Parzinger, Examining the Effect of the Transformational Leader on Software Quality

Approaches used by many software development units include: Capability Maturity Model CMM, Total Quality Management TQM, and ISO 9000-3.. Keywords: leadership, organizational change, tot

Examining the Effect of the Transformational

Leader on Software Quality

St Mary’s University, School of Business and Administration, One Camino Santa Maria,

San Antonio, Texas 78228

College of Business Administration, Creighton University, Omaha, Nebraska 68178

The University of Tennessee at Martin, School of Business Administration, S.B.A 131,

Martin, Tennessee 38238

Abstract Developing and maintaining quality software is paramount in the information-intensive

soci-ety A myriad of concepts, tools and techniques exist that can be employed to improve the quality of software and at the same time increase developmental efficiencies Approaches used by many software development units include: Capability Maturity Model (CMM), Total Quality Management (TQM), and ISO 9000-3 Implementation of these approaches without appropriate management oversight does not guarantee success This study examines the role of the manager vis-à-vis “leadership style” with soft-ware quality Data collected using a questionnaire administered to members of the American Society for Quality (ASQ) - Software Division, suggest that the Transformational leadership style of the manager has a significant positive relationship with the quality of the software developed.

Keywords: leadership, organizational change, total quality management, software development, capability maturity model, ISO 9000-3

Introduction

Enhancing the quality of software is a topic of importance to both researchers and practitioners Software quality, often measured by factors such as program errors, customer approval, meeting user needs, reusable code, and easy expandabil-ity, directly affects the resources needed for system development and maintenance Minimizing time and labor resources while producing a quality product is essential for many organizations in today’s highly competitive environment

Often tools and techniques like prototyping, computer-aided software engineer-ing tools, and structured development approaches such as the Systems Development Life Cycle are not sufficient for meeting software development quality goals Soft-ware organizations have turned toward an analysis of the development process in order to improve product quality Some have approached quality through the pro-cess guidelines provided in ISO 9000-3 Another approach for improving quality

in the software development process is the application of the Capability Maturity

Model (CMM) Total Quality Management (TQM) is yet another approach that addresses not only process analysis but organizational culture

Cultural change, in turn, requires highly committed top management leadership (Huge, 1990) However, the manner or style of even committed management can take various forms Management pundits consider “leadership style” of the man-ager a proxy for management effectiveness Leadership style is often categorized

as Transformational or Transactional A Transformational leader is thought to be more effective in implementing organizational change Though change management depends on leadership to be enacted, there has been very little integration of these issues (Eisenback et al., 1999) Empirical research addressing leadership styles and their affect on TQM implementation success has not been explored in the context

of IS development Thus, it is worth considering whether the Transformational lead-ership style of managers leads to higher quality of software than the Transactional style during organizational change such as TQM implementation The purpose of this paper is twofold First, the characteristics of the Transformational and Trans-actional leadership styles are examined in the context of TQM in software devel-opment Secondly, the effect of the Transformational leadership style on software quality is investigated

Literature review

Problems such as poor system quality, long development lead times, user dissatisfac-tion, and high costs of software development are indicative of the need for change

in IS processes (Ravichandran and Rai, 2000) Organizations have turned to TQM concepts for solutions TQM is defined by Merlin and Parkinson (1994) as “the integration of quality management methods, concepts, and beliefs into the culture

of the organization to bring about continuous improvement.” The implementation

of TQM within an organization or individual department can be difficult Because

it involves an intangible philosophy, as well as tools and techniques, a major change

in organizational culture may be required This cultural change has ramifications for the software developer For example, the focus is no longer on immediate users of information systems (IS) but upon customers, both internal and external Control

is not administered by management but shifted to employees and teams Proactive measures are used to discover potential problem areas rather than waiting for a system to fail (Zahedi, 1995)

