inversely related to the number of individuals who communicate: the higher the number of members involved in the communication, the lower the possibility of conflicts. Concerning the type of conflicts, Wilson divided them into social and task conflicts. In her research, she found that all of the conflicts were task conflicts. Besides poor communication, another frequent source of conflicts Wilson observed is the the lack of understanding of roles and responsibilities. She noticed that such conflicts can have a negative impact on interpersonal relationships, but they can improve the software product quality and stated that one way of reducing these conflicts is by preferring face-to-face communication [6]. Build on this, we have concluded that better communication and some training would be another ways to reduce the conflicts. Eventually, Wilson concluded that software projects were remarkably improved when developers collaborated with designers [6]. Compared to the work by Wilson, the work in this thesis use the same research method, i.e., online survey, though we have used one survey only while Wilson had one survey for designers and another one for developers. In this thesis, we have divided the possible conflicts in three main categories being task, relationship and value conflicts (as opposite to Wilson who used only two groups). Regardless of the number of categories, we have also concluded that the most common type of conflicts are task conflicts. Eventually, in our thesis we have focused on understanding whether or not years of experience might have a stake on the conflicts creation. Wilson did not focus on this aspect and referred to it as a possible future work.

Acuña et al. [19] conducted an empirical study using quasi-experiment for connecting team members personalities (e.g., neuroticism, extroversion, openness to experience, agreeableness and conscientiousness, characteristics of the tasks), autonomy and team processes (e.g., cohesion and conflict). The purpose of this study was to evaluate the connection between the quality of the software under development and relationships between team members. The participants were second-year computing students that were taking Data Structure and Algorithms course. The authors observed that extroversion has a positive impact on software quality. Although motivation and some personality traits can contribute in increasing conflicts between designers and developers, these can be mitigated with a good team cohesion [19]. Similarly to Wilson, Acuña et al. also observed that task conflicts have a more negative impact on the final software product than social conflicts [19]. Unlike these two studies, who focused on the impact of personality traits in arising the conflicts, we have examined socio-cultural traits. Besides, the main difference between our thesis and the work of Acuña et al. is that we have used survey and we have only targeted professionals with proven experience in software development (rather than students).

Jerome’s and Kazman’s [23] study investigated to which extent previous findings of studies in Human Computer Interaction (HCI) could be valid in industry. Hence, their study involved pro-fessionals from several IT organizations (e.g., British HCI Group and the International Software Engineering Research Network).The results showed that the findings were not applicable. They observed that there is a lack of understanding between disciplines and roles and that misunder-standings among designers and developers have a negative effect on the project. Furthermore, they noticed that there is a lack of cohesion in the team when members start to collaborate too late. An interesting finding of Jerome’s and Kazman’s study is that 68% of developers stated that software decisions were made only by them and without consulting the human interaction team [23]. Con-sidering that these decisions can influence the overall product in later stages, developers need to see collaboration with designers as a complementary benefit in the project [4]. Our thesis extends their findings on the misunderstanding as we have analysed that there are more misunderstandings between front-end and back-end developers than between user experience and developers. Besides, our results have confirmed that conflicts are less frequent when team member collaborate from the beginning of the project.

(25)

the connection between communication among developers and identified bugs in the project. To this end, they used the dates of commits that have bugs and the dates of the emails extracted from the archive. The results showed a positive correlation between communication frequency and number of injected bugs in the software. They noticed that developers communicated more when there were bugs in the project. They conclude that healthy communication can contribute to the quality of the software. Our thesis builds on their conclusion and tries to prove whether earlier engagement and appreciation of everybody ideas can reduce project failures.

Javed et al. [25] investigated the effect of various factors that contribute in job satisfaction of software development teams. To analyse their hypothesis, they considered 23 factors and used Multiple Regression Analysis technique. The results showed that 4 of 23 factors (working envi-ronment, quality information provided, quality work and performance) have positive influence on job satisfaction. In our thesis, we have discussed how the overall awareness of the project status (which may depend on information exchange quality) can contribute to projects success. When it comes to the performance, we have noticed that adjustments that were willingly performed had a positive impact on resolving the conflicts.

Patrashkova-Volzdoska et al. [20] shown that the relationship between communication frequency and team performance is curvilinear. More precisely, communication can have positive impact on team performance only if there is the right amount of it. In our thesis, the results have shown that better communication can contribute in avoiding conflicts in the team. Patrashkova-Volzdoska et al. conducted an empirical investigation using Likert scales on co-located and non-co-located teams. When it comes to the co-located teams, they concluded that the email is the only communication method. Our research complements this statement considering new communication media.

(26)

7

Conclusion and Future Work

This study is conducted in order to examine the potential conflicts that can occur between UX designers, front-end and back-end developers during the software development process.

We have investigated different types of conflicts in teamwork, the main factors that lead to these conflicts and the effect that these conflicts have on the work process and final result. The method we have used for this research is an online survey which has been performed with voluntary participants. The results we have derived from the obtained data show that the most identified conflicts are task conflicts. Age, gender and geographical location of a team member do not favor identified conflicts, but gender has an influence on receiving proper information of project’s status. Furthermore, the results have shown that no matter how long the team has worked together and how much work experience each team member has had, all teams experienced task conflicts the most. Also, the results have shown that team training and communication improvement can minimize these conflicts. We have discovered new channels for communication for teams that are spread across different time zones which are Slack, WhatsApp, Discord and Rocket chat. We have also found that gender of a team member has an impact on their specialization.

