Leadership in Information Systems A Literature-analysis Study

(1)

1

Leadership in Information Systems

A Literature-analysis Study

By

LAURENS SMIT

University of Groningen

Faculty of Economics and Business

Master Thesis Business Administration: Business & ICT

Azielaan 236 3626SH Utrecht Tel +31 (0) 643105583 E-mail: Laurens.f.smit@gmail.com

(2)

2 Abstract:

This research paper focuses on leadership and IT literature. The Literature analysis has identified 72 relevant papers from 1990 till 2012. From the relevant papers, the research model that was used is decomposed into variables that were composed into this papers' research model. This research identified the leadership variables that were used in the relevant literature. The results of these relevant articles are discussed in two parts, the first parts look's into the search results, This research found that many papers were discussing the topics of project management, implementation projects and more specifically in virtual team settings and Group decision support systems. The larger second part of this study investigates the leadership attributes and characteristics of leadership, the main contribution of this study is there is a comparison between traditional leadership and IT leadership. This study found that many attributes and characteristics of IT leadership fall into

(3)

3

Introduction

Much research has been done on traditional leadership from various perspectives; traits leadership research dates back to 1849 where Thomas Carlyle's "great man" theory which stated that the history of the world was the biography of a great man (Carlyle, 1849; Zaccaro, 2007). However after the early 1950's many researchers discarded the trait-based leadership approaches as being insufficient to explain leadership and leadership effectiveness (Stogdill, 1948; Mann, 1959). The trait-leadership perspective was abandoned for its inability to offer a clear distinction between a leader and non-leaders, also scholars commented that any trait's effect on leadership behavior will always depend on the situation (Huges, Ginnett, & Curphy, 1996; Yukl & Van Fleet, 1992). A more recent study, driven by greater conceptual, methodological, and statistical sophistication, the trait-leadership approaches have again risen to prominence (Zaccaro, 2007)

A second leader-central approach is the skill based leadership approach (Goleman, 1998; Drennan and Richey, 2012). Skill based leadership is in a manner similar to the trait approach, they differ in the sense that the trait approach focuses on personality traits that are considered inherent where the leadership skill approach can be learned and developed (Northhouse, 2010). Goleman from HBR argues it's not IQ or technical skills, but rather emotional intelligence: a group of five skills that enable the best leaders to maximize their own and their followers' performance. The five skills he proposes are self-awareness, self-regulation, motivation, empathy, and social skill.

Though many writers in the field of leadership and management have been involved in search for the best style of leadership which would be successful in most situations the evidence clearly indicates that there is no single all-purpose leadership style (Hersey and Blanchard, 1976). Hersey and Blanchard argue that successful leaders are those who can adapt their behavior to meet the demands of their own unique environment (Hersey & Blanchard, 1976). This is where Hersey and Blanchard in the early 70's created their situational leadership model theory. It is based upon an interplay among the number of directions a leader gives, the amount of socio-emotional support a leader provides and the maturity level that followers exhibit on a specific task (Hersey & Blanchard, 1976).

(6)

6 broad conceptual map that is worthwhile to use in attempts to understand the complexities of leadership. Based on this approach leaders can assess their actions and determine how they want to change their style to develop their leadership. Though one of the criticisms is that – again – this approach has failed to find a universal style of leadership that could be effective in almost every situation (Northouse, 2012).

Leadership and management for one way look similar to each other, both involve influence, and working with people, both are concerned with effective goal accomplishment. However, leadership is different from management, but not for the reasons most people think. Leadership is not mystical and mysterious (Kotter, 2001), of course, not everyone can be good at both leading and managing. Some people have the capacity to become great managers but not strong leaders. Others have great leadership potential but, for a variety of reasons, have great difficulty becoming strong managers (Kotter, 2001). Leadership can be traced back to Aristotle where management emerged around the turn of the 20th century with the advent of the industrialized society (Northhouse, 2012). Management is about coping with complexity such as planning, organizing, staffing and controlling. Where leadership, by contrast, is about coping with change and setting vision and influencing others. Napoleon allegedly said that no successful battle ever followed its plan. Yet Napoleon also planned every one of his battles, far more meticulously than any previous general had done. Without an action plan, the executive becomes a prisoner of events (Drucker, 2004 ).

Leadership is still one of the most intriguing subjects in research, and it looks like modern literature knows much about leadership from different perspectives as earlier described. However with the emergence of new technology that is becoming more valued, IT is an important facet of the organization. This research looks into the value of IT leadership in IT projects. As there is so much literature about general leadership then IT departments and organizations dealing with IT can still have much to learn or adopt practices from general leadership, also new technologies lead to new communication channels which may require new leadership styles.

(7)

7 IT projects are accomplished by teams, where the team is a group of individuals who are responsible for completing the project work. The project team is assigned to the project manager or leader, however, the variety of views in the project team often leads to conflict in task and

personalities (Chan, 1989).

User participation has widely been acknowledged by literature to be of utmost importance in IT/IS projects and has become a recommended strategy for development and implementation (Wagner & Newell, 2007; Newman & Noble, 1990; Wakefield, Leidner, Garrison, 2008; Jiang, Klein, Discenza, 2002). Evenly important as user participation and commitment is the commitment of top-management support (Jiang, Klein, Discenza, 2002; Neufeld, Dong, Higgins, 2007) where Armstrong and Sambamurthy (1999) also argue that participation and commitment from CIO can enhance the IT knowledge to help with IT assimilation. Thus creating a paradox, of user participation and

commitment, and conflict management. This is where Pre-project partnering research has indicated that the pre-project activities such as, workshops, icebreaking activities mini-lectures on the

principles of teamwork, synergy, approaches to continuous improvement, constructive feedback and problem solving, may lead to more effective teams and managers and eventually to project success (Jiang, Klein, Discenza, 2002).

Although the IS literature has emphasized the importance of commitment to IS projects as a key factor in their success (Elam & Sabherwal, 1995; Ginzberg, 1981), Strong commitment is not warranted in projects that are failing (Keil & Robey, 1999). Clearly IS management must strike an appropriate balance between obtaining commitment and resources for promising IS projects and withdrawing commitment or redirecting projects that are failing.

(8)

8

Research Method

Coding articles process

This research consists out of two sections, first there is a journal and article selection which focuses on the research method of the relevant articles, second is the in depth coding of the articles that will focus on the conceptual model which are used in the relevant articles.

The first step in the journal and article selection was to identify what journals should be reviewed and based on which criteria. The results need to be refined based upon the topic of the article, and if the criteria was part of the main topic of the article, this research is interested in the finds of an article, so whenever the criteria was found in the literature section it could be marked as irrelevant. This step narrowed the search results. The third step is to code the articles based upon the research method which was used in the articles such as research method, research design, and a topic. These results are processed and summarized in this article.