TQM is a philosophy that, if properly implemented, can enhance the quality of products and services However, some organizations adopting the philosophy have reported success while others have obtained less than desirable results (Eskildson, 1994; Cao et al., 2000) Observers of quality programs typically blame implementa-tion processes for failure rather than the underlying concepts (Reger et al., 1994) Past research has emphasized software quality characteristics, software metrics, and quality control techniques and tools (Ravichandran and Rai, 2000) While the technical and engineering aspects of quality control have been stressed, research gaps remain in the organizational aspects of quality management Just recently, attention has turned to organizational factors that impact TQM implementation in

IS development For example, factors deemed critical for successful TQM imple-mentation in the IS arena have been identified (Parzinger and Nath, 1998; Ravichandran and Rai, 2000), and the influences these factors have had on software quality have also been examined (Parzinger and Nath, 2000) Yet much of the cur-rent TQM literature surrounding systems development is anecdotal and prescriptive However, a common thread in all TQM studies is the importance of management’s role in the implementation of a quality program

The lack of success of TQM implementation is often attributed to management’s disregard for the human aspects of implementation (Harris and Purdy, 1998) The attributes of managers are changing In an organization where the manager con-trols processes, not people, attributes required are team-working, high interper-sonal skills, negotiating, consensus management and the ability to handle several issues or projects at a time The leader must develop competence and trust in the employees to make their own decisions (Macdonald, 1998) However, in many IS organizations, inherent distrust presents a barrier to successful employee empower-ment Many managers make all the decisions and blame employees when problems arise (Pearson et al., 1997; Ward, 1997) Leadership style, while not recognized as

a major factor of TQM, may have a significant influence on the contributions of TQM tools and techniques (Bass et al., 1987b; Howell and Avolio, 1993; Niehoff

et al., 1990; Waldman, 1994)

Two forms of leadership often discussed in the literature are Transformational and Transactional The Transformational leader possesses charisma and is able to simply be supportive rather than directive when the situation allows (Hersey and Blanchard, 1969; Hersey and Blanchard, 1977; Hersey and Blanchard, 1984) There

is a religious-like motivation that the leader is able to instill in the employees, whereas, the Transactional leader appeals to employees’ self-interest rather than raise the levels of morality and motivation (Burns, 1978) Transformational leader-ship processes may align followers’ work-oriented values with those of the greater group or organization (Bass et al., 1987a; Burns, 1978; Conger and Kanungo, 1988) The empathetic, facilitative characteristics of the Transformational leader may be necessary for persistent efforts such as required of continuous process improvement (Waldman, 1994)

Leadership studies often focus on a top ranking corporate officer However, Francis J Yammarino notes, “Transformational leaders need not occupy the high-est or most prominent positions to influence others Transformational leaders can occupy a variety of positions at various levels of organizations and be formal or informal leaders.” Leadership qualities can trickle down through layers of the hier-archy In fact, the relationship between the leadership style of the person in charge

of the quality effort and the operating employees may be irrelevant It may be the immediate supervisor’s leadership behavior that influences success rather than the person in charge of the project (Bass and Avolio, 1994)

Leadership style, though not considered a factor of TQM, may have a significant relationship with the success of the quality program In the realm of software devel-opment, the success of the TQM program can be measured in terms of customer satisfaction, the Capability Maturity Model level, and the extent of compliance with

ISO 9000-3 This research examines the following propositions:

Proposition 1 The leadership style utilized during TQM implementation in the

soft-ware development arena will affect the quality of softsoft-ware produced.

Proposition 2 The degree of Transformational characteristics displayed by the leader

will directly affect the level of software quality during TQM implementation.

