What Was I Thinking?

Common traps in administrative decision making of superintendents and principals by Stephen H. Davis

Over the past quarter-century, schools and school districts have become increasingly turbulent, pluralistic and unpredictable places to lead. On any given day, conflicting values, ideas, preferences, interests, needs, demands, problems and solutions descend upon the typical school leader from a shifting array of individuals and groups, each with fluctuating degrees of influence and power.

As a result, many administrative decisions are based on beliefs concerning the likelihood of uncertain events.

Although superintendents and principals work hard to bring order and structure to the arrhythmic nature of organizational life, schools consist of a “mixture of structured and unstructured activities, formal and informal procedures, and controlled and autonomous behaviors,” says E. Mark Hanson, author of Educational Administration and Organizational Behavior. Such challenges, in concert with increased demands for accountability and educational reform, underscore the need for administrators with highly effective decision-making skills and sound judgment.

In school districts, decision makers and their decisions are rarely perfectly objective. More importantly, decisions rarely produce perfect or optimal outcomes. This is because the most important decisions are often highly complex. Information may be incomplete, inaccurate, poorly understood or ambiguous. Organizational and/or decision-maker goals may be unclear or subject to disagreement, and decision-making processes are often uneven.

Moreover, the technology of schooling (that is, what works and how) is very difficult to generalize. What works in one setting or context may not work as well in another. Many variables beyond the control of the decision maker can influence how management decisions are carried out and what effects they might have on school operations and outcomes.

Imperfections in decision making also occur when biases distort the decision maker’s ability to apply objective or logical thinking. Biases can appear in several ways: (a) when the information used by the decision maker is inaccurate, incomplete, one-sided or based on unethical or immoral perspectives; (b) when the decision maker’s preferences or professional ideologies obscure objectivity; (c) when political influences or conflicts cause the decision maker to deviate from normative decision-making approaches or organizational goals; or (d) when the decision maker’s motives are self-serving and not in the best interests of the school or district.

Of course, truncated or flawed analytical or logical thought processes also can diminish the quality of administrative decisions. Given the frequency and intensity of problems and dilemmas that confront school leaders, opportunities for orderly and deliberative thinking can be few and far between, thus increasing the possibility that errors in logic and analysis will contaminate decision making.

Cognitive Traps

Following are 15 cognitive traps common to managers in all types of organizations, but especially schools. They are derived from the large body of literature on both educational and private-sector managerial decision making. It should be noted there are no fail-safe strategies that can help leaders avoid or reduce the damaging effects that such traps so often inflict upon decision-making processes and their outcomes.

However, the following typology and examples drawn from school leadership may help administrators enhance their ability to anticipate, recognize and mitigate the adverse effects that such traps can have on decision effectiveness.

• No. 1: Presumed associations.
Leaders often overestimate the probability that two events will co-occur. For example, some school administrators believe that an increase in the number of minority students will automatically result in an increase in disciplinary infractions, truancy rates and lower test scores. While this may indeed occur in some school settings, such presumed associations can create unreasonable expectations about what minority students will or will not do, thus leading to a perverse and durable halo effect that virtually guarantees an increase in disciplinary infractions, truancy and poor test scores among minority students.

• No. 2: Insensitivity to base rates.
It is not uncommon for leaders to ignore base rates when assessing the likelihood that a particular event will occur. For example, there is an old saying among veteran school administrators that five percent of a school district’s constituency creates 95 percent of a superintendent’s headaches. The emotional intensity of interactions between administrators and disgruntled constituents can belie the fact that the base rate of such interactions is actually quite small. Nevertheless, it is not uncommon for administrators to feel overwhelmed by the perceived frequency of emotionally charged conflicts when in fact they typically represent only a small number of the interpersonal interactions that occur over a given time period.

• No. 3: Insensitivity to sample size.
Leaders often fail to appreciate or understand the role of sample size when assessing the reliability of information. This may cause the leader to make generalizations about a situation, event or group of people based on a handful of observations. I’ve seen this happen in several situations—from the school superintendent who makes premature conclusions about the success of a new reading program on the basis of only a handful of results from one school in the district to the classroom teacher who determines that the new reading program is flawed on the basis of only a small number of student scores on the first few reading tests administered.