Figure 1. Process of coding the search results.

The second part of this research focuses on the conceptual model of each relevant article, the first step is to build a reference research model based on traditional leadership theory models in which each conceptual model of the relevant articles could fit. Second step is to identify each conceptual model from the relevant articles, and code these according to the earlier created research model. Each attribute from the conceptual model is summarized and included in the findings. The last part of this study is to discuss the findings from the conceptual model of each relevant article.

Figure 2. Process of coding the articles.

Journal and article selection Data collection Coding the articles

Analyzing and summarizing search

results

Building a research model

Identifying the conceptual model

of each paper

Codinging the conceptual models from the

(9)

9

Journal and article selection

The research method for this paper is a literature analysis from the "AIS basket of eight" IT journals; European Journal of Information systems, Journal of information technology, Journal of Strategic Information Systems, Information systems Journal, Information systems research, Journal of the Association of Information systems, Journal of management information systems, MIS quarterly. The search query for these journals is set on "leader" or "leadership" as text so that articles containing one of those words would show up in the search results. For the date, the setting was between January 1990 till June, and August 2012. EBSCO host was used as primary search engine; however, EBSCO host did not cover the spanning between 1990 and 2012 for all the journals, to overcome this issue, the primary publisher search engine was used for the additional papers. One issue that arose was that some articles were found by search engine one as well as search engine two or three. The search results in the table below (table 1) show only the unique search results. In the appendix, there is a complete list of the search queries.

Table 1. Search results

Journal Unique Search

results

Search engine

European Journal of Information systems 36 + 219 + 31

248 EBSCO Host, Palgrave Macmillan, ProQuest

Journal of information technology 51 + 144 194 EBSCO Host, Palgrave Macmillan

Journal of Strategic Information Systems 266 264 Elsevier Science Direct

Information systems Journal 71 + 73 + 28 172 Wiley, EBSCO Host

Information systems research 135 + 36 + 39 210 Informs Online, EBSCO Host

Journal of the Association of Information systems 43 + 96 + 27

166 AISEL, EBSCO Host

Journal of management information systems 260 + 97

357 ESCHO Host

MIS quarterly 261 + 111 372 ESCHO Host

Total 1983

Data Collection

In total, the search queries resulted in 1983 articles containing the word ‘leader’ or

(10)

10 author sections are marked as irrelevant. With these criteria, 259 articles are marked as possible relevant articles for this study. A second round of thorough reading showed that only 72 articles are relevant due that many articles use leadership as a given, not researching any of the fore mentioned topics or simply stating that leadership is an important factor, these articles are all marked as not relevant.

Coding the articles

Coding the articles can be divided into three steps according to Corbin and Straus (2008). First there is the "open coding" which is breaking the data apart and delineating concepts to stand for blocks of data (Corbin & Straus, 2008). This research used the coding scheme of Zhang and Li (2005) and Crowston et al. (2012) as starting point. As they also made a literature analysis research and their topic is also in the field of IT.

Details of the relevant literature such as, the research models that are used in the articles are identified and decomposed into independent leadership variables, mediators, moderators and the ultimate effects which are known as the dependent variables, more about the research model can be found in the research model section of this paper.

The third step is that the remaining articles need to be coded. The first and highest distinction is made between an empirical study or non-empirical study, which can be divided into more sub-categories as seen below in table 2.

Table 2. Research methods, Research methods in depth (based upon the work of Crowston et al 2013)

Research method

Research Method in depth Description

Theoretical Theory A well-substantiated explanation of some aspect of the natural world; an organized system of accepted knowledge that applies in a variety of circumstances to explain a specific set of phenomena

Opinion A belief or conclusion held with confidence but not substantiated by positive knowledge or proof Literature The body of written work produced by scholars or

researchers in a given field

Empirical Interpretive case study Interpretive study that generally attempt to understand phenomena through the meanings that people assign to them.

Positive case study Positivist study that generally attempt to test theory, in an attempt to increase the predictive understanding of a phenomenon.

Survey A questionnaire

Interview Conversation with or questioning of a person Delphi Process essentially provides an interactive

(11)

11

'experts' in a field, in order to develop themes, needs, directions or predictions about a topic

Focus Questions are asked in an interactive group setting where participants are free to talk with other group members

Second data Data collected by someone other than the researcher him self

Experimental Study in which all of the risk factors are under the direct control of the investigator.

Explorative Type of research conducted for a problem that has not been clearly defined

Next step is the research design of the articles; this can either be qualitative or quantitative. Qualitative research aims to gather an in-depth understanding of human behavior and the reasons that govern such behavior. The qualitative method investigates the why and how of decision making, not just what, where, when. Quantitative research refers to the systematic empirical investigation of social phenomena via statistical, mathematical or computational techniques. The objective of quantitative research is to develop and employ mathematical models, theories and/or hypotheses pertaining to phenomena.

The next level of the coding scheme is the level of analysis that the research paper used, the level of analysis is the level at which the data is collected and analyzed. This paper used four different levels the first two levels; individual, group/team are more focused on the human side where the latter two levels are more focused on organizational and industry units. According to Zhang (2005) the importance to identify at which level the analysis has taken place is that at the individual level can be those relevant to cognitive styles, individual reactions toward IT, and individual productivity or performance related to IT. Group performance in decision making and group member

conflict/agreement are typical topics for analysis at the group level (Zhang, 2005).

Table 3. Level of analysis (based upon the work of Crowston et al 2013)

Level of Analysis

Level Description

Individual Based upon individual or more people from the same function in different organizations

Group/Team A group of people working towards a same goal

Organizational Organizational wide analysis from different functions who do not necessarily have to collaborate with each other

Inter-organizational Same as above only then researched at more than one organization

(12)

12 where banking, commerce setting are found has been merged with organizational or workplace context. As many research papers have an experimental setting that takes place at an educational context, this study believes that this may be a more important factor that can explain certain situations better, as students may not have any working experience or may have a higher understanding than people with less education working in the market place setting.

Table 4. Context of analysis (based upon the work of Crowston et al 2013)

Context of analysis

Context Description

Organizational /Workplace Context of the study is organizational wide investigating a phenomenon which includes the whole organization. Cultural, national,

geographical

Attitudes and behavior characteristic of a particular social group, nation or geographical setting.

Educational Research has taken place in an educational context with ex-graduate, graduate students or professors.

Online Community Context of the study is in an online community e.g. Forums, Massive Multiplayer Online Games.