Measures of software quality

In order to measure the success of TQM in the software development arena, it

is necessary to define software quality Reeves and Bednar (1994) conclude that

a global definition of quality does not exist, but that different definitions of qual-ity are appropriate under different circumstances In measuring software qualqual-ity, specific characteristics of a system are typically addressed These include flexibility, maintainability, reusability, integration, consistency, usability, reliability, salability, functionality, efficiency, and portability (Dunn, 1990; Arthur, 1993; Yourdon, 1992; Humphrey, 1989; Sanders and Curran, 1994; Budgen, 1994) These characteristics tend to focus on the engineering aspects of software development that ultimately affect user (customer) satisfaction Software quality is multi-dimensional and exist-ing attempts at measurement have many weaknesses (Budgen, 1994) Recently, a threefold approach focusing on the user, the technologies, and the purpose of the software has been suggested for performance measures (Prahalad and Krishnan 1999) In light of these facts, this study employs multiple methods to measure soft-ware quality

Customer satisfaction

The relationship between users and IS personnel is typically viewed as pivotal to the success of systems development projects (Beath and Orlikowski, 1994) It is often the views of customers that dictate the degree of quality in the product However, obtaining a potential customer’s software requirements in the initial stages of devel-opment is often an obstacle affecting the quality of the final product Difficulties in communication with customers can result in late deliveries, budget overruns, and undiscovered defects To heighten the problem, software development personnel do not always recognize its key customers who deserve the most attention (Merlyn and Parkinson, 1994) With TQM’s emphasis on customer satisfaction, it is reasonable

to acquire the customer’s perspective through questionnaires or interviews

Capability Maturity Model

The Capability Maturity Model (CMM) for Software is founded upon works of Humphrey (1988, 1989) The first release was in two technical reports of 1991 but these were revised during 1992 and are entitled “Capability Maturity Model for

Software, Version 1.1” (Paulk et al., 1993a) and “Key Practices of the Capabil-ity MaturCapabil-ity Model, Version 1.1” (Paulk et al., 1993b) The underlying premise of the model is that continuous improvement can occur only through focused and sustained effort towards building a process infrastructure of effective software engi-neering and management practices This is essentially seen in a mature software organization as compared to an immature organization that generally improvises during the course of a project In a mature organization, managers monitor the quality of software products and customer satisfaction The CMM can be used to understand the key practices that are part of effective processes for developing and maintaining software and to identify those key practices needed to achieve the next maturity level in the CMM There are five levels of software process maturity: initial, repeatable, defined, managed, and optimizing Each is described below:

• Initial—The software process is characterized as ad hoc, and occasionally even chaotic Few processes are defined, and success depends on individual effort

• Repeatable—Basic project management processes are established to track costs, schedules, and functionality The necessary process discipline is in place to repeat earlier successes on projects with similar applications

• Defined—The software process for both management and engineering activities is documented, standardized, and integrated into a standard software process for the organization All projects use an approved, tailored version of the organization’s standard software process for developing and maintaining software

• Managed—Detailed measures of the software process and product quality are collected Both the software process and products are quantitatively understood and controlled

• Optimizing—Continuous process improvement is enabled by quantitative feed-back from the process and from piloting innovative ideas and technologies (Paulk

et al., 1993a)

To assess the software development process, a team is selected which administers and evaluates the maturity questionnaire This questionnaire was designed by the SEI and addresses key practices grouped into (1) commitment to perform, (2) ability

to perform, (3) activities performed, (4) measurement and analysis, and (5) verifying implementation The team also visits the sites being assessed and produces a list of strengths and weaknesses The final step involves the preparation of a key process area profile This profile highlights the areas where the organization has, or has not, satisfied goals determined in prior steps

Prior studies reflect the relationship between the various CMM levels and soft-ware quality (Krishnan, 1996; Harter et al., 2000; Anthes and Vihayan, 2001) For example, statistical analysis of data from a large Fortune 100 software develop-ment laboratory found that process maturity significantly increased quality (Krish-nan, 1996) There is also reason to believe that TQM can influence the CMM level Hollenbach, et al (1997) depicts how TQM approaches allowed a business unit to move from Level 1 to Level 3 on the CMM

