Management Information System – Unit II – Based on BSc IT Calicut University

UNIT II (12T+3L)

Conceptual Foundations: Phases in decision Making Process, Concepts of decision Making,

Behavioural models of the decision Maker, Behavioural Model of organizational decision making,
decision making under psychological stress, Methods for decision among alternatives, relevance of
decision making concepts for information system design, Definition of information, quality of
information in decision making, value of information other than in decision, General model of the
human as an information processor, The Newwell-Simon model, tentative limits on human
information processing, Concepts of human cognition & learning, Characteristics of human
information processing performance.

PHASES IN THE DECISION-MAKING PROCESS

The well-known model proposed by Herbert A. Simon 1 will be used as the basis for describing the
decision-making process. The model consists of three major phases:

The decision-making process has three main phases:

 Intelligence: Spotting a problem or opportunity by looking at information and identifying issues.

 Design: Coming up with and evaluating different solutions or actions.
 Choice: Picking the best option and putting it into action.
The decision-making process (Intelligence, Design, Choice) isn't strictly linear; you can loop back to
previous phases if needed. For example, if no good options are found during the "Choice" phase, you
can return to "Design" to create new ones.

The "Intelligence" phase is all about identifying problems or opportunities by constantly scanning
your environment, like an air traffic controller watching for issues or a marketing executive looking
for customer needs. This phase essentially tells you when something needs to change or has the
potential to improve.

Simon's model does not go beyond the choice phase. Some models of decision-making include
implementation and feedback from the results of the decision. For example, Rubenstein and
Haberstroh proposed the following steps:

1. Recognition of problem or need for decision

2. Analysis and statement of alternatives
3. Choice among the alternatives
4. Communication and implementation of decision

1
 Management Information System – Unit II – Based on BSc IT Calicut University

5. Follow-up and feedback of results of decisions

INTELLIGENCE AND DESIGN PHASES

Three important aspects of the intelligence and design phases are problem finding, problem
formulation, and development of alternatives.

Problem Finding

Problem finding is all about noticing a gap between what's happening now and what you want to
happen. You have an ideal in mind (your "model"), compare it to reality, and if there's a significant
difference, that's your problem.

According to Pounds, we use four types of models to set our expectations:

 Historical: What's happened before.

 Planning: What we've planned to do.
 Other People: What others in the organization expect.
 External: What competitors, customers, or professional groups expect.
Essentially, you need a mental model of how things should be to even spot a problem. If your model
improves, so does your ability to find problems.

Problem Formulation

It's crucial to properly formulate a problem to avoid solving the wrong one. This means clarifying the
issue so your solutions are effective.

Often, simply stating the problem clearly is enough. Other times, you need to simplify it using these
strategies:

 Define boundaries: Specify what's included in the problem.

 Examine changes: Look for what might have caused the problem.
 Break it down: Divide the problem into smaller parts.
 Focus on controllable elements: Concentrate on aspects you can actually influence.

Drawing analogies to similar, previously solved problems can also greatly simplify the issue,
showing the value of past experience.

2
 Management Information System – Unit II – Based on BSc IT Calicut University

Developing Solutions

A key part of decision-making is coming up with different solutions or options. This is a creative
process that can be learned and improved with tools like brainstorming, using checklists, or even
thinking of similar past situations (analogies).

To generate good alternatives, you need a good understanding of the problem (domain knowledge)
and the motivation to solve it.

CONCEPTS OF DECISION MAKING

Decisions vary, and these differences influence how we create options, choose among them, and
design systems to support these choices. Key ways decisions differ include:

 How well we know the outcomes: Are the results predictable or uncertain?
 How programmable they are: Can they be automated or do they require human judgment?
 The criteria used: What factors determine a "good" decision?
 Their impact: How significant are the consequences of the decision?

Knowledge of Outcomes

When making decisions, understanding the knowledge of outcomes is crucial, especially when you
have multiple options. There are three main ways we categorize our knowledge about what might
happen:

1. Certainty: This is when you have complete and accurate knowledge of what will happen for
each choice you make. There's only one predictable result for every alternative. In this ideal
scenario, your task is to simply pick the option that gives you the best (or "optimal") outcome,
often based on criteria like least cost. However, real-world situations rarely offer such perfect
certainty.

2. Risk: In this situation, you can identify multiple possible outcomes for each choice, and,
importantly, you can assign a probability to each of those outcomes occurring. When operating
under risk, the goal isn't just to pick the best single outcome, but to optimize the expected
outcome. This involves calculating the "expected value" (probability multiplied by the outcome)
for each alternative and choosing the one with the highest expected value. This assumes a
rational decision-maker.

3. Uncertainty: This is the trickiest scenario. Here, you can identify multiple possible outcomes
for each alternative, but you have no knowledge of the probability of each outcome occurring.

3
 Management Information System – Unit II – Based on BSc IT Calicut University

Since you can't assign probabilities, you can't calculate an expected value. This requires different
decision-making approaches, as standard optimization criteria don't apply. Often, strategies are
employed to either try and estimate probabilities or use rules designed to minimize potential
negative impacts or maximize potential gains without knowing the odds.

Programmed versus Nonprogrammed Decisions

Decisions can be categorized by how much they can be pre-planned:

 Programmed Decisions: These are routine decisions with clear, pre-defined rules or procedures,
like those found in rule books or regulations. They imply certainty because all outcomes are
known, and they can often be delegated to lower levels or even automated.
 Nonprogrammed Decisions: These are unique, often complex decisions without established rules.
They can range from crisis management to recurring problems with constantly changing
conditions. They typically cannot be delegated or automated and require higher-level judgment.

A strategy is to handle normal situations with programmed rules and escalate unusual cases as
nonprogrammed decisions. However, relying too much on programmed methods can lead to rigid
and potentially inappropriate solutions.