Leadership research is done on many different topics in information systems literature, next to the topics aforementioned such as market, industry leadership, etc. The relevant articles can also be divided into different topics. This is where Whitten et al. (2004) proposed that the IT artifact lifecycle should be divided into two main stages: during IT development and after IT development (Whitten et al., 2004). However, this remains a broad area, after reading the relevant articles a list could be made with abstract topics, where each paper could fit in. Together with the categories from Whitten et al. (2004) this research has made nine different topics where a paper must fall into.

Table 5. Topics

Topic Description

Career Development Personal development in respect to their career that the person is willing to achieve

Development Development of Information systems or IT infrastructure. Development & implementation As above only the paper also included the implementation

phase

Group Decision Support Systems Class of electronic meeting systems, a collaboration technology designed to support meetings and group work Group Support Systems Information systems that aim to make group meetings more

productive by offering electronic support for a variety of meeting activities.

(13)

13

IT innovation An innovation is something original, new, and important—in whatever field—that breaks in to (or obtains a foothold in) a market or society

IT leadership Paper that has observed the leadership phenomena that occurred in the organization.

IT use and impact The use and impact a IT change had on the organization Knowledge Management Process of capturing, developing, sharing, and effectively

using organizational knowledge.

Project Management The process and activity of planning, organizing, motivating, and controlling resources to achieve specific goals.

Strategic decision making Paper that is describing the phenomena of interaction in top management about long term planning of the organization Virtual teams Paper which is describing a group of individuals who work

across time, space and organizational boundaries with links strengthened by webs of communication technology. IT De-escalation Paper that is describing how IT project that were escalated or

were about to escalate are managed in order to prevent escalation.

Research model

This research tries to identify characteristics from leaders in IT projects, these characteristics can be extrinsic or intrinsic. Also, group leadership is researched in terms of how do groups reach an agreement, is this trough elected leadership, emerged leadership or no leadership at all? In order to research these characteristics, the articles are analyzed to create a conceptual model.

The first independent variable (IV) is the "Leadership" variable, each relationship has at least one independent variable and one dependent variable (DV). It is normally hypothesized that, in some way the IV causes DV (Blumberg, 2008). This "leadership" variable only lists leadership characteristics that were analyzed in the article. The other IV, the "mediator" variable serves to clarify the nature of the relationship between the IV and DV In other words, mediating relationships occur when a third variable plays an important role in governing the relationship between the other two variables (MacKinnon, 2008). The "moderating" variable is a second IV that is included because it is believed to have a significant contributory or contingent effect on the original IV-DV relationship (Blumberg, 2008). Sometimes it is difficult to distinguish mediators and moderators when forming hypotheses about variables. The definitional difference that a mediator is predicted by the independent variable and a moderator is a separate independent variable, is important but not always obvious (Bennett, 2000).

(14)

14 known as the dependent variable is the outcome of the research model which is mentioned in other studies as observed, measured or explained variable. The figure below (figure 3) is used as a base research model to map and identify elements of the relevant articles.

Figure 3. Research model (based upon the work of Blumberg, 2008)

Findings from the search results

This study has found 72 relevant articles, in this table each journal is listed with the results from the search query and categorized by year. As the table shows, only the Journal of Management Information systems and MIS Quarterly published digital or digitalized papers in the topic of

leadership and IT. Information Systems Research also published two papers in 1990, though, why there is a gap between 91 and 99 needs investigation. An assumption that could be made is that since 99's other journals started to digitalize their journals and that technology made it possible to do so.

Table 6. Relevant articles by year and Journal

Journal 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 Total EJIS 1 2 1 2 1 2 2 1 12 JIT JSIS 1 1 2 1 1 1 7 ISJ 1 2 1 3 7 ISR 2 1 1 1 1 6 JAIS 1 1 1 2 1 1 7 JMIS 2 1 1 2 1 3 2 1 1 1 1 2 18 MISQ 2 1 1 1 2 1 1 2 1 2 1 15 Total 2 2 3 1 1 1 2 0 6 3 6 5 3 2 2 0 4 6 8 6 7 2 72 Leadership Attributes

Mediator(s) Ultimate Effect(s)

Moderator(s) Other factor(s) That affect the mediator(s)

(15)

15

Research methods per year

This study identifies the research method which the relevant articles used in two ways, the first step is to identify if the study was either qualitative or quantities, the results show that there is an even distribution of qualitative or quantitative studies. The second table shows what research method was used in what year, the results show that surveys and case studies were found most in the relevant articles with 39% and 27% also experiments and interviews were found relatively much compared to Delphi and focus studies which were only found once.

Table 7. Relevant articles by Research Method by Year

Table 8. Relevant articles by Research Method in depth by Year

The table nine shows which level of analysis was used in what year, as the results show there is an even distribution between the level of analysis, where organizational and inter-organizational scores a little higher. Also, there is a peak in from 99's till 2002 and from 2007 till 2011, however, no conclusions can be drawn from these statistics.

Table 9. Relevant articles by Level of analysis and Year

(16)

16

Research method vs. Level of analysis

The second table identifies which research method was used for which level analysis, first for organizational level of analysis the case study was preferred as most chosen research method, this is explained that the studies focus on a project that is the case for that particular study. Second is that when studies are carried out in inter-organizational wide context, the survey design was used most often compared to the individual score. The inter-organizational research was interested in various people from more organizations, as the high individual score is explained that only one certain role e.g. CIO was researched within multiple organizations. For group or team level of analysis the experimental method was used most, this can be explained that many experiments were in a group setting in an educational context, e.g. observing the group behavior to a certain leadership style, or how leadership emerges.

Table 10. Relevant articles, Research method vs. Level of analysis (based upon the work of Crowston et al 2013)

Research Method Level of Analysis

Total Individual Group/Team Organizational Inter-org.

(17)

17

Context vs. Level of analysis

Next this study looks at the context of the study compared to the level of analysis. The results show that there is a clear preference for an organizational context with 75% of the research. Studies in an educational setting are responsible for 12% that is explained by the experimental settings in an educational setting, or research done at universities. The online community score can be explained by the technology that was not (easily) available till 2002 as is explained in the next table "topic trends".

Table 11. Relevant articles, Context vs. Level of analysis

Context Level of Analysis

Total Individual Group/Team Organizational Inter-org.

Total 72 13 15 20 24 Organizational ₅₆ 11 7 18 21 Workplace 2 2 Educational ₉ ₁ ₅ ₂ Online community ₅ ₁ ₃ ₁ Topic trends

The final table shows which found topics where popular and if there were any trends in a particular year. The most found topic is about implementation, as IT is confronted with many changes that in many cases is driven by technological advancements that explain why this topic is popular for research due that these projects are often driven by project leaders and how

organizational leaders support these changes. Also, the development topic e.g. software, is often found in a project based group, which may as well be organizational, inter-organizational or even in a global virtual context, these projects were also favored by researchers. The virtual team topic

emerged around 2002 which can be explained by technology that made it possible to collaborate easier and faster.

