WIXLIB

6M. Nakayama and N.G. Sutcliffe2.3 Information SystemsThe execution of leadership transpires through the availability and access of information.Given that, modern organisations critically depend on IS and the appropriate use ofleadership styles, the relationship of IS and its associated Information Technology (IT) toleadership behaviour has received little scrutiny.Past studies analyse the impact of IT on organisational structure and function. The useof IT increases the timeliness and quality of organisational decisions, the autonomy ofmanagers and their span of control (e.g. Huber, 1990; Ferioli and Migliarese, 1996;El Louadi, 1998; Lau et al., 2001; Sor, 2004). As a lateral or horizontal coordinationmechanism for managerial communications, the increased use of IT decentralisesmanagerial decision-making (Ferioli and Migliarese, 1996; Hitt and Brynjolfsson, 1997;Malone, 1997), decreases the size of firms (Brynjolfsson et al., 1994), and reduced levelsof hierarchy (Ferioli and Migliarese, 1996). There is also a fit between computingarchitectures and organisational forms (Fiedler, Grover and Teng, 1996). From ahistorical perspective, down-sized, disaggregated forms of organisations in the late-20thcentury have replaced the large, stable centralised organisations prevalent in the mid-20thcentury (Winter and Taylor, 1996).Receiving little attention is whether the decision-making leadership styles interactwith certain types of IS to influence the effectiveness of that IS. A fundamentaltheoretical underpinning of the above studies is their common perspective; they vieworganisations and the groups within it as an information processing entity. They look onorganisational structures as mechanisms which process information so that theseorganisations can respond to their environment (Galbraith, 1973, 1977).Under such a perspective, we view leaders as the controllers of information formaking decisions. Data are defined as facts. Davis (1999, p.108) defines information as“data that has been processed into a form that is meaningful to the recipient.” Thedistinction between data and information is critical for IS, because information reducesuncertainty for decision-making. This is especially relevant when evaluating the impactof IS on a leader’s behaviour. The key functions of IS are to collect, store, disseminateand present information; the utility of IS for leaders is through what and how IS providesrelevant information to leaders. Accordingly, to assess the IS impact, we regard leaders asagents who consume and manipulate information to make decisions.Past studies provide some insights into how leaders process information. Dearbornand Simon (1958) report that managers’ information processing is influenced bydepartmental (e.g. marketing, finance and accounting) biases. Walsh (1988), in contrast,notes that managers do not necessarily process information just for their departmentalconcerns. The study by Thomas and McDaniel (1990) shows how the informationprocessing structure (that is, the levels of interaction and formalisation) of topmanagement teams influences the way Chief Executive Officers (CEOs) process strategicinformation.In the following sections, we take a brief look at specific leadership styles and IS.Then, we follow with the leadership framework used in this study, the propositions,implications, future research directions and conclusions.

The role of leadership decision styles on the use and effectiveness of IS37Leadership and decision-makingA research consensus developed that most leaders use a variety of leadership styles,ranging from the autocratic to laissez-faire based on the situation in their interactionswith members (Bass and Valenzi, 1974; Flamholtz, 1986, 1990; Scandura, Graen andNovak, 1986; Sutcliffe, 1999). In addition, leaders tend to use a variety of styles when thedecision must have subordinates’ acceptance and/or must be high quality solutions(Heller, 1972). However, Wilcox (1982) found that the perceptions of leaders andsubordinates do not always match. Usually, leaders rate themselves more consultativethan their members rate them. So, the same actions are interpreted differently based onthe position of the person within the group.The one commonality is that all these studies look at the conditions that indicate thata given decision-making leadership style is preferable. These studies do not look at theimportance of a decision-making leadership style for the determination of future IS-basedinformation flows between leader and members.However, the Bass–Valenzi model (Bass, 1976) did look at non-computerisedinformation flows and leadership styles. Here, the leader and members form an opensystem with inputs and outputs that are moderated by the styles and the ‘within-systemrelations’. The inputs consist of attributes of the organisation such as constraints, clarity,warmth; attributes of the workgroup such as intragroup harmony, interdependence;attributes of the task such as clear objectives, routine and complexity; attributes of theattitudes such as fair, assertive and introspective. The outputs consist of work groupeffectiveness, satisfaction with supervision and job satisfaction. The moderating ‘withinsystem relations’ consist of boss power, subordinate power, information that the boss has,information that the subordinates have, the long-term objectives and the structure. Themoderating styles are on a continuum from directive to delegative (similar to laissezfaire). In this model, information is always within the model; it is with the leader or withsubordinates. It does not tell us how the decision-making leadership styles interact withthe IS, which affects the group’s effectiveness in the organisation. Thus, a key questionof this article is: given a leadership style, what are the IS requirements that facilitate useof that leadership style?3.1 Leadership decision-making categories and stylesAs already mentioned, research has identified several leadership styles on the continuumof decision-making authority (Tannenbaum and Schmidt, 1958; Likert, 1961, 1967;House, 1971, 1997; House and Mitchell, 1974; Flamholtz, 1986, 1990; Yuki, 2006).There are also corresponding decision procedures from the LMX research (Scandura,Graen and Novak, 1986). These styles/procedures are based on the degrees of freedomleaders give to their members for decision-making and can group into three categories.The categories are styles are:1directive (autocratic or benevolent autocratic style)2interactive (consultative or participative style)3non-directive (consensus or laissez-faire style, Table 1).

