How Effective Principals Think While Solving Problems

How Effective Principals Think While Solving Problems

Angeliki Lazaridou

ABSTRACT: The problem-solving processes and strategies of ten school principals with reputations for high
effectiveness were studied. Verbal protocols were gathered through unstructured interviews while principals were
working on a set of five cases. Analysis of the data was guided, but not constrained, by a set of processes and
strategies for problem solving suggested by the models of problem solving developed by Leithwood, Voss, and
Mumford. Results describe the categories of problem solving processes that emerged across the five problems, the
frequencies with which these processes were used, graphic representations of the strategies principals used to solve
the problems, as well as the consistency of using the same strategy across problems.

Although leadership has long been fertile ground for research and debate, there are many contradictions in the
accumulated knowledge base, and certain areas still remain unexplored. One neglected area relates to how
administrators solve problems. Before Haller and Strike’s (1986) work, this topic had rarely been considered directly
in the field of educational administration. Up until then, problem solving had been addressed primarily in the context
of decision-making, and investigations failed to provide understandings of how administrators interpret and think
through the problems that give rise to decisions. Recently, it has been suggested that a focus on cognitive processes
may contribute to a more complete theoretical frame for studying how administrators solve problems (e.g., Hart,
1999; Zaccaro, Mumford, Connelly, Marks, & Gilbert, 2000). In this regard, Leithwood has been a pioneer,
maintaining that inquiry into how administrators perceive, interpret, understand, and solve problems lies at the heart
of understanding and improving educational administration:
Efforts to understand the roots of effective administration … will be much more productive if they shift from a focus on
action or behaviour to a focus on thinking or problem solving. (Leithwood & Steinbach, 1992, p. 318)

The study I report here is grounded in the cognitive approach to investigating leadership and focuses on the school
principalship. Specifically, I examined the cognitive processes and strategies highly effective school principals
engage in when solving organizational problems (Lazaridou, 2002).

A Cognitive Approach to Understanding Problem Solving

Over the last decade or so, an increasing number of researchers have argued that the focus of future research on
organizational leadership ought to be more on the cognitive dimensions of leadership and less on leadership styles
(Hart, 1999; Zaccaro, Mumford, Connelly, Marks, & Gilbert, 2000). The results support the contention that
understandings of leaders’ cognitions are important. More specifically, Mumford and his colleagues conducted a
series of investigations to probe the ways in which leaders go about solving organizational problems (e.g., Connelly,
Gilbert, Zaccaro, Threlfall, Marks, & Mumford, 2000; Mumford et al., 2000) and concluded that “successful leaders
are capable of identifying and solving significant organizational problems using an analysis of organizational
requirements and constraints, along with wisdom and perspective taking, to craft viable solutions likely to work within
the organizational context” (Mumford & et.al., 2000, p. 282).

In the domain of educational leadership, Leithwood (1995) concluded that cognitive perspectives make three central
contributions to the study of educational administration: (a) they redefine the meaning of effective leadership by
focusing attention on expert, internal, cognitive processes, (b) they expand understandings of the knowledge base
required to exercise effective leadership, and (c) they enrich understandings of how leadership expertise develops.
Cognitive perspectives have been applied only recently in the field of educational administration and leadership,
highlighting two phenomena that are consistent with contemporary thinking about leadership. The first is that people
can handle only a certain amount of information at any given time, and thus are selective about the things they pay
attention to and process. The second is that people impose their own meanings on environmental events (Ormord,
1995). This notion is related to the concept of constructivism, which proposes that people construct their own
knowledge from their experiences.

Problem Solving Theory and the Notion of Expertise

In cognitive studies of problem solving in organizations, attention has been directed toward the manner in which
information is noticed, interpreted, and processed by administrators. One aspect of studies in this area is a focus on
the concept of expertise and its development among administrators (e.g., Chi, Glaser, & Farr, 1985; Reimann & Chi,
1989, Van Lehn, 1989; Ericsson & Charness, 1994). These studies have identified a number of differences in how
experts and novices solve problems. Specifically, it has been found that experts pay more attention to monitoring and
managing their problem-solving activities; they draw more on their domain knowledge to build more effective
representations of problems than do non-experts; they create more complete, abstract, coherent, and functional
problem representations (Reimann & Chi, 1989, p. 188); they identify and possess more complex goals for problem
solving (Bereiter & Scardamalia, 1986); they spend more time planning their strategies and are able to use a greater
variety of approaches to the task of developing a solution (Berliner, 1986); and they are more sensitive to the social
contexts within which problems are to be solved.