Table 12. Relevant articles, Topic trends

(18)

18

Findings from the Leadership Literature

Findings are discussed based upon the level of analysis used in the research articles relevant for this study. Modeling all attributes that were found in the relevant articles was first done in one large model, the result was a too large model where all attributes were merged, the results did not make any sense. The idea was to find a certain level or context in which the articles from that same context or level could be merged and say possibly something about each other in order to find any relations between them, from which a theory could be build.

First I tried to model it according to Organizational/Workplace,

Cultural/national/geographical, Educational or Online Community context, which didn't make sense as the level of analysis was too different en thus describing or comparing different attributes at different levels did not make sense. Also modeling according to topic resolved in the same problem. Modeling by level of analysis worked quite well, tough the hard part was that the moderators and mediators are in many cases so different – as there can be numerous of different mediators and moderators – that it is difficult to merge them at first hand, therefore, this study has created an in-depth reference model which is an elaborated version from the one in figure 1, based upon the work of Crowston et al. (2013). The leadership attributes are from earlier leadership studies; trait

(19)

19 leadership attributes are found in the studies from Stogdill (1948), Mann 1959), Stogdill (1979), Lord (1986), Locke (1991) and Zaccaro (2004).

Table 13. Leadership traits and characteristics (Adapted from Northouse, 2012)

Skill leadership attributes are found in other literature as well; the work from Northouse (2012) gives a good overview of the different leadership views, one of them is the skill leadership view from Katz (1955). Katz suggested that an effective administration (i.e., leadership) depends on three basic personal skills: technical, human, and conceptual. A more recent work from Mumford, Zaccaro, Harding, et al., (2000) formulated a new skill based model of leadership characterized as a capability model because it examines the relation between a leader's knowledge and skills and the leader's performance (Mumford, Zaccaro, Harding, et al., 2000). In this research attributes from both studies are used to code the articles, from the Mumford article the three competencies, problem-solving skills, social judgment skills and knowledge are used, the social judgment skills are similar to Katz's his human skills; however, the conceptual skills from Katz's study cannot be found in the work of Mumford, which are central to creating a vision, strategic planning and to work with ideas. These attributes can be found in the Mumford model which I believe are keen for a complete skill overview.

Trait Leadership Attributes

intelligence, Stogdill (1948), lord (1986) dominance, Mann (1959), Lord (1986) Alertness, Stogdil (1948) extraversion, Mann (1959) Zaccaro (2004) insight, Stogdil (1948) conservatism, Mann (1959)

responsibility, Stogdil (1948), Stogdill (1974) achievement, Stogdill (1974) initiative, Stogdil (1948) cooperativeness, Stogdill (1974) persistence, Stogdil (1948), Stogdill (1974) tolerance, Stogdill (1974) self-confidence, Stogdil (1948) Stogdill (1974) influence, Stogdill (1974) sociability, Stogdil (1948) Stogdill (1974) drive, Locke (1991)

masculinity, Mann (1959), Lord (1986) motivation, Locke (1991), Zaccaro (2004) adjustment, Mann (1959) interpret, Locke (1991)

(20)

20 Table 14. Skill leadership Attributes (Adapted from Northouse, 2012)

French and Raven (1959) identified five different bases of power: reward, coercive, legitimate, referent, and expert. This initial categorization of power was later revised to two broad types: structural or legitimate power (i.e., having the right or authority to exert power), and personal or expert power (i.e., having more information regarding a certain area important to the group) (Sarker & Valacich, 2010). Therefore, this study uses the five bases of power to categorize attributes related to power.

The mediators and moderators found in the relevant articles are so disperse as there can be various factors that mediate or moderate a particular leadership attribute affecting the ultimate outcome that many mediator and moderator variables come from the identified research articles themselves. This study tried to group them according to context or setting so that when the attributes from the relevant articles are merged a big picture can be seen on which a theory can be build.

However attributes could be considered either a moderator OR a mediator. It depends on the theory being tested, for example; Gender could be viewed as a Moderator e.g. The effect of social support on depression varies depending on gender, but also as mediator e.g. Social support has an effect on depression mainly because of an underlying difference between social support levels of males and females. This explains why a same attribute can be found in the mediator as well in the moderating model. The ultimate effect also known as the dependent variables are gathered from the relevant articles for this study; these attributes have the same interest which is a success in terms of project outcome, implementation, performance, control, or learning. The illustration below gives an overview of the elaborated version of the research model, the conceptual models used in the relevant articles are coded and merged according this scheme done on a Level of analysis.

Skill Leadership Attributes Problem-solving defining

seeing big picture analytical Social Judgment Skill/people Perceptive taking,

Social perceptiveness behavioral flexibility social performance create trust atmosphere Knowledge technical

organizational Conceptual Visioning

(21)

21 The coding results from the conceptual models used in the relevant articles will be discussed in this part of the study. As mentioned before the results will be discussed based upon the four levels of analysis; individual, group, organizational and inter-organizational. Each of these levels will handle the independent variable attributes found, moderators, mediators and ultimate effect attributes, and the relationships between independent variable, moderator, mediator and Ultimate effect.

Figure 2. Research model

Level of analysis: Inter-organizational level Level of analysis: Organizational Level

Level of analysis: Group level Level of analysis: Individual level

Ultimate effects/outcomes

Performance,

IT adoption

Control

Knowledge & Learning

Decision making

Communication, Consensus & Conflict Resolution

Quality Improvement

Moderators

Organizational: Hierarchy, Technology

Experience Of Subordinates, Organization Size, Resources

External Environment: Economic,

Project Characteristics:, Speed of

Consensus, Support, Training, Experience

Group: Planning, Decision, Risk

management Mediators

Technology: Usefulness, Ease of Use Organizational; Facilitating Conditions,

Fit, Culture, Communication, Innovativeness, Vision, Power

Group:, Conflict, Communication,

Performance, Competence, Effort

Follower attributes: Valence, Attitude

Acceptance, Knowledge, Openness, Responsibility, Integrity,

Project Characteristics:, Control,

Leadership Emergence, Collaboration, Behavior, Decision Making, Participation, Social Interaction, Conflict Behavior, Diverseness, Cohesiveness Satisfaction, Motivation, Valence, Attitude

Venture startup Leadership Independent variable

Leadership Traits: Intelligence, Alertness,

Insight, Responsibility, Initiative, Persistence, Self-Confidence, Sociability, Masculinity, Adjustment, Dominance, Extraversion, Conservatism, Achievement, Cooperativeness, Tolerance, Influence, Drive, Motivation, Interpret, Confidence, Cognitive Ability, Conscientiousness, Openness, Agreeableness, Self-Monitoring, Emotional Intelligence