Decision-Making Criteria

There are two main ways to think about how decisions are made:

 Normative (Prescriptive) Models: These models tell you how you should make decisions to
achieve the best possible outcome. They are built on the idea of maximization or optimization
(e.g., maximizing profit, minimizing cost), assuming a completely rational decision-maker.
Examples include linear programming.
 Descriptive Models: These models explain how people actually make decisions. They
acknowledge that decision-makers aren't perfectly rational or fully informed. Instead, people
often "satisfice," meaning they choose the first alternative that is "good enough" and meets their
requirements, rather than searching endlessly for the absolute best (optimal) solution. This
concept is based on "bounded rationality," recognizing that our ability to process information and
find all options is limited.

4
 Management Information System – Unit II – Based on BSc IT Calicut University

BEHAVIORAL MODELS OF THE DECISION MAKER

The way a person examines a problem and makes a decision can be described from several different
viewpoints, depending on the assumptions made. Several models of the decision maker are described
below.

Classical Economic Model of Decision Maker

 The Classical Economic Model describes an ideal, "rational" decision-maker. It assumes:

 Complete knowledge: All options and their exact outcomes are known (or at least their
probabilities).
 Maximizing profit/utility: The goal is always to get the best possible outcome.
 Perfect sensitivity: Even tiny differences in outcomes are noticed and acted upon.

This model is prescriptive, showing how decisions should be made to be perfectly optimal. While
rarely met in reality, it's a useful benchmark, and many decision-making tools are based on its
assumptions.

Administrative Model of Decision Maker

The Administrative Model describes how people actually make decisions, rather than how they
ideally should. It acknowledges that decision-makers operate with bounded rationality, meaning
their ability to be perfectly rational is limited by factors like their knowledge, perception, and
capacity.

Key assumptions of this model are:

1. Incomplete knowledge: Decision-makers don't know all alternatives or outcomes.

2. Limited search: They conduct a restricted search for solutions.
3. Satisficing: They choose the first acceptable option that meets their goals, rather than seeking
the absolute best.
This model suggests people often use "rules of thumb" (heuristics) to make decisions, highlighting
the need for decision support systems that accommodate these natural, less-than-perfect approaches.

Human Expectations and Decision Making

Our expectations significantly shape how we make and react to decisions. Key psychological
theories explain these responses:

5
 Management Information System – Unit II – Based on BSc IT Calicut University

1. Cognitive Dissonance: After making a choice, we often feel uncomfortable (dissonance)

because our chosen option has flaws and the rejected ones had good points. To reduce this
discomfort, we'll often convince ourselves our choice was better, ignoring negative
information about it and seeking out positive reinforcement.

2. Commitment Theory: When we commit to a decision (especially if it's irreversible), we tend

to take more time and be more careful. Once a decision is made and announced, we're
reluctant to change it, and even small commitments can lead to bigger ones.

3. Anticipatory Regret: Before making a decision, we often imagine how much we might
regret it if things go wrong. This "anticipatory regret" can make us more cautious and
encourage us to think through consequences beforehand, which can also soften the blow if
things don't go as planned.

BEHAVIORAL MODEL OF ORGANIZATIONAL DECISION MAKING

The Behavioral Theory of the Firm (developed by Cyert and March, building on Simon's ideas)
explains how decisions are actually made within organizations. Starting with the concept of
"bounded rationality" from the administrative model, it uses several key ideas to describe
organizational decision-making:

 Quasi-resolution of conflict: Organizations handle internal disagreements by localizing them or

dealing with them sequentially, rather than fully resolving them.
 Uncertainty avoidance: Organizations try to avoid uncertainty by focusing on short-term
problems and using feedback to make adjustments rather than trying to predict the distant future.
 Problemistic search: When a problem arises, organizations search for solutions close to the
source of the problem and often use simple, familiar solutions first.
 Organizational learning: Organizations adapt their behavior over time based on experience,
changing their rules and procedures.
 Incremental decision making: Decisions are often made in small, sequential steps, making
minor adjustments rather than radical changes.
Quasi-resolution of Conflict

Organizations are made up of groups with different, often conflicting, goals (e.g., sales wants high
inventory, inventory control wants low). Rather than fully resolving these conflicts, organizations use
"quasi-resolution" through three main methods:

1. Local Rationality: Different departments or subunits are allowed to set their own goals and
make their own decisions within certain agreed-upon limits.

6
 Management Information System – Unit II – Based on BSc IT Calicut University

2. Acceptable-level Decision Rules: Units follow specific rules or procedures to make decisions
that are "good enough" for their local goals.

3. Sequential Attention to Goals: The organization addresses one goal at a time, then another. This
means conflicting goals are never tackled simultaneously, and some conflicts may never truly be
resolved.

Uncertainty Avoidance

Organizations operate in unpredictable environments. The behavioral theory of organizational

decision-making suggests that organizations prioritize avoiding uncertainty even if it means
accepting a lower "expected value." People and organizations often prefer a more certain, slightly
lower gain over a less certain, higher potential gain.

Organizations reduce uncertainty through:

1. Short-run feedback and reaction cycles: Making frequent small adjustments based on
immediate feedback, which lessens the need to predict far into the future.
2. Negotiated environment: Trying to control their external environment through industry
agreements or long-term contracts.
Problemistic Search

Organizations typically search for solutions only when a problem arises, not proactively. This search
follows simple rules:

1. Local First: They look for solutions close to the problem's symptom or the current solution
(e.g., if sales drop, they start with the sales department).
2. Expand to Vulnerable Areas: If local search fails, they then look to areas with "slack"
resources (like overstaffed departments) or hard-to-measure goals (like research).
Organizational Learning

Organizations learn and adapt over time, changing their goals and problem-solving methods based on
past experiences. Their ambition levels (aspiration levels) adjust: they stay slightly above current
performance in stable times, lag behind during growth, and fall slower than performance during
decline.

