MOTIVATING SOFTWARE ENGINEERS

A Theoretically Reflective Model

Nathan Baddoo, Sarah Beecham, Tracy Hall

School of Computer Science, University of Hertfordshire, College Lane, Hatfield, Herts AL10 9AB, U.K.

Hugh Robinson, Helen Sharp

Dept of Computing, Faculty of Mathematics and Computing, The Open University, MK7 6AA, U.K.

Keywords: Motivation, model, theory, software engineer, systematic literature review.

Abstract: We present a model of motivation for software engineers. Our model suggests that software engineers are

motivated by two sets of factors, intrinsic and extrinsic motivators, where a subset of intrinsic motivators

are aspects inherent in the job that software engineers do. It shows that software engineers are orientated

towards these particular sets of motivators because of their characteristics, which in turn are mediated by

individual personality traits and environmental factors. Our model shows that the external outcomes of

software engineers’ motivation are benefits like staff retention, increased productivity and reduced

absenteeism. Our model is derived from a Systematic Literature Review of motivation in software

engineering. We have constructed this model by engaging in practices that reflect good principles of model

building as prescribed by operational research and scientific management discipline. We evaluate our model

for theoretical efficacy and show that our model, in comparison to other attempts at modeling software

engineers’ motivation, reflects

a wide range of the classic concepts that underpin the subject area of

motivation. We argue that, this theoretical efficacy validates the model and therefore improves

confidence in its use. We suggest that our model serves as a valuable starting point for managers

wanting to understand how to get the best out of software engineers, and individuals wanting to

understand their own motivation or who are embarking on career choice.

1 INTRODUCTION

In this paper we present a model of motivation for

software engineers. We derive this model from a

Systematic Literature Review (SLR) of software

engineers’ motivators. We follow model-building

principles from the discipline of operational research

and scientific management to construct this model

and evaluate the model in terms of how it reflects

classic theories of motivation. We show that our

model is strongly corroborated by existing

phenomena in the organizational psychology

literature on motivation.

Our model of motivation is based on findings

from a systematic literature review of 92 published

papers. This means it is rigorously underpinned by

previous work in the area. Our model shows that

there is a complex array of factors that must be

managed effectively to get the best outcomes from

software engineers. Our model also reflects some

classic concepts of motivation, which is an

important outcome of this work, since our related

work suggests that theories are not used well in

existing studies of motivation of software engineers

(Hall et al 2007).

Motivation is increasingly cited as a particularly

pernicious people problem in software engineering.

In DeMarco and Lister’s 1999 survey motivation

was found to be one of the most frequently cited

causes of software development project failure

(DeMarco & Lister 1999). The Standish report

(1994) amplifies this finding by reporting that

having access to competent, hard working and

focused staff is one of 10 success criteria for

software projects. However, until now no

comprehensive model of what motivates software

engineers has been developed. Consequently it is

difficult for managers to know how best to motivate

their software engineers.

174

Baddoo N., Beecham S., Hall T., Robinson H. and Sharp H. (2008).

MOTIVATING SOFTWARE ENGINEERS - A Theoretically Reﬂective Model.

In Proceedings of the Tenth International Conference on Enterprise Information Systems - HCI, pages 174-180

DOI: 10.5220/0001685601740180

 SciTePress

In Section 2, we provide some background on the

value of models. We also present an overview of

motivation theories. In Section 3, we describe our

research method. In Section 4, we present the

findings of our SLR, detailing how these findings

underpin the development of our model. In Section

5, we present our model and evaluate it in terms of

how it reflects classic and conventional concepts of

motivation in. We conclude and discuss future work

in Section 6.

2 THE VALUE OF MODELS

Any model derived to examine a particular area of

study or practice should reflect the classic theories in

that area. Indeed, work that we have done shows that

some models are actually representations of theory.

We provide a summary of theories of motivation. An

extensive overview of these theories is published by

Hall et al (2007). We will, in later sections, reflect

the model presented in this paper in the light of these

theories.