But these findings come from studies outside the educational domain; whether they apply to educational
administrators is largely moot.

In this regard three research programs are of interest here, two outside the educational domain, and one within. Of
the programs outside education, one is headed by Voss, Greene, Post, and Penner (1983) and the other by
Mumford, Zaccaro, Harding, Jacobs, and Fleishman (2000). In the educational domain, the work that stands out is
that conducted by Leithwood and his colleagues.

Voss and Colleagues

Voss et al. (1983) identified two structures in problem solving: a control structure and a reasoning structure, each with
its own set of operators.

The control structure relates to goal attainment, and controls the problem-solving process. It consists of seven
operators that act upon the individual’s knowledge base and generate the problem solutions. These operators are
state constraint, state subproblem, state solution, interpret problem statement, provide support, evaluate, and
summarize. The reasoning structure is complementary to the control structure, and includes the justifications that
subjects use during problem solving. The ten operators of the reasoning structure are state argument, state assertion,
state facts, present specific case or example, state reasons, state outcome, compare and/or contrast previous
statement, elaborate and/or clarify, state conclusion, and state qualification.

Voss et al. (1983) went beyond simply identifying the elements of the problem-solving processes used by experts;
they also described how these processes are organized into strategies that help solvers address the problem. A
problem-solving strategy is a patterned set of cognitive processes. Voss found three strategies or archetypal
combinations of processes:
· Decomposition, in which the main problem is broken down into a set of subproblems, usually no more than three.
· Conversion, in which the given problem is converted into another issue for which a solution may already exist.
· Identifying and eliminating the factors that contribute to the problem.

Mumford and Colleagues

Mumford, Zaccaro, Harding, Jacobs, and Fleishman (2000) have focused on the skills, knowledge, and social
judgements that seem to be related to leaders’ effectiveness in solving organizational problems. Their data indicated
that problem-solving skills and knowledge are the most potent in predicting effective leadership. Accordingly,
Mumford and Connelly (1991) reported seven skills grouped under two general categories:
· Generation skills – problem construction, information encoding, category specification, and category combination
and reorganization.
· Implementation skills – idea evaluation, solution implementation, and solution monitoring.

Leithwood and Colleagues

In the domain of educational leadership, Leithwood and his associates (1989, 1991, 1993, 1995) pioneered
investigations to describe educational leaders’ problem-solving processes. Their typology of problem solving
processes comprises six components that are grouped under three general categories (Leithwood, Steinbach, &
Raun, 1993; Leithwood & Steinbach, 1995; Allison, 1996):
· Understanding – interpreting the problem, setting goals
· Solving – identifying constraints, generating solutions
· Dispositions – considering values, considering mood

Design of the Study

Research Methods

This study was based on the analysis of verbal reports of the thinking and problem solving of selected school
principals. A verbal report is a problem solver’s account of his or her mental processing (Ericsson & Simon, 1993).
Results generated during the problem-solving process are called concurrent verbalizations. For the purpose of
studying problem-solving activity, concurrent verbal reports seem valuable because they provide a record of what the
participant is thinking about and attending to while solving problems, and these data are not marred by changes that
the participant may make after having had time to think more about the task. For this reason, this study involves the
collection and analysis of concurrent verbal protocols.

Participants

Ten school principals (five male, five female) from three schools boards in eastern and western Canada took part in
the study. An office administrator at the board level in each district was asked to nominate principals who were highly
effective school leaders. All sessions were audiotaped and subsequently transcribed with consent of the participants.

Data Collection

Data from participants were collected with a think-aloud procedure while principals were working on a set of five
problems. The problems were short cases based on true events that contained one or more organizational problems.
The think-aloud procedure was introduced to participants with a set of prepared instructions. These instructions
asked participants to first read aloud each case in the order they were provided, articulating all thoughts that occurred
as they did so. Participants were specifically instructed to try to express all their thoughts from the moment that they
were handed the cases until they had finished dealing with them. A practice problem was used to familiarize
participants with the process before actual data gathering began.