8M. Nakayama and N.G. SutcliffeTable 1The leadership categories, leadership styles and decision proceduresLeadershipcategoryLeadership style Definition – leaderDirectiveAutocraticDeclares what is to be done ‘Autocratic’without explanationBenevolentautocraticDeclares what is to be done – None specifiedwith an explanationConsultativeGets opinions beforedeciding on the planpresentedInteractiveParticipativeNon-directive ConsensusLaissez-faireSource:LMX decision procedures‘Minimal DecisionInvolvement’ frommembersFormulates alternativeswith members, then leaderdecides‘Consultation’ more frommembers ‘Collaboration’even more from membersin formulating alternativesGives all in the group haveequal voice in makingdecisions– None specifiedLeaves it to the member(s)to decide what to do‘Delegation’Scandura, Graen and Novak (1986) and Sutcliffe (1999).Among leaders, there are significant differences in the delegation of decision-making. Inthe directive leadership category, managers give marching orders with or withoutexplanation. Only the leader defines problems and their priorities. In addition, the leadergives orders regarding when, what and how tasks are accomplished, and monitors closelythat these orders are followed. In LMX research, only the autocratic is discussed with nomention of the benevolent autocratic where orders are given with an explanation.Next on the delegation continuum, the interactive (consultative or participative)category has leaders seeking inputs and formulating alternatives with their membersbefore making the final decision. The communication is two-way and frequent betweenthe leaders and their members on defining problems, monitoring progress and devisingsolutions. Such communication is on-going.As mentioned before, LMX is interested on the differences between highly valuedmembers (high LMX) and other members (low LMX). LMX posits that leadercommunication and collaboration is much stronger with high LMX members from the‘in’ faction than with low LMX members who make up the ‘out’ faction. So, under LMX,the leader uses the consultation and collaboration decision procedures with the ‘in’faction and the minimal decision involvement decision procedure with the ‘out’ faction.In contrast to the previous styles, leaders using the non-directive (consensus orlaissez-faire) category style do not give direct orders. Instead, they delegate the decisionmaking to their subordinates. It is the members who define problems and determinesolutions while always informing the leader of the steps taken. However, the leader canintervene when necessary. Under LMX, the delegation decision procedure is only usedwith high LMX ‘in’ faction members.Effective managers use the directive category more often when the environment isunsettled and turbulent requiring quick decisions, whereas they use the non-directive

The role of leadership decision styles on the use and effectiveness of IS9category more often when the environment is stable (Bourgeois, McAllister and Mitchell,1978). In other words, there is a life cycle of dominant categories as environmentalstability goes through cycles such as:1organisational startup, growth, maturity and decline2market growth and decline (Muczyk and Reimann, 1987).3.2 Use of multiple leadership styles?A leader is more comfortable using one category during a major task assignment, abusiness/organisational phase or even the leader’s tenure in a job position (Fiedler, 1971).This comfort dictates the assumptions that a leader makes on the information needed tomanage the group. This is not quite the same as the leadership styles used whenmanaging and interacting with subordinates/members. Some claim that effective leadersjust focus on member needs in a situation and effective leaders change the style to meetthat situation. Such a view is found in the contingency theory (Fiedler, 1967) and thesituation leadership theory of leadership effectiveness (Blanchard, 1991; Irgens, 1995;Hersey and Blanchard, 1996). These theories posit that effective leaders change theirdecision-making leadership style as appropriate.If leaders change decision-making leadership styles flexibly and frequently, it wouldbe pointless to investigate the optimal IS requirements to target as a good match for eachleadership style. IS cannot easily change their characteristics. They are relatively stableover time. But, what about the stability of the deter

effectiveness of Information Systems’, Int. J. Information Systems and Change Management, Vol. 3, No. 1, pp.3–15. Biographical notes: Makoto Nakayama is an Associate Professor at the College of Computing and Digital Media in the DePaul University. He holds a PhD from the