• No. 4: Misconceptions of chance.
Leaders often overestimate statistical significance and underestimate the probability of randomness. According to Daniel Kahneman and Amos Tversky, in their article “Intuitive Prediction: Biases and Corrective Procedures” in Judgment Under Uncertainty: Heuristics and Biases: “People often make extreme predictions on the basis of information whose reliability and predictive validity are known to be low.” In school districts, it is not uncommon for administrators to make causal inferences about the relationship between certain student outcomes (like achievement scores on standardized tests) and particular teaching or curricular interventions.

The factors that influence student performance are numerous and highly contextual. However, as a group, school administrators are susceptible to inferring causality without effectively controlling for such extraneous variables as SES, parent education level, peer influence or ethnicity. Moreover, an insufficient sample size diminishes the likelihood of statistical significance and increases the probability that particular outcome measures will be the result of random effects or chance.

One reason why this cognitive trap is so commonplace may be as simple as the fact that most administrators have little or no formal training in statistics or research methods. Of course, wishful thinking can be infectious in an era of increasing public demands for educational quality and accountability.

• No. 5: Regression to the mean.
School leaders frequently fail to recognize that extreme events tend to regress to the mean. That is, over time, big variations in school outcomes like SAT scores, achievement test scores, attendance rates and graduation rates, tend to average out. Severe upward or downward spikes in any such measures of schooling do not necessarily mean an enduring pattern has emerged.

Administrators need to be cautious before attributing too much importance to short-term data when making major decisions about educational programs and processes.

• No. 6: Confirmation trap.
Many leaders seek out or give greater weight to information that confirms long-established or deeply held beliefs. In effect, they see what they want to see and discover what they expect to discover. I supervised an assistant principal once who firmly believed that kids who wore their hair long or wore loose baggy pants and rode skateboards were “bad kids” and should be weeded out (suspended, expelled, or transferred out of the school). His convictions ran so deep that when information came to him that cast a shadow of suspicion on unsuspecting students of this ilk, he would quickly reach the conclusion that the student must be guilty and, as such, dealt with severely.

Understandably, his skewed perceptions and subsequent actions generated widespread complaints from parents and students (and even some teachers), many of whom made their way to the superintendent and members of the school board. Fortunately, he eventually left the profession.

• No. 7: Hindsight.
When a leader overestimates the degree to which he or she would have predicted a correct outcome if given the chance, that leader allows hindsight to obscure the complexities and subtle forces that often complicate decision making. Such thinking can impede the ability of a leader to learn from past experiences and failures and mislead the leader into believing that his or her perceptions, values or beliefs are infallible or incorruptible.

• No. 8: Overconfidence.
It is not unusual for school leaders to overestimate the infallibility of their judgments, the depth of their knowledge or their management skills. Overconfidence may be a manifestation of the arrogance of authority that can infect leaders who assume positions of high status and power. It may also be a product of a charismatic leader’s sense of self-aggrandizement or the result of unbridled or naive optimism yet to be seasoned through the tribulations of organizational leadership.

Overconfidence also can occur when a leader possesses an exaggerated perception of control over the environment or the influence of his or her decisions on the organization.

• No. 9: Incrementalism.
Most school leaders make decisions in highly complex and contentious political environments. Because of this, they rarely seek or achieve optimum solutions to wicked dilemmas. Instead, according to Stanford Professor James March in his book A Primer on Decision Making: How Decisions Happen, they “satisfice.” In their attempt to satisfice, decision makers most often choose alternatives that differ only incrementally from the status quo or they attempt to maximize one aspect of a problem while satisficing on other aspects. This provides them with opportunities to test potential solutions and/or to retreat from a decision trajectory without over-committing themselves or others. However, it also can prevent the leader from taking bold and decisive steps when bold and decisive steps are needed.

• No. 10: Anthropocentric attributions.
Leaders tend to attribute events to the intentional behavior of people and not to chance. For example, students may fail to show up for standardized testing, staff members may get sick, deliveries may be delayed by bad weather and construction costs may exceed budgeted parameters. The list of spontaneous and random mishaps and misfortunes that can bedevil school leaders could go on and on. The problem is that the aftermath of such events often are attributed by superintendents, board members or members of the broader school community to negligence, incompetence or lack of commitment.