Cases

The cases were originally written by graduate students for a graduate Educational Administration and Leadership
course and were modified by permission from the authors in order to be used for this study. The cases reflected the
reality of a principal’s job as the authors had encountered it in schools as teachers, assistant principals, or the
principals themselves. Case 1 involves a conflict between a classroom teacher and a teaching assistant in a class
with special-needs students during a social science exam. Case 2 refers to the issue of changing a school’s
philosophy from a traditional mode to a more inclusive one with a new principal to be in the leading role. In Case 3
the issue is service downsizing and the transfer of a teacher into a classroom with high-need students. The focus of
Case 4 is the impact of financial constraints and low enrollment along with a restructuring policy on staff
empowerment. Finally, Case 5 involves the issue of disciplining a student after his/her use of abusive language
toward a teacher.

Data Analysis

The interviews generated 50 protocols that were analyzed according to protocol analysis suggested by Ericsson and
Simon (1993). The analysis was guided initially by a set of coding categories that were derived from the three models
of problem solving reviewed earlier. It is important to note, however, that the analysis was also emergent in that it
allowed for the addition of new categories as they became apparent in the data. The reliability of coding was tested
through inter-coder reliability checks. Agreement ranged between .9 and .96. Where discrepancies emerged, the
coding was adjusted accordingly.

Findings

Problem-Solving Processes

Table 1 reports the categories and the frequencies of problem-solving processes that were found in the verbal
protocols of principals.
The three categories of processes were labeled problem construction, plan formation and execution, and regulation.
The problem construction category comprehends three problem-solving processes: stating the subproblem,
representing the problem, and stating constraints. From the analysis of the protocols it became evident that the
principals tended to use these three processes to generate a representation and understanding of the problem and to
identify its parameters. The plan formation and execution category consists of eight processes: stating solutions,
elaborating/stating reasoning, stating examples from experience, stating values/beliefs/principles, presenting
alternative/hypothetical scenarios, making assumptions, using metaphors/analogies, and summarizing. These
processes, in concert, allowed the principals to formulate and carry out a detailed plan for addressing the stated
problem or the represented problem. The regulation category comprised two processes, being proactive and
evaluating. Regulation refers to monitoring and controlling the solving process according to the plan that was
formulated earlier, and also evaluating the plan and its appropriateness at implementation.

Following are the specific problem-solving processes evident in my data, organized under these categories.

Problem Construction

Stating subproblem(s)

Subproblems are typically regarded as important but subordinate to a more general problem. All principals identified
subproblems in the five cases they analyzed, but some saw more than others (frequency range = 3-13). This
category was evident in such statements as:

Another problem that is of importance here is the whole dynamic between the superintendent and the principal.

My finding here points to the significance of this category in the process of solving problems (frequency rank = 5).
Although the literature in the domain of leadership does not identify this element as part of the solving process,
evidence from other domains suggests that this category accounts for a large portion of the problem-solving process
(e.g., Voss, Greene, Post, & Penner, 1983).

Principals identified subproblems in two ways. They either stated a subproblem while breaking down the main
problem, or they “encountered” subproblems when they were exploring the implications of a proposed solution. For
example:

I’m thinking [about how] that tension had escalated to such a degree. I guess my thought is that principals need to
walk through their classrooms frequently. They need to talk to their staff frequently.

Representing the problem

Statements classified under this category indicated how principals made sense of the problem before they engaged
in the solving process. This category ranked thirteenth in frequency. It was evident in such comments as the following
two:

The problem in this particular one is probably more related to the communication that hasn’t gone around.

The problem I see straight away is that the superintendent is going to have this teacher put under his influence.

This finding can be understood better by comparing it with the supplementary category “stating facts from
experience.” When principals had encountered similar experiences in the past, they did not articulate this process.
Rather, they indicated that they knew how to solve the problem because of that experience.

Another explanation for this result may lie in the nature of the problems. Problems that principals perceived as “easy”
to solve did not elicit interpretive responses. In contrast, problems that they perceived as “complex” did elicit
interpretive responses. Even though problems were not described as easy or difficult to the principals, principals
inevitably provided their own evaluations of the problems either before they started the analysis or at the end of the
solving process. This finding is consistent with evidence from previous research, which suggests that the approach
an individual takes to solving a problem depends on the nature and the degree of difficulty of the problem (Leithwood,
1989, 1993).