Leadership Skills & knowledge: Problem-Solving Skills: Defining, Seeing Big Picture, Analytical, Social Judgment Skill/People Skill: Perspective Taking, Social Perceptiveness, Behavioral Flexibility, Social Performance, Creating Trust Atmosphere

Knowledge: Technical, Organizational Conceptual; Visioning, Strategy Defining

Power: Referent, Expert, Legitimate,

(22)

22

Individual Level of Analysis

Leadership Characteristics

Traits, the first part of this research discusses the traits, skills and power attributes from the leadership independent variables as shown in the framework. The traits found from the selected relevant studies are, task knowledge, sociability, self confidence, responsibility, problem solving persistence, intelligence, integrity, initiative, drive, cooperativeness, and adjustment

Adjustment, a form of adjustment found in the study of Smits, McLean and Tanner (1993) is in terms of autonomy and relaxed supervision. In their study, they grouped their respondents in high, medium and low achievers and used gender as a factor. The findings for autonomy and supervision are that high achievers are characterized by a desire for creative and challenging work that provides a sense of accomplishment, by task variety, autonomy, and completion, and by work that provides opportunities for advancement within their chosen careers (Smits, McLean, Tanner, 1993). The difference between males and females is that the characteristics they prefer in an I/S job are more alike than different, the only difference they found is that if one accepts the Scheffe Test results, the extent to which females prefer a more relaxed supervision in contrast to typical or close supervision. (Smits, McLean, Tanner, 1993)

Cooperativeness, this study found several articles writing about cooperativeness or in other terms relating to improved or declined cooperativeness, the study of Jiang, Klein, and Discenza (2002) hypothesized that pre-project partnering ensures improved project manager performance resulting in improved project outcomes. Pre-project partnering activities lead to more effective team characteristics e.g. good interpersonal relations among team members and with management. Conclusions similar to other works reviewed earlier are drawn from the strong relations supporting that strong project manager performance improves effective project team characteristics.

Smits, McLean, Tanner (1993) results, show that the high-achieving female and male I/S students describe Themselves lower than other male and female I/S students on the dimension of Cooperation. With males in general rating themselves significantly lower than Females with

moderate or low achievement. Adding to that, the inter-personal insensitivity, Poor people skills, and preference for working alone that Chusmir says is characteristic of people with a high need for achievement, and it suggests that getting high-achieving, entry level IS professionals socialized into the organization and workgroup maybe a real challenge for IS managers.

Napier, Keil and Tan (2007) findings related to cooperation is that their subjects also

(23)

23 Contribution, in the study of Ransbotham and Kane (2011) They researched how members contribute to Wikipedia, first members begin by observing the activity of the community, then start to contribute, and ultimately might end up as moderators and leaders of the community

(Ransbotham, Kane, 2011)

Integrity, although integrity is an important trait of leadership there isn't much written about integrity in the relevant articles, in the study of Napier, Keil and Tan (2007) they write that the IT project managers who demonstrate personal integrity acts in a manner consistent with high ethical standards,, as opposed to self-interest, they believe this is an attribute that is important for

successful project managers. Interesting is that they perceive it as a skill because they believe that an ethical behavior is something that can be taught if a person is provided with the proper foundation and mentoring (Napier, Keil and Tan, 2007), instead of what other literature suggest a trait. That may explain why they find it interesting that personal integrity only accounted for less than 5% of the personal constructs elicited in their study as they claim that it had not been identified in prior studies (Napier, Keil and Tan, 2007), due that other studies would include it in traits. However if they do not find it in other studies who do include it in personal traits it would be an interesting finding as Northouse (2012, p23) describes personal integrity as one of five major leadership traits.

Intelligence, is also one of the five major leadership attributes described by Northouse (2012) though it is only found once at the individual level of analysis as described earlier. Smits, McLean and Tanner, found that High achievers, in contrast to low achievers, describe themselves as more punctual, intelligent,educated, industrious, organized, insightful, and independent (Smits, McLean, Tanner, 1993). Which may explain why the high-achievers also see themselves as more capable of providing leadership to others (Smits, McLean, Tanner, 1993), one can argue that self-confidence is also an important factor – as described in the five major leadership traits in the work of Northouse (2012) - why the high-achievers see themselves as better leaders, however, this attribute cannot be found in the study of Smits, McLean and Tanner (1993).

(24)

24 motivated, what is seen here again is that this study includes a trait motivating as a skill as is also seen earlier with integrity in the study of Napier, Keil and Tan, (2007).

Problem solving, the study of Napier, Keil and Tan (2007) identified problem solving skill as the ability of the IT project manager to address problems efficiently and effectively. They stated that the other literatures were missing the problem solving skills entirely (e.g. Jianget al. (1998), and the El-Sabaa (2001) study did mention the importance of a ‘strong problem orientation'. The Napier, Keil and Tan (2007) research clearly shows that competent IT project managers must proactively address problems. They must know how to analyze the root causes of problems and be willing to take responsibility for any problems that occur. (Napier, Keil and Tan, 2007). Again interesting to see is that this study grouped the problem solving attribute as a skill instead of what general leadership literature study does, group it as traits.

Responsibility, two articles, mentioned responsibly in their research, Napier, Keil and Tan (2007) mentioned that competent IT project managers must proactively address problems. They must know how to analyze the root causes of problems and be willing to take responsibility for any problems that occur. Smits, McLean, Tanner, (1993) state that early opportunities for success and felt responsibility fit very nicely with what is known about persons with a high need for achievement and, therefore, should be given serious consideration by IS managers attempting to integrate, and retain, high-achieving IS professionals into their workgroups (Smits, McLean, Tanner, 1993). One interesting side note is that the article of Napier et al. (2007) relates responsibility as a characteristic of

competent project manager, giving the impression that there is not a clear distinction between leader and manager, a better reading of their article also reveals that they use leadership as part of management skills, with competencies as Communicate goals, Likeable, Communicate a vision, Political savvy, Level headed, Calm, Positive attitude (Napier, Keil and Tan, 2007). This suggests that literature sometimes interchangeable make use of management and leadership, while leadership and management are clearly two different things.

Leadership can be traced back to Aristotle, where management emerged around the turn of the 20th century with the advent of our industrialized society. Management was created as a way to reduce chaos in organizations, to make them run more effectively and efficiently. The primary functions of management, as first identified by Fayol (1916), were planning, organizing, staffing, and controlling whereas the primary function of leadership is to produce change and movement, setting goals and creating visions. (Kotter 1990). Although Northouse (2012), clearly agrees with the difference between leaders and managers, they do, however, treat the roles of managers and leaders similarly and do not emphasize the differences between them.