Americans are not particularly tolerant of leaders whose organizations struggle or falter (as most do from time to time) and as a society we relish playing the blame game. (The national political theater is a prime example.) Moreover, in complex organizations like schools, blame (and praise) is often narrowly and incorrectly distributed. The reality is that sometimes bad things happen in schools despite the best efforts and intentions of administrators, teachers and staff.

• No. 11: Information imbalance.
It is common for leaders to place a disproportionate amount of attention and weight on information acquired early on in the problem-solving process despite its lack of accuracy, comprehensiveness or even relevance. In schools, this is due partly to the fact that administrators tend to be action-oriented—especially early on in the problem-solving process.

Because they tend to take action as the problem-solving process unfolds, the most recently received information is often perceived as the “best” information. Moreover, it is not uncommon for leaders to make a decision early on in the problem-solving process and work to convince themselves (and others) they are right.

Corollary to the information imbalance phenomenon is the propensity of school leaders to over-estimate the frequency of highly publicized events, and visa versa. For example, in one California high school a conflict involving a Caucasian boy and an African-American boy led to a complaint filed by parents with the U.S. Office of Civil Rights. Although this was a relatively ordinary dispute, the extreme publicity it received in the local press led the superintendent and school board to erroneously conclude the school was a hotbed of racial tension. As a result, the principal lost his job.

• No. 12: Wishful thinking.
All people do this from time to time, but none more so than school leaders. Schools and school districts are incredibly hard places to manage with strategic foresight. The ability to reliably predict the outcomes of educational decision-making can be extremely difficult, if not impossible. From year to year, school district budgets can fluctuate dramatically; collective bargaining and/or community pressures can reshape district policy and rapidly shifting demographic trends can render educational programs and instructional practices obsolete or insufficient to meet the needs of all children.

Several years ago, a charismatic superintendent from a high-profile school district in Northern California devised a highly innovative magnet school plan for improving academic achievement among poor minority students. So eager were the superintendent and the school board to implement the plan that they based several critical resource allocation decisions on their expectations of supplemental funding from both private and public sources. Tragically, the funding sources never materialized and the school district went into state receivership after posting a budget deficit of more than $25 million. The superintendent lost his job but was able to repeat this scurrilous feat in two more school districts before his eventual retirement.

• No. 13: Opportunity risk aversion.
Leaders generally take greater risks in response to threats than to opportunities. As a group, school leaders like to play it safe. They are loath to unnecessarily put themselves or their schools under the critical scrutiny of the general public. Staying out of the limelight and covering one’s backside are popular orientations among school leaders who are understandably interested in reducing political frictions and in sustaining career stability. However, threatening events (declining test scores, upsurge in gang activity, increased dropout rates, etc.) are catalysts for administrative problem-solving activities. Administrators love to solve problems. In fact, most seasoned school leaders have a repertoire of solutions to problems that have yet to occur.

But the fixation on threats and problems obscures the ability to recognize opportunities for organizational growth and diminishes the willingness to disrupt the status quo if not absolutely necessary. The irony is that such reactive behavior stultifies entrepreneurial thinking among school leaders by placing them on the defensive.

A mentor of mine once explained that in today’s hyper-competitive environment, complex organizations like schools do one of two things: They move forward or they fall behind. There is no such thing as the status quo anymore because as soon as an organization believes it has reached the pinnacle of success, some other organization comes along and does it better.

• No. 14: Opinion intransigence.
Setting one’s feet in concrete is not an activity exclusive to mafia hit men. Leaders of complex organizations do this all too often. For school leaders, changing a formed opinion can be especially difficult because of the high frequency and intensity of decision events. For example, the typical principal’s workday consists of few breaks in the workload, a wide variety of fragmented and paradoxical activities and requests for assistance, and numerous disjointed conversations. Work is frequently interrupted by relatively trivial issues and concerns that often require (or result in) rapid mood shifts. The constant flow of problems means that principals must simultaneously attend to numerous tasks, few of which ever capture his or her full attention. As a result, principals must form opinions and make decisions rapidly and often without a great deal of deliberation or forethought.

The sheer volume of decision-making events during the typical workday almost guarantees a certain level of intransigence once an opinion has been formed or a decision made. Moreover, decisions in human service organizations like schools are rarely linear in trajectory. Given the high levels of complexity, risk, ambiguity, uncertainty and disputed values that characterize many decision events in schools, formulaic or normative approaches to problem solving generally don’t work very well. Instead, principals often must rely on their intuitions or gut feelings. Dislodging an intuitive feeling through analytical discourse or argument can be very difficult.