Stating constraints

Statements included in this category identified factors that restricted the range of possible solutions. Statements of
this type ranked in a tie for twelfth in frequency, and included statements such as:

You know, either we have the money or we don’t. And this is how it is.

I think there comes a point as well [when] the mandate for the principal [requires] that; and, depending on whether the
school district says you have to do it, you do it.

The number of statements coded as constraints was small. It seemed that principals chose to identify only minor
problems that could be addressed and to pay less attention to obstacles that couldn’t be overcome. Thus, when they
encountered a constraint, they either attempted to convert it to a subproblem or they accepted the constraint as a
given and worked to find another way to solve the problem. The following example illustrates how principals
converted a constraint into a subproblem:

The only avenue we really have is to keep on pressuring for more government money, so that we can drop the class
size.
In principle, then, it is possible for any constraint to be converted into a subproblem – which complicates things for
analysts, since the distinction between subproblems and constraints becomes less definite.

My data also suggested that the principals’ use of the process “identification of constraints” depended on how a
problem was presented to them. When a case explicitly described constraints, the principals attended to them in one
of the two ways presented above. However, if a case did not identify the inherent constraints explicitly, the
participants generally did not devote attention to the “identification of constraints,” and focused on other aspects of
the problem instead. Evidence from similar studies confirms these findings (Leithwood, 1989, Voss et. al, 1983). This
finding is simply a reminder that principals’ problem solving cognitions are subject to how others – including
researchers – represent the problem in the first place.

Plan Formation and Execution

Stating solution(s)

This category included statements that indicated an implicit or explicit solution to a problem or subproblem.

I think the first thing I would do here is to bring the people together to sort of articulate what the teacher’s role is, and
what the teaching assistant’s role is.

I would work at separating the issues, to say to the teachers, “I support you in terms of helping kids take responsibility
for their actions.”

Statements of this type were most frequent in the principals’ responses – principals spent a good portion of their time
on identifying ways to solve the problems. They spoke of both short-term and long-term solutions, and they also
spent a lot of time talking in detail about those solutions. In some cases, the solution process dominated protocols.
An interesting aspect of the findings is that some participants went on to provide the reasoning behind the proposed
solution, and also to explore the implications of their solutions (more details about this below). Previous researchers
have suggested that this tendency in a leader is indicative of expertise (Leithwood, 1989, 1993; Voss, Greene, Post,
& Penner, 1983).

Elaborating/stating reasoning

Principals used this process to justify their proposed solutions, identify a subproblem, or generate a problem
representation. The principals provided elaborations at different stages during problem solving, and differed in the
number of reasoning statements to which they referred. The following examples illustrate how principals used this
process:

When you go into a school, you do not change philosophy. You do not change the focus. You listen, you learn, you
watch, you observe, you soak it up like a sponge. You become part of what is already there. You blend into the walls.
On occasion, you do have to assert in certain areas, and you go with the flow. You see where there are possibilities
of doing things differently.

Because if we don’t allow people to express their views, then I think we don’t help them grow. And it’s certainly very
prevalent in that there are many benefits to them. You help to see, help to broaden stakeholders’ perspectives, and I
think that’s a very healthy thing to do.

Stating examples from experience

This category of thinking was another way the principals sought to support their proposed solutions. Statements of
this type ranked third in frequency and were used by all principals across the five problems. Interestingly, two
principals started the analysis of three of the five cases by referring to examples from previous experiences. These
examples referred either to past experiences with a similar problem or to a current situation at the principals’ schools.
Another principal started his analysis of three of the five problems by referring to similar cases in his experience. Also
interesting was the fact that the principals referred mainly to successful incidents from the past. Examples of
statements in this category are as follows:

What we’d done before was build what we call school family, and with that school family we have the parents and the
kids try to work together as a team.
In our school, we have a lot of people who are so good with the kids that they give the kids some decision-making
powers. They nurture the kids. When we do get some individuals who might be more on the authoritative side, they
stand out.