ISO 9000-3

The International Organization for Standardization (ISO) has developed a series of standards for TQM (Tenner and DeToro, 1992) ISO 9000-3 (1991) is the guideline for the application of ISO 9001 to the development, supply, and maintenance of

software (ISO 9001 is the Model for Quality Assurance in Design, Development,

Production, Installation and Servicing) Because the process of development and

maintenance of software is different from that of most other types of industrial products, ISO considered it necessary to provide additional guidance for quality sys-tems via ISO 9000-3 The guidelines are intended to describe controls and methods for producing software that meet a purchaser’s requirements Many organizations are finding it necessary to be in compliance with ISO 9000 standards in order to con-duct business with potential customers ISO 9000-3: 1991 (E) provides a framework describing responsibility, documentation, audits, and corrective action for software development in order to meet the requirements of ISO 9000

Empirical evidence of the relationship between ISO 9000-3 compliance and soft-ware quality is not prevalent Registration for this certification does not guarantee that a product is certified for quality Certification does give a foundation for achiev-ing quality (Ashrafi et al., 1995) There is, however, a strong correlation between ISO 9001 and the CMM although the level of abstraction is different Both the ISO quality guidelines and the CMM emphasize documented procedures (Paulk, 1995)

Methodology

Data for this research were collected using a questionnaire The results reported in this paper are based on a subset of the data gathered To measure the dominating leadership, Bass’s (1985) instrument was adapted Bass’s (1985) research resulted

in five dimensions or factors of leadership Charisma, individualized consideration, and intellectual consideration are three active dimensions of the Transformational leader Contingent reward is one active dimension of the Transactional leader and management by exception one passive dimension of the Transactional leader From each of these five dimensions, two items with the highest loadings were chosen and included in the instrument This resulted in the following 10 items:

TRANSFORMATIONAL LEADERSHIPSTYLE

• Makes everyone around him/her enthusiastic about assignments

• I have complete faith in him/her

• Enables me to think about old problems in new ways

• Gives personal attention to members who seem neglected

• Finds out what I want and tries to help me get it

• His/her ideas have forced me to rethink some of my own ideas which I had never questioned before

TRANSACTIONAL LEADERSHIPSTYLE

• Tells me what to do if I want to be rewarded for my efforts

• There is a close agreement between what I am expected to put into the group effort and what I can get out of it

• As long as the old ways work, he/she is satisfied with my performance

• He/she is content to let me continue doing my job in the same way as always

Each item is rated in a 5-point scale (1 = Not at All, 2 = Once in a While, 3 = Sometimes, 4 = Fairly Often, 5 = Frequently, If Not Always) While a leader may have some characteristics of either style, a dominating mode usually surfaces The mean of the 6 items indicative of Transformational leadership style was compared with the mean of the 4 items describing Transactional leadership style Data was acquired by requesting the respondent to judge how frequently the respondent’s current immediate supervisor or the supervisor during the Quality program imple-mentation had displayed behavior indicative of transactional and transformational styles of leadership Thus, responses were based on the manager’s style during the implementation phase This eliminated the probability that the respondent was basing the answers on a new supervisor rather than the person in charge dur-ing the implementation The respondent’s manager was assigned to the leader-ship style with the higher mean If the two means were identical, no style was assigned

In addition, respondents were asked to rate the success of their TQM effort in software development using a 5-point Likert scale (1 = very unsuccessful; 5 = very successful) Overall program success and three other measures of software qual-ity were used in this study: Change in customer satisfaction, CMM level, and the Extent of ISO 9000-3 compliance Change in customer satisfaction used a 5-point scale (1 = significant decrease; 5 = significant increase) The Extent of Compliance with ISO 9000-3 also utilized a 5-point scale (1 = not implemented; 5 = complete compliance) Also, each respondent was requested to identify where his/her orga-nization stood with respect to the five stages described in the Capability Maturity Model

The initial draft of the instrument was presented to a panel of MIS doctoral stu-dents and faculty members Modifications were made based upon their discussion and recommendations The revised questionnaire was then distributed to 20 individ-uals, primarily practitioners, for pilot testing This group was chosen from volunteers who subscribe to the QUALITY listserv on the Internet, a bulletin board service provided through Princeton University for the purpose of discussing any facet of quality A hard copy of the questionnaire was mailed to each volunteer along with

a self-addressed stamped envelope The 20 returned responses were evaluated and comments reviewed This feedback led to further clarifications of some items Mem-bers of the American Society for Quality (ASQ)—Software Division, were used to collect data The questionnaire was mailed to 1,990 ASQ members A postage-paid envelope and a cover letter explaining the purpose of the study were included with the questionnaire