2.1 Why are Models Important?

Models are used in all fields of software

engineering, from requirements engineering to

software evolution either to explore possible

consequences of an action before taking that action

or as embedded parts of a system to aid in routine

decision making (Pidd 1999). Overall, our

exploration for a model to represent software

engineers’ motivation aims at deriving some form of

external representation of the dynamics of how

software engineers are/can be motivated. Such a

model will be important to the people who manage

software engineers in terms of helping them to

manager software engineers better.

2.2 Concepts of Motivation

A good model of motivation should address some

classic theories of motivation. There are eight

theories that specifically address how people are

motivated within an organizational context. We

provide an overview of these as follows

2.2.1 Job Characteristics Theory

The Job Characteristics Theory (JCT) states that

there are certain key characteristics present within a

job that makes it motivational to practitioners

(Hackman and Oldham, 1976). These key

characteristics are classified into five ‘core

dimensions’, which are Skill variety, Task identity,

Task significance, Autonomy and Feedback. The

extent to which these five job dimensions motivate

practitioners is dependent on his need for personal

growth and development; Growth Need Strengths

(GNS) (Hackman and Oldham, 1976).

2.2.2 Stimulus Response Theory

Stimulus Response Theory (SRT) describes the

activities that modify behaviour (Skinner, 1976).

These activities are termed stimuli. According to

SRT there are two types of stimuli: punitive and

rewarding stimuli. The theory explains that punitive

stimuli are easier to apply and do have the effect of

producing the required responses in the short term.

However, rewarding stimuli, which are more

difficult to apply and require more ingenuity to

devise, tend to have a longer-term effect in inducing

the correct responses from subjects (Skinner, 1976).

2.2.3 Equity Theory

Adam’s Equity Theory (1963), in an organization

context, is concerned with how to make employees

feel equitably treated in an organisation. It states that

the inputs that people bring into an activity or

organisation, that is their experience, education,

skills and seniority, should be matched by the

outputs that is what they get from that activity or

organisation, which are salary, recognition,

opportunity for achievement etc.

2.2.4 Needs Theory - Maslow

Maslow’s hierarchy of needs roughly translates that

different types of needs motivate people at different

stages in their lives (Maslow, 1954). Such needs

manifest in a hierarchy where physical needs are at

the bottom of the hierarchy and self-actualisation

comes at the top.

2.2.5 Need Theory McClleland

McClelland’s needs theory identifies three

motivational needs: achievement, authority and

affiliation (McClelland 1961). It states that each

individual has a combination of these needs in

various levels of strength. So that the individual

whose need mix is strongly biased by affiliation will

tend to be a more objective in order to increase their

opportunity for bonding with the most number of

people.

MOTIVATING SOFTWARE ENGINEERS - A Theoretically Reflective Model

175

The interesting distinction between McClelland’s

theory and Maslow’s is the absence of the

hierarchical structure on which these needs manifest.

However, it is possible to assume that with time, an

individual’s needs mix will change and assume

different profile depending on where he finds

himself.

2.2.6 Motivation-Hygiene Theory

Herzberg’s motivation hygiene theory classifies

factors that motivate practitioners into two distinct

sets: Extrinsic factors and intrinsic factors.

• Extrinsic factors are those that are external

to the job that practitioners do.

• Intrinsic factors, on the other hand, are the

primary determinants of motivation and

satisfaction.

2.2.7 Goal Setting Theory

This theory states that goals that are hard to achieve,

when accepted, lead to better performance by the

people doing them than goals that are easy to

achieve (Locke 1968). However, in order to do this,

the goal needs to be very well defined, made specific

and measurable, and feedback provided so that the

person tackling the goal will know when it is

achieved.

2.2.8 Expectancy Theory

Much of Vroom’s (1964) expectancy theory is based

around the notion that a individual’s motivation to

engage in certain activity is predicated by the degree

or amount of positive outcomes that he expects from

this activity.

3 RESEARCH METHOD

We analysed the research problem in relation to two

main question areas:

• How are software engineers motivated?