This learning impacts planning (which reflects aspirations) and control (where reports influence
performance based on aspiration levels). Information systems play a role in balancing what's
achieved with what's aspired to.

Incremental Decision Making

7
 Management Information System – Unit II – Based on BSc IT Calicut University

Incremental decision-making is a "satisficing" approach where organizations make decisions through

small, gradual changes to existing policies. The focus is on minor corrections and improvements, not
radical shifts. This is often called "muddling through," especially common in governments and large
organizations where consensus is important and changes are small.

While good for minor adjustments, it's not effective for big policy shifts. Etzioni suggests a "mixed
strategy": using incremental steps for small decisions, but a full review of alternatives for major
policy changes.

DECISION MAKING UNDER PSYCHOLOGICAL STRESS

Traditional decision-making models assume a calm, logical process. However, many real-life
decisions, both personal and organizational, happen under psychological stress due to strong
emotions, conflicting desires, or fear of negative consequences.

This decisional conflict can severely impair decision-making. Information systems designed to
support decisions in such high-stress situations need to account for this altered behavior, as described
by the conflict-theory model of Janis and Mann, to be truly effective.

Decisional Conflict and Psychological Stress

Decision-making under significant psychological stress often arises when all available choices seem
to have serious negative outcomes. This creates decisional conflict, leading to feelings of
apprehension, hesitation, and distress. Even under such stress, decisions are still made, but often
through various coping mechanisms. This highlights the need for decision support systems to be
designed to account for these psychological factors to be truly effective.

Coping Patterns in Stressful Decisions

When facing a difficult decision with potentially serious negative outcomes, people use coping
patterns. These patterns are determined by four questions:

1. Are there serious risks if nothing changes?

2. Are there serious risks if a change is made?
3. Is a better solution realistic?
4. Is there enough time to decide?
If risks are serious and time is short (e.g., a fire), the response might be hypervigilance – acting
impulsively without fully thinking through all consequences. If risks are serious but a better solution
seems possible with time, the coping is more vigilant (search, appraisal, planning).

8
 Management Information System – Unit II – Based on BSc IT Calicut University

Hypervigilance is an urgent, often panicked, response to disaster where a person focuses only on
negative outcomes and rushes a decision without thorough consideration, ignoring information that
contradicts their preferred, immediate action.

Defensive Avoidance

Defensive avoidance is a coping strategy where decision-makers avoid making a difficult decision,
hoping the problem will go away. They might ignore disturbing information, engage in wishful
thinking, or procrastinate.

If they can't avoid the decision, they might use "bolstering tactics" to reduce mental discomfort
(cognitive dissonance) after choosing. These tactics include:

 Exaggerating the positives of their choice.

 Downplaying or denying the negatives.
 Assuming the problem is less severe or someone else's responsibility.
In groups, this can lead to "groupthink," where members avoid real debate to maintain harmony,
often resulting in poor decisions. Symptoms of groupthink include an illusion of invulnerability,
rationalizing bad decisions, believing in the group's inherent morality, stereotyping outsiders,
pressuring dissenters, self-censorship, an illusion of unanimity, and self-appointed "mindguards" who
protect the group from conflicting information.

9
 Management Information System – Unit II – Based on BSc IT Calicut University

METHODS FOR DECIDING AMONG ALTERNATIVES

Many techniques exist to help decide between options. These methods usually assume all choices are
known, even if in reality, people often stop searching too early. This section briefly covers various
methods like:

 Optimization techniques  Decisional balance  Game theory

 Payoff matrices sheets  Statistical inference
 Decision trees  Elimination by aspects

It also touches on methods for situations where probabilities are hard to guess, providing a general
idea of available tools without going into deep detail.

Optimization Techniques Under Certainty

These techniques are used when you know exactly what will happen for every choice you make. The
goal is to mathematically find the single best option (the "optimal" one) for a specific goal (like
maximizing profit or minimizing cost).

Examples of these techniques include:

 Systems of equations  Dynamic programming  Capital budgeting

 Linear programming  Queuing models analysis
 Integer programming  Inventory models  Break-even analysis

Payoff Matrices in Statistical Decision Theory

Payoff matrices are tools used in statistical decision theory to help evaluate options. They organize
decisions by listing available strategies (your choices) in rows and different "states of nature"
(conditions that affect outcomes, like "new competitor" or "highway rerouting") in columns. Each
cell in the matrix shows the "payoff" (the financial result or consequence) if a specific strategy is
chosen and a particular state of nature occurs.

If you are certain which "state of nature" will happen, you simply pick the strategy that gives the
highest payoff in that specific condition.

However, in situations of risk (where you know the probabilities of each "state of nature"), you
calculate the "expected value" for each strategy by multiplying each outcome by its probability and
summing them up. A rational decision-maker would then choose the strategy with the highest
expected value.

10
 Management Information System – Unit II – Based on BSc IT Calicut University

When facing uncertainty (outcomes known, but probabilities unknown), you can't use expected
value. Instead, different rules apply:

 Minimize regret: Choose the strategy that minimizes your potential "regret" (the difference
between what you got and what you could have gotten).
 Maximin rule: A pessimistic approach; choose the strategy that maximizes the worst possible
outcome for that strategy.
 Maximax rule: An optimistic approach; choose the strategy that maximizes the best possible
outcome for that strategy.
These rules offer ways to make choices when probabilities are unclear, but each has its own
advantages and disadvantages.

Decision Trees

Decision trees are visual tools that help analyze a sequence of decisions and their possible outcomes.
They start from an initial decision and branch out, showing subsequent decisions and external events
("states of nature") with their probabilities.

