Special scientific methods in research

Complementarity between analytical and synthetic methods

The study and understanding of phenomena, for example biological, requires the use of different scientific methods of research and reflection. The two major scientific methods, complementary, are reductionist analysis and systemic transdisciplinary synthesis.

Reductionist analysis

Reductionist analysis consists in breaking down biological systems into levels of organization and elementary units, the smallest and simplest possible (it is reductionism, by breaking down into: organs, tissues, molecules, atoms, subatomic particles… ). Then at each level of organization, each of these elementary units is studied in detail by a specialized discipline (physiology, molecular biology, genetics, chemistry, atomic physics, etc.), in order to understand its structure and functioning. This method is used in most scientific laboratories.

For example, the main levels of organization and analysis of the nervous system are, in order of reduction:

Systemic, the study of the nervous system as a whole, consisting of the neural organ and the other organs or systems with which it interacts (association and interactions with the musculoskeletal system to achieve motor skills; interactions with the digestive system for regulation of digestion, energy balance and feeding behavior; interactions with sensory organs for orientation, identification and understanding of environmental cues; etc.).
Organic, the study of the neural organ, made up of all the neurons and associated elements (glial cells, cerebrospinal fluid, etc.).
Neural network, the study of the structure and functioning of neurons when several of them are interconnected.
Neuronal, the study of the neuron, which is the elementary cellular unit, basic constituent of neural structures.
Synaptics, the study of the structure and functioning of connections between neurons.
Molecular, the study of the different molecules that exist in the nervous system.

Systemic transdisciplinary synthesis

Systemic transdisciplinary synthesis consists of bringing together data from different disciplines and different levels of analysis, then performing a synthesis of all this information, in order to develop general models of the functioning of systems. The objective of the systemic transdisciplinary synthesis is to arrive at an overall understanding of the system studied. But this method is still little used, particularly in anthropology, and its development remains more theoretical than practical.

Specific methods

In addition to the general methods described above, there are more specific methods that apply to particular disciplines.

Double-blind method

Faced with the disease, some recover spontaneously, others react more or less to medication, and moreover, the very fact of taking a treatment without effect can sometimes have effects (beneficial or negative) (placebo effect). The so-called “double-blind randomized” studies in medicine, psychology or social sciences, thus make it possible to avoid certain biases.

Replication method

Particularly in the complex emerging fields of biological and ecological effects, especially at low doses, and/or when there are industrial and/or regulatory issues at stake (eg standards concerning radioactivity, GMOs, radiofrequencies, additives …), or when the subject is controversial (eg certain vaccines, exposure to certain radio frequencies, etc.), many study results, sometimes unexpected or counter-intuitive, will only be accepted as “valid” by the scientific community. ‘after the study that highlighted them has been replicated once, or even several times, and by different laboratories. This is why the “good” scientific publications contain (sometimes in appendix) a precise description of their protocols, conditions and materials used. Obtaining an identical result is a necessary but not sufficient condition for validation, because during replication a discreet and unidentified bias can also simply be reproduced. It may be useful to replicate a study by varying certain parameters (for example the same agent may prove to be toxic for the rat but not for the mouse, or only for the fetus, or only for males, or only when is combined with another element, etc.).

Dual validity assessment

In the 1950s, Donald Campbell proposed the concepts of internal validity and external validity to estimate the degree of confidence to be given to the result of a scientific experiment in the social sciences.

History and idiographic approach

In essence, the historical fact is “past” and “singular”: the wheel may have been invented many times and in many places, but each time in a singular context; and it seems that there was only one Russian Revolution, only one ancient Egypt. The analysis of these singular events is therefore based on an idiographic approach. Authors including Karl Marx and Carl Hempel, however, have attempted to develop a nomothetic approach to history, arguing that there are “laws of history”. This conception of history was notably criticized by Karl Popper in The Poverty of Historicism.

See also the articles on consilience, Darwin and then shaping the ideas of the theory of evolution for another example of historical type science. It is also criticized and rejected as science, at one point, by Popper as shown in his article.

This shows that the sciences of historical types, although as powerful as other scientific methods, are just as much misunderstood, poorly mastered and often rejected by many specialists. The latter prefer stereotypical methods such as physics or other experimental sciences. However, they often only confirm (and psychologically validate) the results of the historical type method. This was also the case for Alfred Wegener and his theory of continental drift confirmed by physics and thus recognized by all after 40 years of heated controversy, ranging from simple disruption of classes by teachers “against drift” to rifle shooting in the colleague’s empty office door; although the various and multiple equally scientific sources that were brought during all this period had validated it long before the proof by physics.

Social sciences

Two major currents of methodology come to account for reality in the social sciences. Qualitative methods and quantitative methods. These analysis methods can be used alone or independently of each other, in complementarity, they can each provide different elements of answers or can even be used jointly to analyze the same aspect twice and thus validate it. According to Gilles Gaston Granger, in Qualitative models, quantitative models in scientific knowledge, qualitative methods, usually perceived as excluding scientificity, have made a significant contribution to contemporary science by making it possible to account for the structures that can be observed and these forms, although measurable, are first of the order of qualitative assessment:

“The evolution of awareness of the deeper nature of scientific thought could be symbolized very schematically, by three mottos, each of which in some way reinterprets and rectifies the previous one. It was first proclaimed that there was no science but the universal; then that there was no science except the measurable. We should say today: there is no meaning except the structurable. A profession of faith that in no way rejects the two preceding ones, but relativizes them, and gives a new meaning to the universal and the measurable”.

Thus, it is appropriate to recognize the contribution of these methodologies in the social sciences. It is also possible to experiment with social network analysis. The study of the small world represents the most popular experiment of its kind.

The majority of contemporary economists agree that the methods used in economics should approach as closely as possible the methods of the physical sciences. The economists of the Austrian School maintain, on the contrary, that economics must, like mathematics and logic, be constructed by pure logical derivation from irrefutable axioms.

Unit issue

In the human sciences (economics, ethnology, psychology, etc.), the experimental approach is delicate, the predictive aspect of the method applied to human phenomena often being faulted.

Faced with this difficulty, two opposing attitudes have emerged:

The emergence of the human and social sciences from the end of the 19th century and into the 20th century led to questioning the antiquated model of the scientific method, which defines the notion of science in a reductive way.
For other authors, such as Michel Foucault in Words and Things, one must, on the contrary, be wary of the tautology which consists in defining a discipline as scientific because its name contains the word science. It would therefore be desirable that, like philosophy, these disciplines assume themselves as a rational approach to the study of reality without possible experimentation.

Universality

The method evolves over time. It also evolves in space.

The activities carried out in the different experimental sciences are so diverse that it would be futile to try to model them. On the other hand, scientific approaches, supposed to create knowledge, have common and universal characteristics that it is possible to exhibit. The model of the experimental approach has two complementary and inseparable descriptions:

The experimental approach goes through three stages. Those are :

a questioning phase;
a phase of finding an answer to the question asked;
a validation phase of the response found.

The fields of validity of the answers found being limited, the validation cannot be done by a direct confirmation, but by a succession of non-infirmations.

The experimental approach necessarily involves three areas:

the “real” domain: for physics it is identified with the material world which includes “objects” participating in events whose evolution we want to describe;
the theoretical domain includes the intellectual tools forged to answer questions: theories, concepts, etc. ;
the technical field includes experimental devices, measuring devices, etc. Measurements or the determination of measurable indicators are an integral part of the process.

(Includes texts from Wikipedia translated and adapted by Nicolae Sfetcu)