In the future, we could include more participants in the survey in order to gather more data. We would primarily focus on finding UX designers, as well as persons who have had more ex-perience in working with teams distributively in different time zones. Furthermore, the survey could be performed by conducting interviews with intention to collect more in depth qualitative data. Additionally, potential future work of this research could include performing experiments, e.g. examining the outcome of different teams working on the same project. The outcome could be analyzing the final product of the projects and examining under what conditions the product is more successful, e.g. product which has been developed by team that worked locally or by the team that consist of people with more work experience.

(27)

References

[1] I. Sommerville, Software Engineering, 7th ed. Addison Wesley, 2004.

[2] M. Soegaard, The Basics of User Experience Design. Interaction Design Foundation, 2002. [3] E. Ries, The Lean Start-up: How Today’s Entrepreneurs Use Continuous Innovation to Create

Radically Successful Businesses, 1st ed. Crown Business, New York, 2011.

[4] T. Springer and P. Boon, “Usability first – from the frontend to the backend,” 2014, User-Centred Software Design.

[5] J. Huffman Hayes, “Do you like piña coladas? how improved communication can improve software quality,” 2003, IEEE Software.

[6] M. K. Wilson, “Examining the conflict between user experience and software development in industry,” M.S. thesis, Rochester Institute of Technology, 2019.

[7] T. Punter, M. Ciolkowski, B. Freimut, and I. John, “Conducting on-line surveys in software engineering,” 2003, Fraunhofer IESE, Sauerwiesen 6, D-67661 Kaiserslautern, Germany. [8] International Organization for Standardization, Ergonomics of humansystem interaction

-part 210: Human-centered design for interactive systems (formerly known as 13407), 2019. [9] IBM, Good design is good business,https://www.ibm.com/ibm/history/ibm100/us/en/

icons/gooddesign/, Online; Accessed 30-April-2021.

[10] D. Lesaca, Understanding ux design & front-end development,https://www.codefellows. org/blog/understanding- ux- design- front- end- development/, Online; Accessed

02-February-2021, 2014.

[11] M. Volpe, “What do 76% of consumers want from your website?,” 2011, HubSpot.

[12] L. Deng, D. E. Turner, R. Gehling, and B. Prince, “User experience, satisfaction, and con-tinual usage intention of it,” 2010, European Journal of Information Systems.

[13] D. A. Norman, The design of everyday things, Revised and expanded edition. Basic Books, 2013.

[14] User Testing, What’s the difference between user interface and user experience? https : //www.usertesting.com/blog/ui-vs-ux, Online; Accessed 02-February-2021, 2018.

[15] T. F. Iskandar, M. Lubis, T. Kusumasari, and A. R. Lubis, “Comparison between client-side and server-side rendering in the web development,” 2020, IOP Conference Series Materials Science and Engineering.

[16] R. M. Almuttairi, “What’s the difference between the frontend and back-end?,” 2017, Uni-versity of Babylon journals.

[17] J. M. Six, “The integration of user experience into software development,” 2011, UXMatters. [18] J. Nieters, A. P. Uday, and M. Shankar, “Great user experiences require great front-end

development,” 2012, UXmatters.

[19] S. T. Acuña, M. N. Gómez, and d. L. Juan, “Empirical study of how personality, team processes and task characteristics relate to satisfaction and software quality,” 2008, ESEM, New York.

[20] R. Patrashkova-Volzdoska Ralitza, A. McComb Sara, G. Green Stephen, and D. W. Compton, “Examining a curvilinear relationship between communication frequency and team perfor-mance in cross-functional project teams,” 2003, IEEE Software.

[21] D. S. Shaver and Z. Zhiyi, Beginning Statistics, 1st ed. Creative Commons, 2012. [22] C. Butler, Statistics in Linguistics, 1st ed. Oxford: Blackwell, 1985.

[23] B. Jerome and R. Kazman, “Surveying the solitudes: An investigation into the relationships between human computer interaction and software engineering in practice,” 2005, Springer. [24] R. Abreu and R. Premraj, “How developer communication frequency relates to bug intro-ducing changes,” 2009, Department of Computer Science, VU University Amsterdam, The Netherlands.

(28)

[25] T. Javed, M. Manzil E, and S. D. Qaiser, “A survey to examine the effect of team communi-cation on job satisfaction in software industry,” 2004, National University of Computer and Emerging Sciences, 852-B, Faisal Town, Lahore, Pakistan.

References

Related documents

We will implement our solution to each matching iteration problem with a specific method. In this phase, we are going to plan the solution of each matching method

It presents information about the Art_Value concept, online auctions, user experience design and front end development tools.. 2.1

A pre-feasibility study is a preliminary systematic assessment of all critical elements of the project – from technologies and costs to environmental and social impacts. It is

We have developed Sensors framework using Android API, which gives latest value of all possible sensors used in mobile phones, and notify end user programming about sensor

Visitors will feel like the website is unprofessional and will not have trust towards it.[3] It would result in that users decides to leave for competitors that have a

How does the development team define and capture end user needs?, As part of answering the MRQ, RQ2 is answered by describing how the different projects investigated work

One study suggests all five of the big five personality types are related to preceded job strain, where Neuroticism is associated with high job strain and Openness,

In order to benefit as much as possible from the experience factories Pine and Gilmore (1999) suggest that it is important that the whole experience supports the company and