Stating values/beliefs/principles

The principals used this kind of thinking quite frequently – frequency rank was fourth out of 13. Overall, 12 types of
values were identified during the analysis of the protocols. These are the values of care, collaboration, confidentiality,
consideration, effectiveness, fairness, integrity, mission, nurturing, personal values, professionalism, and
transactional values. Principals referred to professional values more often than any other values. Such statements
are exemplified by the following three, which feature the values of care, collaboration, and professionalism:

I think they realize that we are here for the kids. Our primary responsibility is to the students. And when someone is
unable to deal with kids, I have a responsibility as a principal to make sure that that is corrected, and to do it in a
compassionate way.

I think it’s important for me as the principal to know every child in the school and to know them personally, and I don’t
mean just knowing their name. And when we get too big, then we can’t do that. When we’re too small, we don’t have
the resources to do some of the kinds of things we’d like to do.

These findings support previous research that identified values as a major component of the problem-solving process
of school administrators (Leithwood, 1991; Leithwood, Begley, & Cousins, 1992). This study suggests that values
influence and shape principals’ thinking significantly when solving problems.

Presenting a hypothetical or alternative scenario

The principals used alternative or hypothetical scenarios when contemplating how to get around the problem of
having insufficient information. All principals used this category across the five problems. Two statements that
illustrate this category follow:

Another way of dealing with this would be to not say to the staff that we are going to make this change immediately,
but to say that we have a mandate to look at inclusive schooling, and to engage the staff in actually exploring the
whole concept of inclusive schooling, and how we might, as a school, achieve the model of inclusive school.

Now, what’s probably happened here is this: The specialist didn’t come ‘til September or maybe late September, so
the teachers already had had to teach three weeks [without having any] idea how to apply this mode.

Making assumptions

Principals sometimes made inferences about parts of the problems that were unclear or missing. For example, the
principals made assumptions regarding the timing of events: “I’m assuming that this is happening in June.” In some
instances, the assumptions they made were grounded in past experiences, and they were either explicit or implicit.
Six principals made use of this type of thinking, which ranked seventh in frequency of use.

I’m assuming that this school policy should have the support, if it’s under the umbrella of the jurisdiction policy, of the
parents and the staff, and would have been communicated very clearly to students, staff, and parents.

I’m going to assume, furthermore, that this school had a school council and in my jurisdiction one of the things that
the school council does is to review the school discipline code, and so I’m going to assume for the purpose of this
that the behavior code of the school was probably drawn up mostly by staff.
Using metaphors/analogies

The principals used metaphors and analogies to illustrate their approaches to problems. Five of the eight principals
used this category, and it tied for eighth in frequency. One principal used this category six times across the five
problems. This category is illustrated in the following statements:

I think that people who want to make changes in schools seem to think that teachers could just do this and they’re
going to do it forgetting that the teachers are the experts. I mean, you wouldn’t do that to doctors. You wouldn’t say to
a doctor, now, “We’re going to change the medical procedure.”

And I often ask them to take a look at what happens in their home with budgeting. You’ve got a pay cheque that
comes in, we only have so much money to run the family, and yes, we’d like to take that trip to Hawaii, but we make
some decisions about what we can and cannot do, and then we say sorry, we don’t have the money to do that.

Summarizing

Providing brief summaries at the end of proposed solutions seems to have had an integrative function; the principals
used this kind of thinking to put closure on a series of previous statements, in effect stating, “Here is where I am so
far.” This process also helped principals to catalogue and synthesize in their minds all issues that they considered
important for solving particular problems. Such statements occurred infrequently in the protocols, and only three out
of the eight principals engaged in this kind of thinking. This kind of statement is evidenced in the following:

So, through supervision, checking in myself, and also probably getting the school involved, [I would] see if we can all
just be monitoring that.

Although this category of thinking during problem solving is not evident in research in the educational domain, it has
been documented in research outside education (e.g., Voss, Green, Post, & Penner, 1983). That research suggests
that this category is employed by subjects not so much to describe the solution process as to summarize (perhaps
reinforce or confirm) a large portion of the “think aloud” protocols they have generated.

Regulation
Being proactive

This category accounted for a small number of statements made by four out of the eight principals. Principals used
this category when they wanted to emphasize the things they would have done to prevent the problem from occurring
in the first place. The following excerpts are examples:

The only thing that probably could have stopped this happening in the first place is, at the very beginning, prior to the
teacher aide taking that role, sitting down with the teacher aide and saying, this is what we are expecting in the
school and having them understand how they fit into the school situation.

This is a situation that certainly could have been avoided had proper planning taken place.