How to use them:

1. Build the tree: Start with your main decision, then add branches for each alternative and for
possible events (like government actions or market conditions). Assign a probability to each
event.
2. Assign values: Put a numerical value (e.g., profit or loss) at the end of each final branch.
3. Work backward: Calculate the "expected value" at each branching point (node) by
multiplying outcomes by their probabilities. At decision points, choose the path with the
highest expected value.
This process helps determine the best overall strategy by systematically evaluating all possibilities.
While they use probabilities, these are often subjective estimates, but the tree forces you to make
those estimates explicit.

Ranking, Weighting, or Elimination by Aspects

In decisions, each factor is given a relative importance (weight). Options are scored on how well they
meet each factor, and this is multiplied by the factor's weight. The total weighted scores rank the
options. For instance, in computer selection, if software is weighted at 40 points, a vendor meeting
90% of requirements gets 36 points (0.90×40), and one meeting 70% gets 28 points (0.70×40).

Alternatively, Tversky describes "elimination by aspects." Requirements are ranked by importance.

Starting with the most important, any options not meeting that aspect are eliminated. This continues

11
 Management Information System – Unit II – Based on BSc IT Calicut University

for all aspects. For example, if IBM compatibility is crucial for a microcomputer, non-compatible
alternatives are immediately removed.

Game Theory

Game theory helps analyze competitive situations where one person's gain is another's loss. In "zero-
sum" games, what one player wins, the other loses exactly. While good for understanding conflicts
and bargaining, it's not very useful for everyday organizational decisions.

Classical Statistical Inference

This is a way of using numbers to make decisions, focusing on "objective" data rather than personal
opinions. It's great for preparing information for choices.

Here are some common tools used:

 Sampling: Taking a small group from a larger one to learn about the whole group (e.g., finding
the average height of students by measuring a few).
 Probability Distributions: Using mathematical models (like "normal" or "Poisson" curves) to
understand how data is spread out, which helps predict outcomes. The "Chi-square test" checks if
your data fits a model well.
 Regression & Correlation Analysis: Finding relationships between different factors.
o Correlation: Shows if two things move together (e.g., as study time increases, grades
increase).
o Regression: Helps predict one factor based on another.
 Testing Hypotheses: Checking if a statement or idea (a "hypothesis") is likely true or false based
on data.
Decisional Balance Sheet for Decision Making under Stress

Janis and Mann suggest a tool called the "Decisional Balance Sheet" to help make choices, especially
when under stress.

How it works:

For each possible choice you have, think about the positive and negative outcomes (gains and
losses) it might bring, specifically in these four areas:

1. For Yourself: What you personally gain or lose (e.g., money, time, effort).

2. For Others: How it affects people around you (e.g., family, friends, colleagues).

3. Self-Approval: How you will feel about yourself (proud, guilty, satisfied).

12
 Management Information System – Unit II – Based on BSc IT Calicut University

4. Social Approval: How others will see you (respected, criticized, accepted).

After listing these, you can weigh how strong each gain or loss is to help you decide.

Beyond the Balance Sheet:

Janis and Mann also suggest other techniques to help decision-making and avoid common pitfalls:

 Decision Counseling: Getting advice to better evaluate your options.

 Role Playing (Psychodrama): Practicing scenarios to understand what might happen.

 Emotional Inoculation: Preparing yourself emotionally for potential setbacks after the decision.

 Avoiding Groupthink: Using methods to prevent groups from making bad decisions by
conforming to each other's opinions without critical thinking.

RELEVANCE OF DECISION-MAKING CONCEPTS FOR INFORMATION SYSTEM

DESIGN

It's crucial for both system designers and users to understand decision-making concepts. Some of
these concepts should be built directly into systems that help with decisions, while others help define
what such a system can realistically achieve. These ideas are important for designing effective
decision support systems.

Support for Decision-Making Phases

The Simon model breaks down decision-making into three phases, and information systems can
support each one:

1. Intelligence (Finding the Problem):

o Systems should help scan and analyze data to spot issues or opportunities.

o They should flag unusual situations for humans to investigate.

o They also need to have ways to communicate these findings effectively to the right people.

2. Design (Finding Solutions):

o Systems should include decision models to process information and suggest different
solutions.

o They should offer tools like checklists or scenarios to help analyze these options.

13
 Management Information System – Unit II – Based on BSc IT Calicut University

3. Choice (Picking a Solution):

o Systems are most effective when they present information clearly to help make the final
decision.

o After a choice is made, the system's role shifts to collecting feedback and assessing the
outcome.

Support for Programmed versus Nonprogrammed Decisions

Decisions can be:

1. Programmed (Structured): These are routine, repetitive decisions with clear rules.

o Example: When to reorder office supplies.

o Challenge: The main issue is ensuring these rules are applied correctly and don't get used in
situations they weren't designed for (e.g., using a "fast-moving" inventory rule for slow items).
Systems should allow human oversight here.
2. Non-programmed (Unstructured): These are new, complex, and unique decisions without
predefined rules.

o Example: Deciding to launch a completely new product.

o Support Systems: Decision Support Systems (DSS) are specifically built to help with these
complex, one-off decisions.
Both types of decisions still involve intelligence (finding the problem), design (finding solutions),
and choice (picking one).

Relevance of Models of the Decision Maker

Computers can optimize decisions, but only if they have complete information and the decision is
purely logical. This rarely happens in real life. Trying to force optimization with incomplete info,
political issues, or human biases often makes computerized decision models fail.

A better approach is to assume information is incomplete and costly, that we often "satisfice" (find
a good enough solution) instead of perfectly optimize, and that human psychology plays a role.

In this "decision support" approach, the computer helps the human decision-maker by handling data
(computing, storing, analyzing). The human can decide what tasks the computer does, making the
process more flexible and realistic.

Application of the Behavioral Model of Organizational Decision Making

14
 Management Information System – Unit II – Based on BSc IT Calicut University

The behavioral model of decision-making reminds system designers that real-world decision-makers
face organizational limits. While designers might aim for perfect logic, decision-makers often
prioritize avoiding uncertainty.