(25)

25 Napier et al. (2007) states that The IT project manager must be clear about the vision and goals of the project and are able to communicate that to others. Communication skills refer to the ability of the IT project managers to effectively speak, write and listen to. The perfect IT project manager knows how to communicate both with project team members and senior management. (Napier, Keil and Tan, 2007). Matsuoa, Wong and Lai (2008) also state that Both IT consultants and project managers felt that customer relationship skills were needed to perform their jobs, though they placed different levels of emphases on different sub-categories of customer relationship skills. For example, IT project managers focus on acquiring skills for communicating with customers and trouble-shooting, while IT consultants direct their efforts on building relationships with customers and understanding customer's problems. (Matsuoa, Wong, Lai, 2008)

Skills

As this study already has defined what the definition of skills is, it remains vague for literature what the difference is between skills and traits. If a study uses a trait as skill or vice versa this study handles it as defined in the research model (figure 2), which is based upon general leadership research.

Analytical, Matsuoa, Wong and Lai (2008) describe analytical skills as part of consultants In contrast to project managers, placed emphasis on analytical skills for problem solving and making propositions. One consultant explained his methodology of identifying the root causes of problems. (Matsuoa, Wong, Lai, 2008)

(26)

26 Vision/strategy, although an important factor of leadership it is only to be found once at the individual level of analysis, Napier et al. (2007) are very short about it stating that a successful IT Project managers must be clear about the vision and goals of the project and be able to

communicate that to others (Napier, Keil and Tan, 2007).

Coordinating, can be viewed as a typical management skills as Rost (1991) Northouse 2012 describes management a skill directed toward coordinating activities in order to get a job done. Having stated this, one can argue whether or not it should be included in leadership skills however, a leader must also coordinate his managers.

Napier et al. (2007) states that meetings can be a source of wasted time; the ideal IT project manager must know how to determine whether a meeting is required. If so, he or she should know how to effectively facilitate meetings. The ideal IT PM can save time by being organized and by reusing templates for project plans. The ideal IT PM also knows how to manage others.

Matsuoa et al. (2008) studied what the differences are between IT consultants and project managers in the context of tasks, where he found that the differences in experience-based learning processes between IT consultants and project managers may be due to differences in their skills and responsibilities. As a project manager's objective is to complete projects on time within a given budget, group management skills become necessary for managing their subordinates with different roles and skills involved in a project to ensure the projects can be completed on time and within budget. (Matsuoa, Wong, Lai, 2008)

Power

Expert power, is based on followers' perceptions of the leader's competence. In the study of Sarker and Valacich they hypothesized that the acknowledged expert within a group had a significant effect on group valence, their empirical test of the model highlights that the expert's opinion plays a key role in shaping the valence of the group toward a particular technology. While the observed phenomena is an encouraging result, highlighting that the group values the opinion of experts, the intention of the expert should also be monitored (Sarker, Valacich, 2010).

Legitimate power, Sarker, Valacich, (2010) opinion of the individual possessing structural power within a group in favor of a particular technology will positively affect the group's valence toward that technology, not supported.

Finally, while their study examines the nature of the influence of different types of subgroups and high-status individuals on the group's opinion/preference, it does not inform researchers

(27)

27 between an expert and the majority, the majority's view (which has a higher beta weight) would hold more strongly (Sarker, Valacich, 2010).

In terms of CIO power Leidner, Preston and Chen (2010) did a study regarding the role of CIO in IT assimilation at Hospitals. According to their literature, there is a wide body of research on CEO power, but the CIOs in hospitals play a role similar to IT consultants wherein the knowledge

differential between them and the Board provides them with power to influence perceptions on what is, and is not, technically possible as well as what is even a technical versus a social system. The Board may be particularly important in influencing IT decisions in the healthcare industry because many Hospital Boards are largely, or sometimes fully, comprised of MDs – representatives of the major, powerful user group that often hinders IT implementations. Support from the Board for IT projects could go a long way in helping win support among the physician staff (Leidner, Preston, Chen, 2010)

Referent power, refers to the ability of a leader to influence a follower because of the follower's loyalty, respect, friendship, admiration, affection, or a desire to gain approval. The only small part on referent power that the study found at the individual level is that the opinion of the individual possessing personal power in favor of a particular technology will positively affect the group's valence toward that technology which was supported in the study of Sarker and Valacich (2010)

Experience, Dos Santos, Patel and D'Souza found in the context of venture capital funding that the extant literature suggests that business risk and agency risk consisting out of startup experience, industry experience, leadership experience, and management experience can be important considerations in venture capital funding. Strikingly, their study found that agency risk is not a significant factor in VC funding of IT ventures at start-up (Dos Santos, Patel, D'Souza, 2011)

(28)

28

Moderators

As discussed earlier on in this chapter, this study states that as there can be numerous moderators, it can be difficult to group them and elaborate them to a possible theory. However with the reference model this study uses it tries to summarize and group moderators accordingly.

Organizational, organizational factors contain characteristics, or policies unique to

companies, in order to group the findings at individual level of analysis the following attributes were used, size, hierarchy, culture, innovation/IT impact and competitive environment where all attributes are internal variables except the which is an external environmental attribute.

Size, company size, is used in the study of Armstrong and Sambamurthy (1999) they hypothesized that company size has an impact on IT assimilation as larger businesses possess more slack resources and, therefore, a greater capacity for assimilating innovations and technologies (Armstrong, Sambamurthy, 1999). However, their findings did not show that organizational size had a direct impact on IT assimilation (Armstrong, Sambamurthy, 1999).

Hierarchy, The same study also used organizational hierarchy as moderator in terms of the hierarchical distance from the CEO to the CIO in the firm. Their study found that rich interactions between technical and managerial personnel did contribute to increased levels of IT-based innovativeness. (Armstrong, Sambamurthy, 1999)

Competitive environment, The study of Dos Santos et al. (2011) used the moderator

competitive conditions in the hypothesis that, the business risk of a new IT venture is determined by the market for the firm's products and the competitive environment the firm will encounter. Where a less competitive environment would mitigate the business risk for the firm's products. The

hypothesis was confirmed and in addition, their results show that the underlying effect of timing affects the market for the product and the competitive environment. (Dos Santos, Patel, D'Souza, 2011)

Group

Risk management, The results from Barki, Rivard and Talbot (2001) show that deviations from an ideal Risk Management Profile were negatively correlated with Performance, supporting the contingency model of Figure I and the research hypothesis. The closer a project's Risk Management Profile was to the empirically derived perfect profile, the higher its performance, both in terms of Cost Gap and System Quality. (Barki, Rivard, Talbot, 2001).

