The role of model and modelling in scientific research

Dr Nguyen Hoang Tien CONFERENCE on THE ROLE OF MODEL AND MODELLING IN SCIENTIFIC RESEARCH “Methodology for writing scientific and research dissertation” Abstract summary: This article

Dr Nguyen Hoang Tien

CONFERENCE on

THE ROLE OF MODEL AND MODELLING IN SCIENTIFIC RESEARCH

“Methodology for writing scientific and research dissertation”

Abstract summary: This article discuss the role of modeling in scientific research,

how to create, verify and apply elaborated model, especially in economic and management science The real necessity and both positive and negative sides of economic modeling are here to be discussed

Tóm tắt: Bài viết này thảo luận về vai trò của mô hình hóa trong nghiên cứu khoa

học, làm thế nào để xây dựng, kiểm duyệt và ứng dụng mô hình, đặc biệt trong lĩnh

vực khoa học kinh tế và quản trị Tính cấp thiết và cả những lợi thế và bất lợi của việc

mô hình hóa trong kinh tế cũng được đề cập tới và phân tích thảo luận

1 Defining model and modelling

The reality and the real world are rather complicated, unpenetrable and sometimes very hard to understand In today’s society and business world, the environment, the whole business context of all the organizations seem to be exceptionally and increasingly incomprehensible, erratic and unpredictable It is very frustrated for the researchers when it comes to the dealing and solving certain business problems and decision making Most often, we have no idea on how to solve a concrete problem confronting us More over, we are not able to properly formulate a problem, to precise its core issues That is how to locate problem in certain time and space horizon, at the same time, comprehending its size, internal structure and consequences caused or may be caused for the whole or parts of the firm’s activities [1 p 94]

Solving a properly formulated problem, from both content’s and form’s side,

is based on a sequence of activities which could be well decomposed into logical

2

problem’s modelling stages: analysis, synthesis, and assessment Anyway, the problem’s modelling is vitally important to the correctness of its solution

Model of a phenomenon is called by W Flakiewicz as “a description of a fragment of the reality, taking into consideration only those essential elements of

it while omitting the less important ones” [1 p 105] This is, in general, a justification of the need and the necessity to apply models in the scientific works The reality surrounding us and the phenomena existing out there are featured with high complexity We, the researchers, are not able to identify and describe all the elements and the relationships between them However, we are able somehow to make a selection, that means to choose only some important elements that are easy for us to spot The selected elements of our created model have a subjective character Simply taking, model of a real world is a simplified version of it, reduced in terms of the level of complexity, with highlighting the important to us elements and dimensions

A.Rappaport wrote: “Models in their broadest meanings may be considered as abstractions from the reality with the purpose to establish an terminology order to our complex environment” [2 p 90] According to R L Ackoff, “Models are the presentations of states, subjects, objects or events They are idealized in a sense that they are less complicated than the real world and hence easier to use in the empirical researches It is easier for us to use and manipulate models than real research objects The simplicity of models in comparison with the reality is implied from the fact that it takes into consideration only those features of the reality which are important in a given case” [3 p 142]

Upon the process of defining a model, we should use such terms as: analogy, symmetry, affinity, representation, mapping However, it is hard to define the notion of model not referring to the term of isomorphism “There, where a scheme

of relations between elements of the X system is the same as in the Y system, we are to deal with isomorphism of the two systems” [6]

In the research processes, there are in use different kinds of instruments to facilitate understanding, familiarizing and explaining the researched subjects The notion of model is such a tool which helps to carry out certain simplifications in order to better understand a research subject The cited O’Shaugnessy wrote: “if the research of the X system’s structure is useful from perspective of understanding Y system then X is a model for Y The ideal is having an isomorphic model Perhaps the contemporary pace of scientific development is

owed to the findings of mathematical models corresponding to the structure of physical world” [6 p 203]

It is hard or nearly impossible to imagine the progress achieved in the scientific development without creating and using model Most of the doctoral dissertations

in their intentions tends to propose or modify an applied model before or so far It

is accented in the title of the dissertations or as a purpose of scientific elaborations

As we see further, the notion of model is used in each phase of problem solving The processes of description and explanation the reality are expressed using a model (or models) Different forms of models are used in the phase of prediction and implementation of the research results In each case, it is about certain representation of relations between elements of a system In other words, it is about certain set of correlated and impacting on each other parts which’s structural rules determine their essence or mutual conditionings