This model helps design information systems that:

 Support methods for reducing uncertainty.

 Recognize that organizational goals are often inconsistent (not perfectly rational).

 Aid managers in "local searches" for solutions and help them see when exploring more options is
beneficial.

 Are flexible enough for changing goals and support continuous learning.

 Can prevent managers from making only small, incremental decisions when a deeper analysis is
needed.

Support for Decision Making under Stress

Information systems can help with stressful decisions by:

 Having pre-programmed responses ready for specific crises.

 Allowing interactive exploration of options with the decision-maker.

 Using tools like the decisional balance sheet to quickly organize thoughts and options.

Support for Alternative Techniques

Decision support systems should offer various ways to help make decisions. Sometimes the best
technique can be chosen automatically, while other times, several options might be provided. The
system's goal is to help decision-makers find the best approach.

When dealing with probabilities or estimates, it's helpful for the user to easily change values and
instantly see how those changes affect the results. For example, if estimating new product sales, the
system should let the user quickly adjust sales forecasts and observe the impact, ensuring consistency
in the final estimates.

Support for Quality of Decision Making

It's hard to judge if a decision itself is good, but we can judge if the process used to make it was
good. Janis and Mann identified seven criteria for a quality decision-making process:

15
 Management Information System – Unit II – Based on BSc IT Calicut University

A good decision-maker (using their best ability and information):

1. Explores many options.

2. Considers all goals and values.

3. Weighs all costs, risks, and benefits.

4. Actively seeks new, relevant information.

5. Uses new information correctly, even if it contradicts initial preferences.

6. Re-examines all options (even disliked ones) before deciding.

7. Plans thoroughly for implementation, including backup plans for risks.

Information systems can help reinforce these steps, prompting users to follow a high-quality decision
process.

DEFINITION OF INFORMATION

Data is raw, unprocessed facts (like raw materials).

Information is data that has been processed into a meaningful form for the recipient, providing real
or perceived value for current or future decisions and actions.

 Information reduces uncertainty and can change the likelihood of outcomes in decisions.
 The value of information is tied to its use in decision-making. If it doesn't affect a decision, it has
no value.
 Data processed for one level of an organization can become raw data for another level.
 Information resources (stored data) are reusable and gain value through use.

16
 Management Information System – Unit II – Based on BSc IT Calicut University

 A useful general definition of information for the purpose of information systems is the
following: Information is data that has been processed into a form that is meaningful to the
recipient and is of real or perceived value in current or prospective actions or decisions.
 An information processing system processes data into information. More precisely, the
information system processes data in an unusable form into a usable form that is information to
the intended recipient.
 Data, the raw material for information, is defined as groups of nonrandom symbols which
represent quantities, actions, objects, etc.

QUALITY OF INFORMATION

Even clear information might not be used effectively. Information quality is judged by how well it
leads to action and good decisions.

The true value of information is how much it changes decision-making for the better. We'll explore
this further, but first, we'll look at how decision-makers feel about information's quality, considering
its usefulness, satisfaction, and any errors or biases.

Utility of Information

Andrus says information's value (utility) affects how much it's used, beyond just its accuracy. He
identifies four ways information is useful:

1. Form Utility: Information is more valuable if it's presented in a way that perfectly suits the
decision-maker.
2. Time Utility: Information is more valuable when it's available exactly when needed.
3. Place Utility: Information is more valuable if it's easy to access and get (e.g., online).
4. Possession Utility: The value of information is also affected by who controls it and how they
share it.
Cost vs. Value: Information costs money to get. If the cost of getting information is more than its
value, you can either:

1. Increase its value (make it more accurate or useful).

2. Reduce its cost (by accepting less accuracy or utility).

Information Satisfaction

17
 Management Information System – Unit II – Based on BSc IT Calicut University

Since it's hard to directly measure how much a piece of information improves decisions, we often use
"information satisfaction" instead. This means how happy the decision-maker is with the
information they get from a formal system.

How it works:

 If a decision-maker gets the info they need easily from the system, their satisfaction with it goes
up.

 If they can't find it easily and have to search elsewhere, their dissatisfaction with the system
increases.

This satisfaction can be used to judge how good an information system is.

Errors and Bias

Managers prefer information quality over quantity. Quality suffers from bias and errors. Bias
comes from how information is presented; detecting it is the main challenge, as knowing it makes
adjustments simple.

Error is a more serious problem because there is no simple adjustment for it. Errors may be a result
of:

1. Incorrect data measurement and collection methods

2. Failure to follow correct processing procedures
3. Loss or nonprocessing of data
4. Wrong recording or correcting of data
5. Incorrect history (master) file (or use of wrong history file)
6. Mistakes in processing procedure (such as computer program errors)
7. Deliberate falsification

Often, when you receive information from a system, you don't know if it's biased or contains errors.
The precise-looking numbers in reports can give a false sense of accuracy.

Example: An inventory report might say "347 widgets," but this is likely based on ongoing records
that can accumulate small or even large errors from various mistakes (like wrong entries, receipts,
etc.). This is why businesses regularly do physical counts to correct their records.

The difficulties with errors may be overcome by:

1. Internal controls to detect errors

18
 Management Information System – Unit II – Based on BSc IT Calicut University

2. Internal and external auditing

3. Addition of "confidence limits" to data
4. User instruction in measurement and processing procedures, so users can evaluate possible errors

There are two main ways to deal with information errors:

 Reducing Uncertainty (First two remedies: Internal controls, Auditing): These methods directly
try to fix or prevent errors, making the information itself more reliable and increasing its content
value.

 Showing Uncertainty (Last two remedies: Confidence limits, User training): These methods don't
remove the errors, but they help the user understand the likelihood of errors or the possible range
of values, increasing their confidence in using the information despite potential inaccuracies.

Both approaches ultimately add value by making the information more trustworthy.