• No. 15: Susceptibility to simple arguments.
Last, but not least, is the propensity for leaders to be swayed by simple arguments or singular, scanty, irrelevant and unreliable data. In school districts, the aforementioned turbulent decision environment undoubtedly contributes to this tendency. School leaders often act in order to think and frequently use heuristic thinking (that is, rules of thumb) to drive decision making.

Ironically, the more expert or experienced a leader is, the more likely he or she is to rely upon heuristic thinking. Although the nature of heuristic thinking is to reduce problem complexity and enhance decision-making efficiency, heuristics don’t always hold up well when there is a high requirement for decision accuracy.

The quality of a heuristic decision (and arguably all decisions) depends largely upon the nature of the learning structures that influenced the leader—that is, the quality of information used by the decision maker in the formulation of particular skills, experiences, or dispositions. Learning structures can be either wicked or kind.

Wicked learning structures are, as it sounds, constructed upon biased, flawed, irrelevant or incomplete information. They also may be constructed upon ethically and morally perverse perspectives or environmental influences. Kind learning structures, on the other hand, are based on accurate, complete, reliable and relevant information and ethically and morally acceptable perspectives and environmental influences.

Predictably, heuristic decisions based upon wicked learning structures are most likely to turn out poorly. Take, for example, the principal who, at the start of each school year routinely groups minority students into peer support teams in order to facilitate feelings of cultural identity and the implementation of targeted literacy development strategies. The principal’s intentions may be honorable, but the grouping strategy is based on the simplistic (and faulty) assumption that all minority students are in need of social adjustment and specialized literacy training.

Understanding Dynamics

The biases and cognitive traps that impede good decision making are numerous and inevitable. From time-to-time all leaders commit them and they most often occur without conscious effort or awareness. The latter point is the crux of the issue. School leaders generally have precious little time to think about decision biases, cognitive traps or decision mechanics. As stated earlier, they often act in order to think.

As a result, school leaders are especially prone toward using heuristic or truncated thinking processes when making complex decisions. This is because the frequency, volume, intensity, complexity and novelty of important decision events in schools often defeats deliberative or reflective thinking.

Heuristic thinking is not always a bad thing. In fact without it administrative decision making in schools would come to a grinding halt. However, it is particularly important that school leaders understand the types of biases and cognitive traps (heuristic and otherwise) that can impede good decision making and use this knowledge to guide their retroactive and prospective reflections.

The attributes of reflective practice in educational administration are well understood by most practitioners. However, to engender enduring changes in leadership behavior, reflections must be honed against comprehensive mental models of decision making that include the cognitive barriers and constraints that can reduce decision effectiveness in schools.

Simply put, self-awareness and a fulsome understanding of the dynamics of decision making are precursors to the development of new and more effective decision-making behaviors. The conceptual framework provided here is one important tool that can help school leaders interpret the quality of their decisions, diagnose the root causes of faulty decision making and develop better decision-making processes.

Stephen Davis is associate professor of education at Stanford University, CERAS 429, Stanford, CA 94305. E-mail: shdavis@stanford.edu


Additional Resources

Stephen Davis suggests the following books and articles related to his article:

How Can I Fix It? Finding Solutions and Managing Dilemmas: An Educator’s Road Ma p by Larry Cuban. Teachers College Press, New York, N.Y.

The Intuitive Dimensions of Administrative Decision Making by Stephen H. Davis and Patricia B. Davis. Rowman and Littlefield, Lanham, Md.

A Primer on Decision Making: How Decisions Happen by James G. March. The Free Press, New York, N.Y.

The Psychology of Decision Making: People in Organizations by Lee Roy Beach. Sage Publications, Thousand Oaks, Calif.

“Decision Making and Problem Solving” by Herbert A. Simon, in Research Briefings 1986: Report of the Research Briefing Panel on Decision Making and Problem Solving. National Academy Press, Washington, D.C.

“Decision Making Under Ignorance: Arguing with Yourself” by Robin M. Hogarth and Howard Kunreuther in Research on Judgment and Decision Making: Currents, Connections and Controversies by William. M. Goldstein and Robin. M. Hogarth. Cambridge University Press, Cambridge, United Kingdom.