Evaluating

Statements of this type indicated the degree to which the principals supported a solution. They were small in number.
Principals used this category in three ways: to evaluate a solution in relation to a constraint, to evaluate a solution in
terms of what it would accomplish, and to evaluate a problem before attempting to solve it. This category included
such statements as:

I think the whole approach here is faulty.

For me, this doesn’t seem to be terribly complex.

Although there is no previous evidence in the area of education that supports this finding, research outside education
does support it (Voss, Green, Post, & Penner, 1983,1989).
Problem-Solving Strategies

To investigate whether the principals used distinctive strategies, I developed flow charts from the protocols to show
how each participant combined problem-solving processes to address each of the five cases. These charts were then
examined for patterns or regularities – in and across protocols – that could be called strategies. Principals were
expected to use the three problem-solving strategies identified in the literature – decomposition, conversion, and
neutralizing factors that contribute to problem. However, I concluded that four problem-solving strategies were
evident in my data. I designated them conversion, decomposition, reversion, and solution oriented.

Conversion

In conversion, the principals converted a stated problem into a new issue that they subsequently solved. Five
principals used the strategy of conversion during problem solving. After comparing the charts, it became apparent
that this strategy is characterized by the process “representing problem” early in the solving. In Figure 1, I present an
example of how one principal used this strategy to solve one of the problems.

Decomposition

In decomposition, principals broke the problem into one or more subproblems, which they addressed by proposing
related solutions. After comparing the charts, it became apparent that the process “stating subproblems” was the
dominant feature of this strategy. In Figure 2, I present an example of this strategy.
Reversion

The reversion strategy tends to be initiated when a principal starts the analysis of a problem by focusing on the steps
he/she would have taken to prevent the problem from getting to its present state. The definitional feature of this
strategy is early use of the process “being proactive,” which was used by all eight principals in at least some of their
problem solving. An example of this strategy is presented in Figure 3.
Solution Oriented

With the solution-oriented strategy, the principals focused exclusively on identifying a particular solution to the
problem. Although identification of a solution is at the core of principals’ problem solving, regardless of the strategy
used, some of the protocols were dominated by the plan(s) proposed for solving the problem. Salience of the process
“stating a solution” is the hallmark of this approach to problem solving. There were thirty-four incidences of this
strategy in the principals’ protocols, and it was therefore the most frequently used problem-solving strategy. An
example of this strategy is presented in Figure 4.
Table 2 summarizes the findings regarding the frequency with which the principals used these problem-solving
strategies. It is clear that the most frequently used strategy was the solution oriented followed by the strategies of
conversion, decomposition, and reversion.

One final note: With the exception of the one principal who used the same strategy across all five problems, no
consistency in the use of strategies was found. Different principals used different strategies to analyze the five
problems.

Conclusions

The findings reported here come from an investigation of the problem-solving processes and strategies of effective
school principals. To guide that investigation, I adopted an integrated problem-solving model derived from the work of
Voss, Mumford, and Leithwood. This integrated model was used to guide analysis of the verbal protocols of ten
reputationally effective principals. The emergence of hitherto undocumented processes necessitated adjustment of
the model.

A number of features of the study recommend caution in using the results. In particular, the small sample size, the
reputational method of sample selection, and the lack of a non-expert comparison group. Overcoming these
limitations in subsequent research would add considerably to the knowledge base. Notwithstanding these reasons for
caution, the results, in my view, provide a useful beginning for further research in the area of expert problem solving
and leadership.

My findings suggest that the problem-solving processes of effective principals are similar to those of experts in other
areas of the social sciences. Principals’ thinking consisted of two central elements: the goals to be achieved and the
reasoning to support those goals. Differences were found in principals’ approaches regarding the reasoning
component. Complex reasoning was characterized by the use of elaborate statements in support of an action plan. In
these statements principals used metaphors and analogies to emphasize their points, arguments to justify their
preferred courses of action, and evaluations of their proposed solutions. Previous research has suggested that this
aspect of the solution process is indicative of expertise (Leithwood, 1989, 1993; Voss, Greene, Post, and Penner,
1983).

Principals, like experts in other fields, devoted significant time thinking through a solution process and formulating a
detailed plan of action. They were able to detect similarities and differences with past situations and to use that
knowledge to solve the problems more effectively and efficiently. Their values, beliefs, and assumptions guided their
thinking and influenced the way they analyzed the problems to a great extent.