VALUE OF INFORMATION OTHER THAN IN A DECISION

If the value of information were based only on identified decisions, much of the data that
organizations and individuals prepare would not have value. Since the market for information
suggests that it does have value, there must be other bases for the value of information. Some other
reasons for the value of information are motivation, model building, and background building.

Motivation

Some information acts as feedback, showing people how they are performing. This kind of
information can indirectly encourage certain decisions by:

 Confirming existing beliefs or models.

 Reassuring them that things are going well.
 Helping individuals learn from the results of their actions.

Model Building

Managers and staff operate using mental "models" of their organization (how things work).

 Information they receive can either change or strengthen these mental models.
 This is how organizations learn and build expertise.
 Since these models help identify problems, new or changed information can directly affect how
and what problems are recognized.
19
 Management Information System – Unit II – Based on BSc IT Calicut University

Background Building

Information is most valuable when you have the background knowledge to use it effectively.
Experienced decision-makers often need less new information and can make better or cheaper
decisions because their past knowledge already reduces uncertainty.

It's hard to separate the value of specific information from a decision-maker's overall knowledge.
Much of what we learn isn't for one specific problem but builds a rich "background" of expertise.
This accumulated knowledge allows us to make better decisions or solve problems more efficiently.

This highlights the importance of:

 Broad education: To build a strong base of knowledge.

 Continuous learning: To enhance our ability to find problems, formulate solutions, and draw on
past experiences.

GENERAL MODEL OF THE HUMAN AS AN INFORMATION PROCESSOR

Our brains are like computers:

 How we process: Our senses (eyes, ears, etc.) take in information, send it to our brain for
thinking, and then we create responses (like acting or speaking).

 Limited Capacity: Just like a computer, our brain can only handle a certain amount of
information at once.

 Information Overload: If we get too much information, our ability to respond slows down, and
we don't perform as well. Even though we use multiple senses together, there's still a point where
we get overloaded.

20
 Management Information System – Unit II – Based on BSc IT Calicut University

Our brains receive more information than we can handle. To prevent overload, we use a "filter" to
block out some input. This filter is shaped by:

 Our personal frame of reference (what we already know).

 Normal decision procedures.
 Stress (which changes what we focus on).

We use "filters" to manage the information we receive, based on:

 Our experience and background.

 Our usual decision-making methods.
 Stress: When under pressure (e.g., a crisis), we filter more heavily, focusing only on the most
important details and ignoring less critical information to reduce overload.

21
 Management Information System – Unit II – Based on BSc IT Calicut University

Frame of Reference: How Our Brains Process

Our brains use existing "frames of reference" (mental patterns or categories built from experience) to
understand new information. This helps us:

 Process efficiently: We don't have to learn everything from scratch.

 Handle more input: By using established patterns, we can process more data.
 Develop Expertise: A strong, relevant frame of reference is a key part of being an expert.
The Downside: Our filters (based on these frames of reference) can sometimes:

 Block relevant data: Especially if it doesn't fit our existing views.

 Cause errors: Leading to us missing, distorting, or misinterpreting information.
This means what one person intends to communicate can be understood differently by another,
increasing uncertainty and reducing the message's true meaning.

THE NEWELL-SIMON MODEL

22
 Management Information System – Unit II – Based on BSc IT Calicut University

Newell and Simon suggested we can understand human problem-solving by comparing it to how
computers process information. This isn't to say humans are computers, but the comparison helps us
understand how our brains handle information.

The Human Information Processing System.

The human information processing system consists of a processor, sensory input, motor output, and
three different memories: long-term memory (LTM), short-term memory (STM), and external
memory (EM). The system operates in serial fashion rather than in parallel. This means that the
human can perform only one information processing task at a time, whereas a computer may operate
in either serial or parallel design. A good example of computer parallel processing is the
simultaneous addition of all pairs of bits in two computer data words. Normally, humans add serially
a pair of digits at a time from right to left. Three processing operations to be described in the
following list are also illustrated in Figure 8-6.

Humans are mostly "serial processors" (doing one main task at a time), but we can multitask by
quickly switching between tasks. We also use "pattern matching," which is unique.

Our system has three types of memory:

1. Long-Term Memory (LTM):

o Capacity: Essentially unlimited.

o Content: Stores information in "chunks" (like words, images).
o Speed: Quick to recall (read) but slow to store (write/memorize).
2. Short-Term Memory (STM):

o Capacity: Very small, holding only 5-7 items.

o Speed: Very fast to read and write, used for immediate tasks (like remembering a
phone number just long enough to dial).
3. External Memory (EM):

23
 Management Information System – Unit II – Based on BSc IT Calicut University

o Examples: Paper, whiteboards – anything outside our heads.

o Speed: Fast to read, relatively fast to write, making it efficient for problem-solving
and easing the burden on STM.

The processor of the information processing system contains three parts: the elementary processor,
the short-term memory, and the interpreter, which interprets part or all of the program of instructions
for problem solving (Figure 8-7). The program used by an individual will depend on a number of
variables, such as the task and the intelligence of the problem solver.

24
 Management Information System – Unit II – Based on BSc IT Calicut University

Task Environment and Problem Space

The task environment is the problem as it exists in the real world. The problem space is how a
person mentally represents that problem to solve it.

When faced with a problem, a person creates a mental model (problem space) based on the actual
situation (task environment). The way this space is structured affects how the person solves the
problem.

The problem space is like a network of knowledge states—each state is what the person knows at a
certain point. Solving the problem means searching through these states until the goal is reached.

Well-structured problem spaces help by showing patterns or predictions, making the search process
more efficient.

Heuristics

25
 Management Information System – Unit II – Based on BSc IT Calicut University

Humans often use heuristics—mental shortcuts or "rules of thumb"—to make decisions instead of
analyzing all possible options. This approach is based on experience, intuition, and judgment and
helps save time by narrowing down choices.

