Focus on Ordering

Two Diagonal Sets: Approach Duality

It is evident from the TET diagram that the layout generates two sets of methods lying along the two diagonals. These two sets handle research in opposite ways as follows:

  • Bottom L — Upper R: These focus on ordering to develop a consensus.

  • Bottom R — Upper L: These focus on questioning existing relationships.

Both requirements are so central to inquiry, that researchers draw on methods from each diagonal-set as explored later.

See below: Comparison of methods on the Ordering diagonal.
(Next Topic: Comparison of methods on the Questioning diagonal.)

Ordering

The ordering of reality through organising observations and/or concepts into a consistent coherent structure is fundamental to science. Perception and demonstration of order is the focus for methods that lie along the diagonal that stretches from bottom left to top right.

  • Analytic method (L'2)
    Analysis attempts to bring order to a confused picture. This may involve clarifying concepts or disentangling differences in viewpoints. The researcher can pick and choose what to include in the argument so long as it generates a coherent doctrine, theory, account or framework for further inquiry with other methods. Ordering values consistency and coherence while tolerating imperfections and omissions.

  • Formal method (L'6)
    Pure reasoning is used to order pure abstractions: mathematics is the prime example. Mathematical formalisation can impose a precise ordering on either abstractions from reality or on the outputs of other methods. Algorithms are precisely defined and sequentially ordered procedures that solve problems and are the bases for computer programs. That means completeness is valued and omissions must be avoided.

  • Empirical method (L'1)
    Data based on careful observations must be ordered as part of empirical inquiry. Correlations are sought and laws determined by the use of induction. It may be possible to use mathematics to create an empirical model that captures more complex order: data are then collected to suggest the right model, to calibrate the model and to test the model. Unlike mechanistic models, such ordering makes no assumptions about relationships, nor about presumed causal processes.

  • Holistic method (L'5)
    Here inquiry produces an ordering of critical factors and forces that represents a bounded socio-physical situation. This system model is sufficiently complete if it can be used to manage change involving people. Sometimes, simulations can bring system models to life.

ClosedMore on Modeling

Everyday inquiry attempts to manage psychosocial situations. Scientific inquiry attempts to represent reality correctly regardless of any psychosocial realities. That is why science-based thinking regularly comes into conflict with political and ethical positions within society.

Representing reality is currently believed to reside within the Change Domain at Representation-PH3L6. In that level, there is a Principal Typology of Modeling Methods (PH'3).

These modeling methods-PH'3 reflect perspectives on change, and each is particularly suited to different situations. Resonance with the Research Methods is noticeable:  Dualistic method-PH'3L4, which posits that reality is represented by two opposing forces, resonates with the Dialectic research method-PH'2L4. Causal method-PH'3L3 which posits that reality is represented as a multiplicity with a focus on certain elements that interact, resonates with the Explanatory research method-PH'2L3.

Is the order really there? (i.e. is it true?)

This is a question. To test is to question. Questioning belongs on the other diagonal.

Order is always hypothetical. So the urge to test drives methods towards the Explanatory Method in the lower right quadrant of the TET.

Analytic theories, formalizations, empirical correlations and system models are preferred if they are simple, elegant and parsimonious. However, many accounts deemed correct are rather complicated.

Of greater importance is the range of observations that can be interpreted (explained, understood) in terms of the ordering. Definite predictions should flow and be tested and possibly falsified by observations either made in the past or organized in the future.

Formal ordering cannot usually be tested in that way, but in some cases computer simulations can do the testing.

Increasing Inclusiveness

Inclusiveness is the core issue in ordering i.e. where is the boundary of interest placed? what is covered and what excluded? The methods tolerate different degrees of coverage, with inclusiveness increasing in moving up the diagonal. The diagonally higher method draws freely on lower methods, but the reverse is not the case.

  • Analytic method is least inclusive because the scope of study may be altered at will and evidence may be selected. So boundaries are arbitrary and not everything within them needs attention. Arguments are mostly affected by what will be viewed as reasonable.
  • Formal method is more inclusive because it restricts itself deliberately to a few starting assumptions, axioms, strict logical methods, and formerly proven theorems. Boundaries are sharper and everything within should be included. Any formal model typically works in a tightly defined class of application.
  • Empirical method is significantly more inclusive because ordering draws on formal methods (like statistics), uses reasonable argument as to what counts as data or facts, and is open to additional relevant observations that define or extend the association being developed or tested.
  • Holistic method is the most inclusive because it explicitly attempts to identify and represent all relevant factors: physical/technical, personal/social, abstract/mathematical, within/without the system boundary.

Methods on the Other Diagonal

The need for ordering (or modeling) is recognized in all inquiry, but consistency and coherence do not have to be the defining quality of a research method. Inconsistency or incoherence does not weaken methods on the questioning diagonal. Instead, the findings are accepted and developed for further testing using other appropriate research methods.
ClosedDetails:

Originally drafted: 10-Apr-2015. Last amended: 21-Feb-2022.