The model developed by Dos Santos, Patel en D'Souza (2011), suggests that virtual

(29)

29 in start-up funding for IT ventures. However, they did not find agency risk being an important

consideration in start-up funding for IT ventures. (Dos Santos, Patel, D'Souza, 2011)

Mediators

As for mediators this study did not find a large amount of mediators at the individual level of analysis. Therefore it can be hard to say something about them at this stage, this sub-chapter does however discuss the findings from the relevant articles.

Organizational mediators,

Fit, The article of Barki, Rivard and Talbot (2001) mentioned the attribute in context of, the better the fit between the level of risk exposure of a software project and its Management profile, the higher the project's performance. Adopting a profile deviation view of fit. The research model was tested and supported (Barki, Rivard, Talbot, 2001).

Vision, this attribute, is also found as mediator in the sense that CIOs' business and IT experience significantly influence their firms' IT assimilation; this influence was found to be significant across firms with both the trans-form and inform down visions. (Armstrong, Sambamurthy, 1999)

Group mediators,

Performance, Jiang, Klein and Discenza (2002) state that pre-project partnering activities lead to improved project manager performance, Concluding that similar to other works reviewed earlier are drawn from the strong relations supporting. Also strong project manager performance improves effective project team characteristics, and strong project manager performance improves project outcomes, (Jiang, Klein, Discenza, 2002)

Project

Project characteristics mediators,

(30)

30 participation in the top management team have significant associations with each other (Armstrong, Sambamurthy, 1999).

The study of Barki et al. (1999) developed an integrative contingency model of software project risk management. Their findings are that when project performance was measured with system quality, high risk exposure projects had significantly higher levels of User Participation than low risk exposure projects. It suggests that, in such cases, project leaders should encourage users to assume greater responsibilities and to play a more active role in the project. They do stress however that he success of a project needs to be clearly identified from the outset (Barki, Rivard, Talbot, 2001).

Valence, Sarker and Valacich stated that the effect of the members' attitudes toward the technology on the group's valence would be moderated by the communication media used by group members, with a priori attitudes having a stronger effect in the case of computer mediated (CM) groups was. Supported. Also they hypothesized that the effect of the majority's opinion toward a technology on the group's valence was also moderated by the group communication media, with the influence of the majority being greater in face-to-face than in CM groups (Sarker, Valacich, 2010), which was supported. Also they stated that group's valence toward a technology would have a significant effect on the group's adoption of the technology, the observed relationship was supported (Sarker, Valacich, 2010).

Conflict, Relating conflict to valence the same study of Sarker and Valacich (2010) found that substantive conflict did not have a significant effect on the group's valence toward the technology, which was not consistent with their earlier hypothesis.

Perception, predicting the positive effect of the group's perceptions about the fit of the technology with the group task, and predicting a negative effect of the complexity of technology on the group's valence were confirmed (Sarker, Valacich, 2010).

(31)

31 Heng, Tan and Wei (2003) investigated when subjects are more willing to continue with de-escalated projects, their findings are that the use of positive feedback yielded higher willingness to continue with the project than negative feedback (Heng, Tan, Wei, 2003). However Under the condition of pessimistic framing feedback direction had no effect, positive and negative feedback Yielded comparable willingness to continue with the project. Pessimistic framing produced higher willingness to continue with the project compared to optimistic framing. Collectively, these two sets of analyses revealed that, when accountability was low, subjects were more willing to continue with the project when they were provided with optimistically framed positive feedback and, to a smaller extent, pessimistically framed negative feedback. Under conditions of high accountability, feedback direction and optimism appear to independently affect willingness to continue with software

projects. But under conditions of low accountability, feedback direction and optimism seem to jointly affect willingness to continue with software projects (Heng, Tan, Wei, 2003).

Ultimate Effect

Outcomes are based on reoccurring themes, venture startup, project management, implantation and IT adoption, control, performance and effectiveness, consensus and resolution, communication or knowledge and learning. At the individual level of analysis the following findings were found.

Venture startup, The study of Dos Santos et al. (2011) describe IT venture capital funding success factors, they wanted to find out what the importance is of a young company's business risk and agency risk to the company's success. While VCs try to invest in companies with low business risk and low agency risk, different VCs claim to weigh one or the other more heavily at the margin. The results suggest that for new IT ventures, VCs place less importance on agency risk. It suggests that the requirements for VC funding may differ on another dimension, namely, the type of business or industry (Dos Santos, Patel, D'Souza, 2011)

Project management, In the Article of Heng, Tan, Wei (2003) study they describe that giving positive feedback indicates good prospects for success, framing positive feedback optimistically may induce managers to persist with their projects and keep them from prematurely terminating these. Given that negative feedback indicates poor prospects for success, framing negative feedback pessimistically may discourage managers from persisting with their projects and possibly succumbing to escalation of commitment tendencies. (Heng, Tan, Wei, 2003).

Chen, Preston, Xia (2010) contribute to existing literature by advancing the understanding of the nature of CIO leadership, the individual and organizational factors that facilitate the CIO's

(32)

32 further support for a staged model in which CIO leadership mediates the effect of key antecedent variables on the impact of IT.

Napier, Keil, Tan (2007) their research contributes significantly to the understanding of skills associated with successful IT PMs, first their study confirms the importance of many skills previously described in the literature, while providing a richer understanding of several other skills that were narrowly defined previously.

Implementation, or assimilation of a technology whether this is due innovation or consolidation of systems is a reoccurring theme in scientific research. First Armstrong and Sambamurthy (1999) looked at the factors that influence IT assimilation at firms, their findings mainly from senior level executives are that Senior business executives' IT knowledge did not have a significant influence on IT assimilation. Interactions between the CIO and members of the top management team did not have a significant influence on IT assimilation, Organizational size did not have a direct impact on IT assimilation The sophistication of IT infrastructures has a significant influence on IT assimilation (Armstrong, Sambamurthy, 1999)

Performance, Sarker and Valacich (2010) found that only a small percentage of the variance in task performance was explained by the strength of adoption of the technology. This indicates that the adoption of a particular technology in itself may not be sufficient to generate a higher quality flowcharting solution for the group. The results also indicated a strong effect of the groups' adoption of the technology on members' satisfaction, and the overall task performance. (Sarker, Valacich, 2010).