• What models of motivation exist in software

engineering to explain how software

engineers are motivated?

We conducted a SLR, following the guidelines of

Kitchenham (2004). We formulated the following

set of research questions:

RQ1: What are the characteristics of Software

Engineers?

RQ2: What (de)motivates Software Engineers to

be more (less) productive?

RQ3: What are the external signs or outcomes of

(de)motivated software Engineers?

RQ4: What aspects of Software Engineering

(de)motivate Software Engineers?

RQ5: What models of motivation exist in Software

engineering

The rest of the procedure followed in the SLR can

be found in (Beecham et al 2006).

3.1 Overview of Model Construction

Process

Pidd et al suggest that a good model must be simple.

Such simplicity can be derived from how transparent

the model turns out to be. Transparency, in turn, can

be achieved by how well defined the problem is. In

our model construction the problem we explore is

tightly defined by the research questions formulated

for the SLR. RQ1 to RQ4 provide us with a clear

definition of the building blocks to the overall

question on what motivates software engineers. The

aim is that the solutions to the questions would

provide the rationale for and the parameters for

building the model. The overall structure of the

model itself will be derived from similarities of

other tried and tested work.

The rationale for building our model, in light of

the research that indicates that models are important

in helping us think and reflect about a phenomenon

before acting, is to establish whether such a model

already exists. In the context of this research, this

line of enquiry is catered for by the inclusion of

Research Question 5, above, in the SLR. The

answers to RQ5 should help us understand what the

gaps are in the area of models of software engineers’

motivation.

The next stage is to simplify the problem by

decomposing it into simple components. This

process involves a lot of knowledge of the problem

area. This requires that in formulating questions that

will provide solution to the line of enquiry, we also

needed to know how that problem area could be

decomposed into simple modules. In doing this, we

are also ensuring that should certain components of

the resultant enquiry be found to be erroneous, they

can be replaced without having to replace or re-do

the entire model.

This simplification through decomposition is

provided by the breakdown of the research enquiry

into software engineers’ motivation into four

research question. Figure 1 illustrates this point.

ICEIS 2008 - International Conference on Enterprise Information Systems

176

Figure 1: Framework of research strategy for SLR

(Beecham et al 2007).

The resultant data from the chain of enquiry

represented in Figure 1 allowed us to parameterise

the components of the individual modules of the

emerging model and the overall structure of the

model was then derived from analogies and

similarities with other tried and tested work on

generic concepts of motivation and by analysing

other models of motivation in software engineering.

We explain the tried and tested work and other

models of motivation in Section 5, when we present

and discuss the model.

The final stage in the model construction

exercise was to ascertain how the resultant model

reflected both classic and conventional theories of

motivation.

4 WHAT MOTIVATES

SOFTWARE ENGINEERS?

In the following section, we present findings from

the systematic literature review to the Research

Questions 1 to 5 presented earlier.

4.1 What are the Characteristics of

Software Engineers?

Table 1 shows that the SLR reports a software

engineer to be growth oriented, i.e. he is an

individual who likes challenges and likes to learn

new skills. It also shows that the SLR reports a

software engineer to bean introverted individual

with low need for social strengths. The SLR also

reports a software engineer to be autonomous,

creative and technically competent.

Table 1: Software Engineer Characteristics (Beecham et

al 2007).

Software Engineer Characteristics %

Ch.4 Growth oriented (e.g. challenge, learn new skills) 9

Ch.6 Introverted (low need for social interaction) 8

Ch.11 Autonomous (need for independence) 8

Ch.1 Need for stability (organisational stability) 5

Ch.3 Achievement oriented (e.g. seeks promotion) 4

Ch.12 Need for variety 4

Ch.14 Need for challenge 4

Ch.16 Need to be sociable/identify with group/organisation 4

Ch.2 Technically competent 3

Ch.5 Need for competent supervising 3

Ch.8 Need for feedback (needs recognition) 2

Ch.10 Need to make a contribution (job is worthwhile) 2

Ch.13 Marketable 2

Ch.15 Creative 2

Ch.7 Need for involvement in personal goal setting 1

Ch.9 Need for Geographic stability 1

4.2 What (de)Motivates Software

Engineers to be more (less)

Productive?

