Mason & Mitroff (1973)“A Program for Research on MIS”

The core idea of the paper is that a Management Information System (MIS) isn’t just a computer program or a set of reports. It’s something much bigger, involving these five key parts that work together:

1. The PERSON: Who is using the system? What are they like?
2. The PROBLEM: What kind of tough situation are they trying to solve?
3. The ORGANIZATION: What’s the structure and context of the company they work in?
4. The EVIDENCE: How is the “knowledge” or “information” created or proven to be reliable?
5. The PRESENTATION: How is that information shown to the person?

The authors’ main point is that most MIS designs at the time focused on only one limited option for each of these five parts, completely ignoring many other possibilities. They propose a research plan to study all these other ways.

Let’s look at each part in simpler language:

1. The PERSON: Different Thinking Styles (Psychological Type)

People’s minds work differently. The paper uses ideas from psychology (like Jung’s types, known today partly through the Myers-Briggs Type Indicator). It focuses on how people take in information (perceive) and how they make decisions based on it (evaluate).

• Perceiving:
  • Sensation: Focuses on the immediate, practical, “just the facts” kind of details. They like concrete data and are cautious about guessing beyond it. Good for managing daily operations.
  • Intuition: Looks for the big picture, possibilities, and connections. They see beyond the facts and are good at coming up with strategies and new ideas.
• Evaluating:
  • Thinking: Uses logic, rules, and abstract reasoning to make decisions based on what is true or false. They like to build systems and models.
  • Feeling: Uses personal values, empathy, and considering what is good or bad, pleasant or unpleasant. They are concerned with moral issues and human impact.

Combining these gives us types like Thinking-Sensation, Thinking-Intuition, Feeling-Sensation, and Feeling-Intuition.

The big problem the authors see is that most MIS designers tend to be Thinking-Sensation types themselves, and they design systems that only suit that style. They assume everyone wants information like “raw data,” “hard facts,” or abstract models. This means MIS has largely ignored the needs of Feeling and Intuition types, who might prefer information presented as stories, possibilities, art, or human drama. The first guiding principle they suggest is that managers need information that matches their own psychological style, not just the designer’s. Designers need to understand both the manager and themselves.

2. The PROBLEM: Simple vs. Messy Situations

Problems aren’t all the same.

• Structured Problems: These are neat and tidy. You know what the choices are, what the possible results are, and how likely they are to happen (or you’re certain). Think of optimizing a production schedule where all the variables are known.
• Unstructured or “Wicked” Problems: These are messy, unclear, and hard to define. You might not know all the options, the possible outcomes, or even what exactly the problem is. There’s no single “right” answer, and you often just have to decide when you’ve found a “good enough” one. These problems often require the manager to gain a new understanding or “appreciation” of the situation.

The authors say that MIS design has focused almost completely on helping with the structured, well-defined problems that operations research is good at. Very little work has been done on how to create information systems that help managers deal with the messy, unstructured “wicked” problems they face in the real world. The second guiding principle is that managers need ways to get information that suits their specific problems, including the wicked ones.

3. The EVIDENCE: How Information is Created and Proven (Inquiring Systems)

Where does the information come from, and why should we trust it? The paper talks about different “inquiring systems,” which are like different ways of gathering and validating knowledge. They discuss five main types based on different philosophies:

• Lockean (Data-Based): This system builds knowledge from observations and “raw data.” It’s trusted if experts agree on the facts. Examples are databases, accounting reports, and statistics. Good for clear problems where everyone agrees on the basic facts.
• Leibnitzian (Model-Based): This system builds knowledge from logical rules and formal models. It’s trusted if it’s logically sound and consistent. Operations research models are a classic example. Good for clearly defined problems that can be put into a mathematical form.
• Kantian (Multiple Models): This system uses more than one approach, combining data and models. It uses multiple perspectives or theories to look at the same problem. It’s trusted if the models’ predictions match the data. Might be better for problems that are moderately messy.
• Hegelian (Conflict-Based): This system uses two completely opposite ideas or models to look at the same data. The goal is to create a conflict that forces everyone to see the hidden assumptions behind each idea. It’s trusted because the intense debate is supposed to reveal deeper truths. This is seen as potentially best for very messy (“wickedly” ill-structured) problems where there’s already lots of disagreement.
• Singerian-Churchmanian (Learning Systems): These are complex systems that learn and adapt over time. They can help turn messy problems into clearer ones and vice versa. They can also help us understand how the other systems work.