Leidner et al. (2010), found that hospitals that are more innovative in their approach and use of IT show greater IT impact that contributes to hospital performance. They observed that firms that are innovative in their use of IT are better able to create IT capabilities that impact the contribution of IT to the performance of the hospital. In their model, they found that IT innovation explained 39.3% of the variance in IT impact, which consequently explained 10.6% of the hospital's level of financial performance (Leidner, Preston, Chen, 2010). Barki et al. (2001) results show that when meeting project budgets is the key Performance criterion, high-risk projects call for high levels of internal integration and High levels of formal planning. On the other hand, when System quality is the key performance criterion, high-risk projects call for high levels of user participation. This suggests that, in such cases, project leaders should encourage users to assume greater

(33)

33 Knowledge & Learning, This indicates that project managers tend to acquire knowledge and skills through apprenticeship training or in community of practice, while consultants tend to learn by themselves. Their results suggest that the acquired skills and experience, and the experience-based learning processes are different among the consultants and project managers in different career stages. (Matsuoa, Wong, Lai, 2008). Ransbotham and Kane (2011) provide empirical evidence that in the Greece Wikipedia setting for researchers, the study provides moderate levels of membership turnover positively affect collaborative success. Some membership stability is necessary to retain the information and knowledge generated by the community, but turnover also is desirable to introduce new information to the community. Moreover, knowledge retention in online communities is fundamentally different from knowledge creation (Ransbotham, Kane, 2011)

Group level of Analysis

Leadership Characteristics

Traits, first thing that is interesting is that there aren't many papers describing individual leader attributes at the group level of analysis.

Decision quality, Leadership was found to have no effects on consensus and decision quality in both unsupported and GSS groups, and both identified and anonymous groups. This lack of effects has been attributed to possible confusion arising from the overlapping missions of facilitators and leaders, as found in the study of (Tan, Wei, Partridge, 1999). In a study of Ho and Raman (1991) On the other hand, the use Of a GDSS does not appear to increase decision quality for preference and cognitive Conflict tasks (Ho and Raman, 1991).

Integrity, though there are not many studies that describe individual leadership attributes there are, however, three that discuss the importance of trust, the study of Sarker, Ahuja, Sarker, and kirkeby (2011) suggests that trust remains critical even when it comes to explaining or predicting the performance of the individual team member. The results indicated that in the high trust

centrality set, the interaction between communication and trust had a positive effect on

performance, whereas in the low trust centrality set, the interaction between communication and trust had a negative influence on performance (Sarker, Ahuja, Sarker, and kirkeby, 2011)

Gallivan's (2001) contribution to the trust attributes was that the first observations regarding his content analysis was that there were relatively few references to trust in the nine case studies. Five of the nine studies lacked any reference to trust at all, whereas the other four studies included only a few references to trust. It appears that trust and control may operate independently of each other to shape the behavior of individual contributors, as well as to deter-mine the overall

(34)

34 than to trust in the nine case studies, he argues that the OSS movement appears to rely on explicit forms of control to a much greater degree than on trust (Gallivan, 2001).

The study of Davis, Khazanchi, Murphy, Zigurs, and Owens (2009) briefly mentions trust in their literature and in one proposition but only answer it in vague terms, their question was; How does the ability to mask or make anonymous a person's out-world persona, gender, or culture impact interaction and trust. Their answer is that the ability for avatars to interact with each other and to provide immediate feedback on the performance of others is unique in their environment and can serve to enhance coordination. Therefore, they do state that future research is needed to determine how individuals use specific metaverse technology capabilities to improve participation and

interaction (Davis, Khazanchi, Murphy, Zigurs, and Owens, 2009).

Sociability, Sarker, Ahuja, Sarker, and kirkeby (2011) results indicate that the additive model does not explain the role of trust and communication on performance in distributed teams, given that both trust and communication do not have an effect on individual performance (i.e., only trust had a significant effect). The results cast doubt not only on the "twin predictions" made in the literature (regarding the effect of trust and communication on performance), but also on the "babble hypothesis," where "talkative" individuals are argued to be perceived as key contributors to the team. Indeed, their results show that the mediating model best explains the impact of trust and communication on individual performance in distributed teams. The strong support for the mediation model emphasizes the point that communication's effect on individual performance is through trust (Sarker, Ahuja, Sarker, and kirkeby, 2011).

Kayworth and Leidner (2001) state that, although communication may still be important in traditional teams settings, it may take on added importance in distributed groups as indicated in the study. Likewise, building and maintaining a proper social climate may become a vitally important activity for leaders of virtual teams (Kayworth and Leidner, 2001)

George, Easton Nunamaker and Northcraft (1990) found that there were main effects for Communication medium for three of the dependent variables; consensus, time to decision, and participation. Manual groups took less time to reach a decision, taking an average of 15.8 minutes, while GDSS groups took an average of 30 minutes. Manual groups were more likely than GDSS groups to reach consensus. On the other hand, GDSS groups were much more likely than manual groups to have equal participation during the group process (George, Easton Nunamaker and Northcraft, 1990).

(35)

35 towards and increase in PI group outcome quality. A negative effect of EC support indicated in the causal model in Figure 2 is that it apparently increases group discussion ambiguity, by, for example, eliminating non-verbal cues that add extra meaning to words and sentences, as well as adding feedback delay into the communication. Postings tend to take some time to be replied to, which may lead to some confusion as to what they are referring to when they eventually reach the group. Although EC support is perceived as reducing the duration of PI groups when compared with no GSS support at all, it seems to foster more reflection on the part of the members (Kock 2001).

Warkentin and Beranek (1999) research shows that: teams are not receiving the VTC training started out with fairly high evaluations in terms of the three relational variables, but these

measurements steadily decreased throughout the project. Individuals that received VTC training started with lower evaluations in terms of the three relational variables, but these measures steadily increased throughout the project until they were much higher than those in the control group. This indicates that VTC training led to increasing perceptions of cohesiveness and satisfaction with process over time (Warkentin and Beranek, 1999)

Skills

Conflict Solving, Sarker, Valacich and Sarker (2005) modeled substantive conflict and developed measures to highlight conflict, however, they state that conflict was not a significant attribute for shaping group dynamics (Sarker, Valacich and Sarker, 2005). Tan, Wei, Lee-Partridge (1999) Groups with leadership had marginally higher consensus than groups without leadership. However it is valuable to have a leader who pro-motes mutual respect among group members, encourages them to participate constructively, and helps them to resolve conflicting opinions through consensus (Tan, Wei, Lee-Partridge, 1999).

Perceptive taking, Kayworth and Leidner (2001) results to their research question - what factors contribute to effective leadership in virtual team environments? - yielded some interesting results. Virtual team leaders rated as effective by their members, demonstrate first and foremost a mentoring quality characterized by concern for the members, understanding, and empathy. Al-though these qualities may also describe effective leaders in face-to-face environments, the difference might well be in the ability of the virtual leader to project these qualities (Kayworth and Leidner, 2001)

Leadership in Information Systems A Literature-analysis Study