Table 2 shows that some of the most widely reported

motivators of software engineers from the SLR are

the ability to identify with the tasks, employee

participation, career paths, good management,

variety of work and a sense of belonging.

Table 2: What motivates Software Engineers? (Beecham

et al 2007).

Motivators in Software Engineering %

M.17 Identify with the task (clear goals, personal interest, know

purpose of task, how it fits in with whole, job satisfaction;

identifiable piece of quality work)

M.10 Employee participation/involvement/working with others 17

M.4 Career Path (opportunity for advancement, promotion

prospect, career planning)

M.6 Good management (senior management support, *team-

building, good communication)

M.3 Variety of Work (e.g. making good use of skills, being

stretched)

M.7 Sense of belonging/supportive relationships 15

M.1 Rewards and incentives (e.g. scope for increased pay and

benefits linked to performance)

M.12 Recognition (for a high quality, good job done -different

to M1 which is about making sure that there are rewards

available).

M.2 Development needs addressed (e.g. training opportunities

to widen skills)

MOTIVATING SOFTWARE ENGINEERS - A Theoretically Reflective Model

177

Table 2: What motivates Software Engineers? (Beecham

et al 2007)(cont).

Motivators in Software Engineering %

M.11 Feedback 11

M.15 Technically challenging work 11

M.16 Job security/stable environment 11

M.18 Autonomy 9

M.8 Work/life balance (flexibility in work times, caring

manager/employer, work location)

M.21 Making a contribution/task significance (degree to which

the job has a substantial impact on the lives or work of other

people)

M.5 Empowerment/responsibility 5

M.19 Appropriate working conditions/environment/good

equipment/tools

M.14 Trust/respect 4

M.13 Equity 3

M.9 Working in company that is successful (e.g. financially

stable)

M.22 Sufficient resources 2

4.3 What are the External Signs

or Outcomes of (de)Motivated

Software Engineers?

Table 3 shows that the most widely reported

outcome of motivated software engineers is

retention. Other external outcomes are

improvements in project delivery time and

productivity, adherence to budgets, low absenteeism

and improved project success.

Table 3: External signs of (de)motivated software

engineers (Beecham et al 2007).

External signs of motivated and

de-motivated software engineers

% of studies

Ext1: Retention 11%

Ext2: Project delivery time 2%

Ext3: Productivity 5%

Ext4: Budgets 1%

Ext5: Absenteeism 1%

Ext6: Project Success 1%

4.4 What are Aspects of Software

Engineering (de)Motivate Software

Engineers?

Table 4 shows that the variety of aspects reported

from SLR ranged from problem solving, change,

challenging nature, science to the experimental

aspect of the discipline.

Table 4: Motivational aspects of software engineering

(Beecham et al 2007).

Motivating Aspects of software engineering field %

Asp1: Problem Solving (the process of understanding and

solving a problem in programming terms)

Asp2: Team Working 2

Asp3: Change 4

Asp4: Challenge (Software Engineering is a challenging

profession and that in itself is motivating)

Asp5: Benefit (creating something that is of benefit to

someone or enhances well-being)

Asp6: Science (making observations, identifying,

describing, engineering, investigating and theorising,

explaining a phenomena)

Asp7: Experiment (trying something new, experimentation

in order to gain experience):

Asp8: Development practices (Object Orientated, XP and

prototyping practices)

Asp9: Software process/lifecycle – Software development,

project initiation and feasibility studies, and maintenance

(note maintenance was also found a de-motivating activity)

4.5 What other Models of Motivation

Exist in Software Engineering?

Table 5 provides a summary of the different models

of motivation and job satisfaction developed

specifically for the Software Engineering. It shows

that almost all the models draw on and build upon

some classic theory of motivation, however, none of

the models substantially encapsulates all the factors

that underpin motivation. In effect, none of the

models reflects all the concepts of motivation

expressed in the literature as presented in Section 2.3.