With respect to the strategies principals used, it was found that they are similar to strategies used by experts in other
domains. Although no reported study has dealt with this issue in the education domain, this study demonstrated that
principals used four strategies to solve the problems (conversion, decomposition, reversion, and solution oriented).
The first two strategies are similar to the ones that Voss, Green, Post, & Penner, 1983) found in their study. The last
two did not occur in the literature to date and present new findings.

Two points are worth mentioning about the use of strategies as they appeared in the principals’ protocols. First, no
one set of processes is associated with any one of the strategies. Instead, a single process seemed to appear
consistently in the same strategy across problems and participants. For example, in “conversion” the process that
appeared consistently was representing the problem, in “decomposition” the process was stating subproblems, in
“reversion” the process was being proactive, and in “solution oriented” the process was stating solutions. In this last
strategy, this process dominated the protocols almost exclusively. Due to lack of supporting or contradictory evidence
in the literature, these findings must remain tentative for now. Voss et al. (1983) did not link certain processes with
specific strategies. From a cognitive psychology point of view, this presents an opportunity for further investigations.

Second, use of strategies was not consistent across the five problems. This point, along with the fact that problem-
solving strategies cannot be descriptive of a consistent set of problem-solving processes, indicates the nonlinearity of
problem solving. It could be argued that differences in the nature and the difficulty of a problem elicit different
strategies. Further research, however, is needed to confirm this.
 Implications

This study provides an exploration of one aspect of the leaders’ reality. By studying the problem-solving skills of
school leaders, we acquire better insights into what goes on in their minds before they make decisions. Future
studies can focus on the realities of other constituencies and how these realities are negotiated to lead to more
competent and effective organizations. One of the characteristics of the cognitive approaches to the study of
leadership is its assumption that competent leaders are more likely to be found among those who are familiar with the
dynamics of the psyche rather than those who are not (Maslow, 1965; Vaillant, 1977). As Schott (1991) has written,
the better we know ourselves, the better our chances of understanding others.

Problem-solving skills may be an appropriate focus for leader training programs. Prospective leaders should be
exposed to organizational problems and taught processes and strategies to solve them effectively. If pattern
recognition is important for expertise in other fields¾such as medicine and physics¾it should also be important in
expertise for educational leaders.

From the instructional point of view, the cognitive skills that leaders should acquire are:
· Considering all potential sources of information when facing a problem,
· Comparing new information to what is known or inferred,
· Sifting through information to separate the relevant from the irrelevant,
· Identifying gaps in information and possible means for filling in those gaps,
· Considering the possible causal relationship between observed behaviour, inferred information, and other factors,
· Formulating hypotheses to tentatively explain the relationship among those factors,
· Formulating questions to test and evaluate these hypotheses,
· Assessing the nature of tasks that need to be performed,
· Formulating specific goals, and
· Evaluating the results of actions taken and taking this information into account in subsequent planning.
The present study underscores the importance of mental processes in the exercise of effective leadership.
Accordingly, training programs ought to be devised that use case studies, problem-based instruction, and other
instructional techniques that encourage leaders to entertain alternative definitions of a problem situation or to
articulate multiple potential consequences of an event.

References

Allison, D. J. (1996). Problem processing and the principalship: Theoretical foundations and the expertise issue.
ERIC Document Reproduction No. 412 602.

Bereiter, C., & Scardamalia, M. (1986). Educational relevance of the study of expertise. Interchange, 17(1), 10-19.

Berliner, D. C. (1986). In pursuit of the expert pedagogue. Educational Researcher, 15(7), 5-13.

Chi, M. T. H., Glaser, R., & Farr, M. J. (Eds.) 1985. The nature of expertise. Hillsdale, NJ: Erlbaum.

Connelly, M. S., Gilbert, J. A., Zaccaro, S. J., Threlfall, K. V., Marks, M. A., & Mumford, M. D. (2000). Exploring the
relationship of leader skills and knowledge to leader performance. Leadership Quarterly, 11(1), 65-86.

Ericsson, K. A., & Simon, H. A. (1993). Protocol analysis. Cambridge, MA: The MIT Press.

Ericsson, K. A., & Charness, N. (1994). Expert performance: Its structure and acquisition. American Psychologist, 49,
725-747.