While heuristics can be efficient and effective, especially for those with experience (like expert chess
players), they can also sometimes lead to poor decisions if used incorrectly.

This behavior is explained by the idea of bounded rationality:

 Humans have limited thinking capacity, so they create simplified models of real situations.
 Their decisions might make sense within the model but not always in the real world.
 Factors like age, education, and attitude also affect decision-making abilities.

Skilled people are better at forming useful heuristics and applying them to solve problems
effectively.

TENTATIVE LIMITS ON HUMAN INFORMATION PROCESSING

The Newell-Simon model proposes that humans have limitations as information processors.

These limitations, supported by evidence, affect how we solve problems using heuristics.
Specifically, our abilities are limited in:

 Processing data, due to the constraints of short-term memory.

 Detecting differences.
 Generating, integrating, and interpreting probabilistic data.

Limits to Short-Term Memory

 Psychologist George A. Miller proposed that humans can hold 7 ± 2 "chunks" of information in
short-term memory.
 This means people typically remember 5 to 9 items at a time.
 This limit is evident in tasks like memorizing numbers, symbols, or codes.

Supporting Studies:

Study Findings
Crannell & Parish Immediate recall of numeric data is slightly better for 7 digits.
Error rate increased with digit length:
Conrad
8 (30%), 9 (44%), 10 (54%).
Most frequent errors with 9 and 12-digit codes. Errors increased with code
Chapdelaine
length.

26
 Management Information System – Unit II – Based on BSc IT Calicut University

Study Findings
Osowitz &
Errors rose when alphabetic and numeric characters were mixed.
Sweetland

Extending Memory Using Other Methods:

 Graphic Representation: A graph is a chunk that helps visualize trends and relationships.
 Spatial Analogies: Memory improves by linking numbers to known visuals (e.g., room furniture).
 Visual Strategy: People use visual images or object placement (like desk items) to remember
better.
 Trained Memory: With practice, some individuals can recall up to 75 digits!

Just Noticeable Differences

 JND refers to the smallest difference in a stimulus that a person can detect.
 It's important for identifying errors and noticing changes in data or values (e.g., profit or cost
differences).
 Weber’s Law explains that:

This means the difference needed to notice a change depends on the original size (proportional).

Handling Probabilistic Data

Decision makers often need to assess uncertain events, but humans are not naturally good
intuitive statisticians. Research shows several key deficiencies:

 Poor understanding of sample size effects – People fail to grasp that smaller samples lead to
higher variability (e.g., 2 defects in 10 vs. 20 in 100 is not the same).
 Inability to identify correlation vs. causality – People assume causality from mere co-occurrence.
 Use of biasing heuristics – Estimations are influenced by mental shortcuts (e.g., availability or
recency).
 Poor integration of information – Difficulty in combining data logically.

27
 Management Information System – Unit II – Based on BSc IT Calicut University

Sample size significantly affects data reliability: larger samples are more accurate. Humans often
misinterpret small samples, leading to incorrect conclusions because they don't grasp the inherent
unreliability of limited data.

In decision-making, identifying correlation and causality is crucial, as it helps predict one variable
based on another. However, people often mistake simple co-occurrence for causality, even when no
real link exists. For example, they may blame their own failures on external factors like luck but
view others' failures as due to personal flaws. This misjudgment is known as illusory correlation,
where individuals see a relationship between variables just because they expect one. Additionally,
human judgment is affected by biases—availability bias (giving more weight to easily recalled
events) and recency bias (favoring recent information). These cognitive limitations lead to flawed
probability estimations and decisions.

When people know an outcome occurred in the past, they tend to overestimate its chances of
happening again. This is known as hindsight bias and can hinder learning by making people rely too
much on past events instead of evaluating current situations objectively.

Humans often make inconsistent choices when given different types of information. Research shows
that people give more importance to information presented in the same units as the decision being
made — this is called cue-response interaction. For example, in financial decisions, people are more
influenced by data shown in money terms than by simple counts. Saying "$110,000 in sales were
lost" has more impact than just stating "10 sales were lost." How information is presented can greatly
affect judgment.

Human Information Processing Strategies

To manage their limits in processing information, humans use strategies like concreteness and
anchoring and adjustment. Concreteness means people rely mainly on information that is clearly
presented and easy to access, rather than searching memory or transforming data. As a result, they
focus on obvious, visible details and may ignore important background or context, which can lead to
poor decisions.

Anchoring and adjustment is a judgment strategy where people start with an initial value (anchor)
and adjust from it. This simplifies decision-making but can lead to errors. People often choose
anchors based on past experience or similar units, even if they’re not suitable. Also, they may not
adjust enough from the anchor, giving too little importance to new information. This is common in
budgeting, planning, and pricing.

28
 Management Information System – Unit II – Based on BSc IT Calicut University

CONCEPTS OF HUMAN COGNITION AND LEARNING

Much of the research in human cognition and learning is important for information system design in
that it explains how human capabilities can be enhanced or supplemented by information systems.
Some relevant topics are cognition theory, cognitive style, and learning theory.

Cognition Theory

Cognition is how individuals resolve differences between their internal understanding and what they
actually observe in the environment. Cognitive models aim to explain how people think and solve
problems.

 Newell-Simon model explains human problem-solving behavior.

 Cognitive dissonance theory shows how people adjust past beliefs to match current choices.

In information systems, Shneiderman’s model of programmer behavior is important. It explains that

skilled programmers rely on two types of knowledge:

1. Semantic knowledge – General programming concepts (e.g., loops, algorithms), useful across all
languages.
2. Syntactic knowledge – Specific grammar and rules of a programming language.

Shneiderman emphasizes that semantic knowledge is more crucial for programming expertise.
Experienced programmers can easily learn new languages because their semantic knowledge is
strong and transferable. Programming methods like top-down design and structured programming
help develop this kind of knowledge.