Table 5: Models for motivating SEs (Beecham et al 2007).

Models of motivation #

1: Job Characteristics Model (JCM) of Software Engineer

(SE) Motivation (development, enhancement or validation)

2: Models focusing on Software Engineer Job Satisfaction 6

3: Models of Open Source Developer SE Motivation 3

4: Models of leadership influence on SE motivation 3

5: Model drawing on expectancy theory, goal-setting theory,

and organizational behaviour specific to the software

development process

6: Model of Task Design influence on SE motivation 1

7: Model of Career Progression influence on SE motivation 1

8: Social support influence on Software Engineer turnover 1

Overall, we suggest that though these models

above provide valuable insight into software

engineers’ motivation, they are disparate and do not

ICEIS 2008 - International Conference on Enterprise Information Systems

178

encompass all the factors that our understanding of

the literature leads us to expect of a model on

motivation.

5 MODEL CONSTRUCTION

PROCRESS

In Section 3, we presented the framework used to

explore the issues in this research. This framework

summarised the research strategy used to frame our

research questions. From this framework, we were

able to evolve a model of motivation, based on the

answers to the Research Questions 1 to 4. We

present our model of motivation as a series of stages

in a process. In this model, for example, we classify

a software engineer’s motivation as two sets of

factors: aspects inherent in the job that software

engineers do, for example the problem solving

nature of software engineering and some general

factors, which can also be sub-categorised into

extrinsic and intrinsic factors as illustrated in Fig. 2.

Figure 2: Model of motivation in software Engineering.

Our model also presents a set of characteristics of

software engineers. It shows that these

characteristics are mediated by individual

personality traits and environmental factors. We

suggest that these characteristics orientate software

engineers towards their motivators, particularly, the

set of inherent job characteristics. Our model shows

that software engineers’ motivation leads to external

outcomes like staff retention, increased productivity

and reduced absenteeism.

Having established the overall structure of the

model, it can be parameterized with data from the

SLR that we have presented in Section 4.

5.1 Refining the Model

A final stage in our model construction phase was to

compare our model with the other models of

motivation in software engineering in order to refine

the structure and dynamics of our model. Our

analysis of the structure of these other models

showed that the relationships between components

are more complex than Figure 2 suggests. For

example, we were able to discern that contextual

factors have a direct effect on motivators and how

effective they are. It also became clear that the

balance between organisational intrinsic and

extrinsic motivators and the motivators inherent in

software engineering have an effect on software

engineers’ characteristics, and their reactions to

different motivators. The full discussion of these

finding are presented in (Sharp et al 2007). The

results of this analysis led us to refine our initial

model. Figure 3 presents the refined model.

Figure 3: Refined model of motivation in software

engineering.

Overall, the model we present in Figure 3 is a

process model because it explains the dynamics of

motivation in a series of steps and stages as shown

earlier. We suggest that this is so because the

majority of studies in our SLR from which we have

derived the above model, concentrated more on

process theories than content theories. However, the

notion of a process theory is not mutually exclusive

from that of a content theory, so the model we have

will also reflect some content theories of motivation

as will be demonstrated next.

5.2 Theoretical Efficacy

The dynamics of the model presented here and the

constituents of the components of the model strongly

reflect three classic motivation theories of Job

MOTIVATING SOFTWARE ENGINEERS - A Theoretically Reflective Model

179

Characteristics, Motivation Hygiene and Task

Design. Our model also moderately reflects Adams’

Equity theory, because the model parameters include

the factor of equity as a motivator. We argue that

even though our model fails to directly show a

transition in needs or mix in needs, it identifies a set

of motivators that can be said to reflect the Needs

theories of Maslow and McClelland. The least

reflected theory in our model is the Stimulus

Response theory. Overall, we suggest that this strong

reflection of many of the classic motivation theories

in our model gives the model the theoretical efficacy

that has been wanting in many models of software

engineers’ motivation.