Haller, E. J., & Strike, K. A. (1986). An introduction to educational administration: Social, legal, and ethical
perspectives. New York: Longman.

Hart, A. W. (1999). Educational leadership: A field of inquiry and practice. Educational Management and
Administration, 27(3), 323-334.

Lazaridou, A. (2002). An exploratory study of school leaders’ thinking and problem solving through protocol analysis.
Unpublished doctoral dissertation, University of Alberta, Canada.

Leithwood, K. A. (1995). Cognitive perspectives on school leadership. Journal of School Leadership, 5, 115-135.

Leithwood, K. A., Begley, P. T., & Cousins, J. B. (1992). Developing expert leadership for future schools. London, UK:
Falmer.

Leithwood, K. A., & Hallinger, P. (1993). Cognitive perspectives on educational administration: An introduction.
Educational Administration Quarterly, 29(3), 296-301.

Leithwood, K. A., & Stager, M. (1989). Expertise in principals' problem solving. Educational Administration Quarterly,
25, 126-161.

Leithwood, K., & Steinbach, R. (1991). Indicators of transformational leadership in the everyday problem-solving of
school administrators. Journal of Personnel Evaluation in Education, 4, 221-244.

Leithwood, K., & Steinbach, R. (1992). Improving the problem-solving expertise of school administrators. Education
and Urban Society, 24(3), 317-345.

Leithwood, K., & Steinbach, R. (1995). Expert problem solving: Evidence from school and district leaders. New York:
New York University Press.

Leithwood, K., Steinbach, R., & Raun, T. (1993). Superintendents’ group problem-solving processes. Educational
Administration Quarterly, 29(3), 364-391.

Maslow, A. (1965). Eupsychian management. Homewood, IL: Dorsey Press.

Mumford, M. D., & Connelly, M. S. (1991). Leaders as creators: Leader performance and problem solving in ill-
defined domains. Leadership Quarterly, 2, 289?316.

Mumford, M. D., Zaccaro, S. J., Harding, F. D., Jacobs, T. O., Fleishman, E. A. (2000). Leadership skills for a
changing world: Solving complex social problems. Leadership Quarterly, 1(1), 11-36.

Ormord, J. E. (1995). Educational psychology: Principles and applications. Englewood Cliffs, NJ: Merrill.

Reimann, P., & Chi, M. T. H. (1989). Human expertise. In K. J. Gilhooly (Ed.), Human and machine problem solving.
New York: Plenum Press

Schott, R. L. (2001) Administrative and organizational behavior. Some insights from cognitive psychology.
Administration and Society, 23(1), 54-73.

Vaillant, G. (1977). Adaptation to life. Boston: Little, Brown.

Van Lehn, K. (1989). Problem solving and cognitive skill acquisition. In M. I. Posner (Ed.), Foundations of cognitive
science. Cambridge, MA: MIT press.

Voss, J. F., Greene, T. R., Post, T. A., & Penner, B. C. (1983). Problem-solving skill in social sciences. The
Psychology of Learning and Motivation: Advances in Research and Theory, 17, 165-213

Voss, J. F., & Post, T. A. (1988). On the solving of ill-structured problems. In M. T. H. Chi, R. Glaser, & M. J. Farr,
(Eds.), The nature of expertise. Hillsdale, NJ: Lawrence Erlbaum Associates.
Voss, F. J. (1989). Problem solving and the educational process. In A. Lesgold and R. Glaser (Eds.), Foundations for
a psychology of education, (pp. 251-292). Hillsdale, NJ: Lawrence Erlbaum Associates.

Zaccaro, S. J., Mumford, M. D., Connelly, M. S., Marks, M. A., & Gilbert, J. A. (2000). Assessment of leader problem-
solving capabilities. Leadership Quarterly, 11(1), 37-65.

Author Note

Angeliki Lazaridou holds a doctoral degree in Educational Administration and Leadership from the University of
Alberta, Canada, and is currently teaching educational administration at the Higher Institute of Technical and
Vocational Education in Athens, Greece.

Correspondence concerning this article should be addressed to:

Dr. Angeliki Lazaridou
Higher Institute of Technical and Vocational Education
28 Octovriou 36
Kato Pefki 151 21
Athens, Greece
alazarid@teledomenet.gr