Of the 1,990 questionnaires mailed, 247 completed questionnaires were returned, resulting in a response rate of 12% Of these 247 questionnaires, 145 respondents indicated implementation of a formal quality program and were used in the analysis This section reports the results of the study

Profile of the respondents

Table 1 reports the profile of the responding organizations One hundred forty-five (59%) reported having a formal quality program in place; 14% were considering it; 25% did not have one, and 2% did not report this information Of those having

a formal quality program, nearly 83% had implemented it throughout the organi-zation and the rest had incorporated it only in the software development process Nearly 3

4 of the organizations with TQM implementation were manufacturing, soft-ware development, or service Each represented about 25% of the total Only those organizations indicating a significant commitment to Total Quality Management or

a similar Quality program were included in the analysis

To determine the average Transactional leadership score, the 4 items indicative

of this style were averaged Due to the option of a ‘not applicable’ response to an item, it is possible for the average to be less than 1 The mean score is 2.6 with

a standard deviation of 0.74 See Figure 1 for the Frequency Distribution of these scores

The same procedure was performed to determine the average Transformational leadership score The 6 items reflecting the Transformational style were averaged and the Frequency Distribution of the results is displayed in Figure 2 The mean score for Transformational behavior is 3.1 with a standard deviation of 1.00

Table 1 Profile of respondents and organizations

Existence of a formal

∗ Percentage may not add to 100 because of missing data.

5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 5

50

40

30

20

10

0

Std De v = 74

Me an = 2.6

N = 143.00

Figure 1 Frequency Distribution—average Transactional Score.

5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0

30

20

10

0

Std Dev = 1.00 Mean = 3.1

N = 143.00

Figure 2 Frequency Distribution—average Transformational Score.

Typically, individuals can display characteristics of both leadership styles How-ever, the frequency of behaviors indicates a dominant style reflected in subordi-nates’ perceptions Table 2 shows the dominant leadership style of the managers

as determined by comparing the means of the two dimensions: Transformational and Transactional The mean of the 6 items indicative of Transformational

lead-Table 2 Dominant leadership style

ership style was compared with the mean of the 4 items describing Transactional leadership style A manager is assigned to the category with the higher score Out

of the 145 respondents, 65% reported their manager having dominant traits of a Transformational leader and 31% reported their manager showing dominant Trans-actional leadership characteristics In 3% of the cases, the means were identical and

no dominant leadership style could be determined

To determine whether organizations with Transformational leadership had sta-tistically significantly higher mean scores than those for organizations with

Trans-actional leadership, two-sample t-tests were performed Table 3 shows the mean, standard deviation and the results of the t-test Note that at the 01 level,

signif-icant differences were found for all measures except Extent of Compliance with ISO 9000-3 That is, organizations with a Transformational leader exhibited a sig-nificantly higher degree of success compared to organizations with a Transactional leader

Table 3 Descriptive measures of success by leadership style

Transformational Transactional

Change in level of customer satisfaction 351 069 302 070 223 009

Capability maturity model level 225 090 174 088 234 005

Extent of compliance with ISO 9000-3 218 152 226 138 123 224

∗ Scale of 1 to 5.

Correlations between the level of Transactional leadership characteristics and measures of quality in software development were insignificant However, the level

of Transformational leadership characteristics was found to be significantly corre-lated with Overall Success of the TQM program, Level of Customer Satisfaction,

and CMM level at alpha = 01 In essence, the more frequent the leader behaved in

a manor characteristic of the Transformational style, the greater the level of success (see Table 4)

Table 4 Transformational leadership style and success correlations

Change in level of customer satisfaction 278 344

Note: Numbers in parentheses represent the level of significance.