Cognitive Style

Cognitive style refers to how individuals gather, organize, and evaluate information during decision-
making. People may reach the same decision but often use different thinking processes.

Cognitive style can be described along two dimensions:

1. Information Gathering:
 Perceptive individuals focus on patterns and relationships to understand the big picture.
 Receptive individuals focus on details to gain specific knowledge.
2. Information Evaluation:
 Systematic (analytic) individuals follow a structured, logical process.

29
 Management Information System – Unit II – Based on BSc IT Calicut University

 Intuitive (heuristic) individuals use trial-and-error, act spontaneously, and respond to

nonverbal cues.

In designing Management Information Systems (MIS):

 Systematic users prefer structured systems with clear methods.

 Heuristic users need flexible systems that support exploration, user control, and multiple data
formats (visual, verbal, graphical).

However, cognitive style is not fixed. It exists on a spectrum, and people can adapt their style based
on the task. Training and experience can influence decision-making more than natural tendencies.

Left Brain-Right Brain

Research shows that the brain has two specialized sides:

 Left brain: Handles logical, analytical, and rational thinking.

 Right brain: Handles intuitive, creative, and visual thinking.
Although people can use both sides, education and experience often make one side more dominant
in problem-solving. This affects how individuals approach tasks.

Some believe cognitive style mainly reflects left-brain activity, even in intuitive people. However,
most decision support systems are designed for left-brain users and don’t fully support right-brain,
creative thinking.

To support right-brain users, systems should use visuals, graphics, natural language, color, and
even unexpected options. While this idea is helpful in understanding differences, some researchers
question its accuracy.

Learning Theory

Information and information systems support individual learning, which involves four key elements:

 Drive – The motivation to learn (e.g., curiosity or need).

 Cue – A signal or stimulus that guides a response.

 Response – The action taken based on the cue.
 Reinforcement – Feedback (reward or punishment) that helps learning stick.

30
 Management Information System – Unit II – Based on BSc IT Calicut University

Adults learn not just from natural curiosity but also from acquired drives. Some responses help label
problems (e.g., calling it a "rate of return problem"), which then guide further actions. Information
systems help learning by offering cues, responses, and ways to organize or label problems.

Artificial Intelligence (AI) aims to perform tasks that seem "intelligent" when done by humans. It
helps in two ways: understanding human thinking and making computers better problem-solving
tools. A key goal is to simulate expert thinking through expert systems.

CHARACTERISTICS OF HUMAN INFORMATION PROCESSING PERFORMANCE

Need for Feedback

In both computer and human systems, feedback is essential to confirm that output or input has been
received. In humans, feedback also fulfills a psychological need for assurance.

In online systems, feedback is often measured by response time—the time between a user action and
the system’s reply. Good response time improves speed and reduces errors.

 Too long = user loses focus.

 Too short = user feels rushed, causing mistakes.
 Ideal: 2–3 seconds for data entry tasks; longer is okay for complex tasks.

If response time exceeds 10 seconds, the system should display a message showing it’s still working.
Also, keeping response times consistent improves user experience.

Psychological Value of Unused Data

Organizations often store large amounts of data that may never be used. This happens for several
psychological and symbolic reasons:

1. Confidence Boost – Decision-makers feel more confident just knowing extra data is available,
even if it's not needed.
2. Unused Opportunities – People value options they might use, like living near cultural facilities
they rarely visit. Similarly, stored data is valued for its potential future use.

3. Information as a Symbol – Having access to information symbolizes competence and rational

decision-making, even if the data is never used.

Feldman and March add more reasons:

31
 Management Information System – Unit II – Based on BSc IT Calicut University

 Gathering info is often separated from using it.

 Some data is collected just for monitoring, not decisions.
 Information can be used to persuade or mislead.
 Most importantly, having data signals professionalism, regardless of actual use.

For system design, this means giving users access to information (like through terminals) may satisfy
their need without printing or storing unused reports.

Information Overload

Humans have limited capacity to process information and use natural filters to manage it. However,
due to rapid technological change, we now face information overload, which leads to stress and
health issues, as noted by Alvin Toffler.

With cheaper technology and better storage, organizations collect and share more data than ever.
Managers often ask for more information, thinking they lack enough, but the real issue is too much
irrelevant information.

To manage this, information systems should include tools to filter, summarize, and reduce data,
helping users focus only on what’s important.

Individual Differences

People differ in how they think and use information systems. Cognitive style is one such difference,
but there are many others that affect system use and satisfaction.

While it’s hard to create systems tailored to each person, designers can either:

 Build systems based on the most common user traits, or

 Include options to suit a range of user preferences.

Considering these differences helps improve system effectiveness and user satisfaction.

Nonverbal Information Input

While information is often shared through words or writing, nonverbal communication (like body
language) also plays a key role, especially in face-to-face settings where more than half of the

32
 Management Information System – Unit II – Based on BSc IT Calicut University

message can be nonverbal.

Nonverbal cues—like facial expressions, gestures, posture, and eye contact—can change or reinforce
the meaning of words. For example, saying "I'm angry" with a smile, scowl, or wink changes the
meaning each time.

There are seven main types of nonverbal communication:

1. Hand movements (e.g., gestures, pointing, fidgeting)

2. Facial expressions (e.g., smiles, frowns)
3. Eye contact (shows attention or interest)
4. Posture (open or closed body positions)
5. Proxemics (use of space shows relationship)
6. Body rhythms (movement patterns, speaking turns)
7. Speech tone and word choice (show emotion or involvement)

Cultural differences affect how body language is used and understood. For example, nodding may
not always mean agreement in every culture.

Face-to-face communication is often preferred because nonverbal cues help improve understanding.
Research shows that voice and face-to-face interactions are faster and more effective than
handwriting or typing for problem-solving, especially in emotional or complex situations.

33