The authors state that most MIS design has heavily relied only on the Lockean (data) and Leibnitzian (model) approaches. This means they haven’t explored creating MIS that uses multiple views (Kantian) or thrives on conflict and debate (Hegelian) to tackle the really tough, messy problems.

4. The ORGANIZATION: Where the Problem Lives

The structure of the company matters for MIS.

• One way to look at it is by the level of problems the organization deals with:
  • Strategic Planning: Big-picture, long-term decisions about goals and overall direction.
  • Management Control: Making sure resources are used well to achieve goals.
  • Operational Control: Managing the day-to-day tasks efficiently. The paper says most MIS has been designed for the lowest level, operational control. Less has been done for management control, and very little for big-picture strategic planning.
• Another way to look at it is how the company structure itself (like hierarchy, specialization) can cause information problems (like hiding or twisting information). The paper notes that there hasn’t been much study on how different organizational structures affect MIS design. Also, information means power in an organization, which can cause resistance to new systems.

5. The PRESENTATION: How Information is Shown

How information looks and feels when it’s delivered makes a difference.

• Impersonal: Things like standard reports, abstract models, or computer printouts. The paper notes that MIS designers have often just assumed the computer printout is the best way to show information.
• Personal: Things like graphics, stories, discussions, role-playing, or even art.

The authors question if computer printouts are always effective, especially for different psychological types. They suggest that for Feeling or Intuition types, abstract numbers from a computer might not be helpful, and they might respond much better to more personalized, visual, or interactive ways of getting information. There needs to be more research on how these different presentation methods affect how well an MIS works.

The Big Picture

When you think about all the different options for these five parts (different types of people, problems, ways of getting evidence, organizational settings, and ways of showing information), you realize there are many, many possible combinations for what an MIS could be. The paper calculates at least 240 distinct ways based on their basic categories.

The main point is that past MIS design has been very narrow, generally assuming it’s for a Thinking-Sensation person, dealing with a structured problem, using data or models as evidence, in a traditional operational control setting, and getting information via an impersonal computer printout.

The authors ask: What if you change these assumptions? How would you design an MIS for an Intuitive-Feeling manager facing a wicked problem, using evidence from intense conflict, in a strategic planning, team-based company, and showing information through a personalized drama? Would the old ways of designing MIS still work?

They conclude by stressing that studying information systems is important and needs a more systematic approach, considering all these different parts and how they interact. It’s about designing systems that truly fit the people, problems, and organizations they are meant to serve, rather than fitting everyone into one narrow box.

For comps:

The Mason and Mitroff paper argues that Management Information Systems (MIS) research and design have been far too narrow, defining information broadly as “knowledge for the purpose of taking effective action”. They state that an MIS involves five key components: the PERSON using it (their psychological type), the PROBLEM they face (structured vs. unstructured), the ORGANIZATIONAL CONTEXT (level and structure), the method of EVIDENCE generation (different inquiring systems), and the mode of PRESENTATION (personal vs. impersonal). The authors contend that most MIS designs to date have assumed only one specific combination of these components – typically a Thinking-Sensation person, facing a structured problem, using data or models as evidence, in an operational control setting, and receiving impersonal computer printouts. Their central thesis is that a comprehensive research program must explore the many other possible combinations of these variables (potentially 240 ways based on their categories), emphasizing that managers need “information” and evidence-generating methods that are geared to their own psychology and problems, not just those of the system designer.