6 CONCLUSIONS

In this paper we have presented a model of software

engineers' motivators. Our previous work suggested

that no rational model of software engineers’

motivation existed. The few that exist make a fairly

disjointed use of classic motivation theories. In this

work, we have conducted a comprehensive review

of studies on software engineers' motivators and

extracted a model of motivation from this review.

Our model shows that software engineers are

motivated by two sets of factors, intrinsic and

extrinsic motivators, where a subset of intrinsic

motivators are aspects inherent in the job that

software engineers do, for example, the problem

solving nature of software engineering. Our model

also shows that factors that orientate software

engineers toward these particular motivators are

their characteristics and that these characteristics are

mediated by individual personality traits and

environmental factors. Our model shows that the

environmental or contextual factors can have a direct

effect on the effectiveness of motivators. Also, that

the balance between organisational intrinsic and

extrinsic motivators and the motivators inherent in

software engineering have an effect on software

engineers’ characteristics, and their reactions to

different motivators. Finally, our model shows that

the external outcomes of software engineers’

motivation are benefits like staff retention, increased

productivity, and reduced absenteeism.

Overall, we suggest that our model serves as a

valuable starting point for managers wanting to

understand how to get the best out of software

engineers, and an individual wanting to understand

his own motivation, or who is embarking on career

choice. We suggest that it also provides a platform

from which subsequent researchers can base their

empirical studies, thereby providing a well-founded

basis on further motivation work.

ACKNOWLEDGEMENTS

This work was supported by the UK's Engineering &

Physical Sciences Research Council under grant

number EP/D057272/1.

REFERENCES

Adams, J. S., 1963. Towards an Understanding of

Inequity, Journal of Abnormal and Social Psychology,

67, 422-436.

Beecham, S., Baddoo, N., Hall, T., Robinson, H. and

Sharp H., 2006. Protocol of a Systematic Literature

Review of Motivation in Software Engineering,

Technical Report No. 452 School of Computer

Science, Faculty of Engineering and Information

Sciences, University of Hertfordshire.

Beecham, S., N. Baddoo, T. Hall, H. Robinson, and H.

Sharp 2007/8, Motivation in Software Engineering: A

Systematic Literature Review. Journal of Information

and Software Technology, Elsevier (accepted

manuscript).

DeMarco T and Lister T., 1987. Peopleware - Productive

Projects And Teams. Dorset House.

Hackman, J. R. and Oldham, G. R., 1976. Motivation

through the design of work: test of a theory. New

York, Academic Press.

Hall, T., S. Beecham, N. Baddoo, H. Robinson, and Sharp

H., 2007, Motivation Theory in Software Engineering

ACM Transactions on Software Engineering and

Methodology (TOSEM), (accepted manuscript).

Kitchenham, B., 2004. Procedures for Performing

Systematic Reviews, Keele University and National

ICT Australia Ltd.: 1 - 28.

Locke, E. A., 1968. Toward A Theory Of Task

Motivation And Incentives, Organisation Behaviour

And Human Performance, 3, pp 157-189.

Maslow A., 1954. Motivation And Personality. Harper &

Row, New York.

McClelland, D. C., 1961. The achieving society.

Princeton: Van Nostrand.

Pidd, M., 1999. Just Modeling Through: A rough Guide

To Modeling, Interfaces, 28(2), 118-132

Sharp, H., Baddoo, N., Beecham, S., Hall, T., and

Robinson, H., 2007. Models of motivation in software

engineering, Information and Software Technology,

(accepted manuscript)

Skinner B. F., 1976. Walden Two. Macmillan, New York.

Standish, 1994. The CHAOS Report, The Standish Group

Tomayko, J. E. and Hazzan, O., 2004 Human Aspects of

Software Engineering, Charles River Media,

Vroom, V. H.,1964., Work And Motivation, Wiley, New

York.

ICEIS 2008 - International Conference on Enterprise Information Systems

180