The term of model competes with the term of theory In many fields, it is easier

to create a model than to propose a certain theory (or a set of theories) Sometimes,

it seems that the path of development is coming from the model to the theory But

in reality it is conversely The scientific development is to run from the theory to model Creating a model is, for many fields of science, a significant step forward, sometimes a quite big step enough to be called a breakthrough

It is worth to remind that in the beginnings of the seventies, it was so fashionable to define model so broadly that to embrace each of the physical or abstract representation of a problem This approach dominated no longer in the methodologies of science However, in a unlimited number of efforts to solve the problems, it is assumed that the explicit or implicit usage of model are tolerated And all the unveiled relations become elements (parts) of model Even the classification system can be seen as a model The fact that almost everything in the fields of science are possible to be described, explained and predicted using model will cause a lot of optimism

2. Models and modelling in social, economic and management science

In the fields of social sciences and also in economics and in management science the term of model becomes gradually a subject of banalization, rather than

a key to solve important problems There are many reasons responsible for that: a) Firstly, it is rarely possible to elaborate models fulfilling isomorphism’ criterion The more the reality is complicated, i.e the whole economy, the more the further simplification is needed for its understanding and interpretation Such

4

simplification may be quite important weakness if the process of formalization is

to take place on the cost of understanding the real world

b) Secondly, simple (with the necessity to be seriously simplified) model bring about nothing or nearly nothing to the process of scientific discovery, it could even falsify or darken this whole process There always exists a problem to determine the adequacy level of the model from the perspective of its operationalization

c) Thirdly, each of the researchers is to confront with choices whether to start from a simple model and then gradually introduce more realistic (and more complicated) assumptions or also to start from complex model and then, by reduction, to simplify to the moment when the model is possible to be used conveniently It seems that there is no one universally good solution Both approaches have their own advantages and disadvantages

Many scientists and researchers have warned against the possibilities and temptations of oversimplification of the researched reality During the process of conducting a research, it is advisable not omit the preliminary, descriptive stage showing the diversity and complexity of events to be reflected by each theory or model, if they have to be adequate to a certain extent If, in the process of familiarization and understanding, this stage is forced to be removed or overly simplified, unadequate models may appeared, corresponding to the strict rigour, but not leading to the fullest understanding of interesting us phenomena O’Shaughnessy reminded that models of setting price created by economists are evolving towards limitation of new approaches Also in sociology, it is easy to see that circulative conceptional models of society are focusing on the problems of balance, mutual dependencies and standard variations, but ignoring problems of change and social development

In management science there are different useful models developed to use by decision makers and managers High level of operationalization is associated with decisional models based on theories of limited rationality H A Simon [7 pp 174-176], models of strategic management, or more comtemporary models of knowledge management The attractiveness of each new field of knowledge, new approach is assessed in the category of proposed conceptional models, new heuristiques, or new classifications In practical sciences, it is important to propose new solutions to the problems We are discussing here about positive

side of the process of comprehension or implementating the knowledge into practice However, one should remember that even when the model is used explicitly, there is a real danger of ignoring its assumptions and wrongly taking it (model) as a reality Models may be only academic toys,which’s manipulation delivers us intelectual excitements but nothing more Models of this kind do not reflect strictly the reality such as musical notes do not reflect strictly the whole melody

3 Models’ construction and verification as an essence of scientific research

In empirical sciences the process of model’s construction runs through an well determined scheme (see the illustration below) The starting stage of this process

is to summarize the observations that constitute the event explained and an attempt

to select essential variables As the next, a stage of data analysis is carried out in order to establish mutual relations between variables and arrange data in diverse possible models which may explain changes in relations to the problem

Each model is analyzed from the respect of internal logics, realism and its relation to the existing knowledge The choice of one from many models is made and a forecast for the resulting consequences is carried out The final stage of this process is to verify through confirming the forecasted consequences The first stage of model’ construction is the only that may cause a lot of difficulties Once

we are familiar with structural rules of certain system and the features of its elements, we can, through the method of deduction, prove the impact of those structural rules on the elements of system For the representatives of social and management science, the structural rules are not simply given once forever Essentially, they are empirical laws (i.e the relations between advertisement and the quantity of sales) which must be discovered Similarly, the essence of the variables of the system must be precisely unveiled Because, the level how they are understood will put a limit on the prediction of their mutual impact upon the acceptance of those structural rules as given Some management science’s representatives pay too little attentions to the experience They behave as though they believe (alike the representatives of rationalism) that the reflection of pure reasoning over the problem, without taking advantage of the observational data may deliver enough foundations for truthful convincements

6

4.Conclusions

What kind of purposes the model may serve The model serves not only the managerial orientation in the real business environment Moreover, the model can replace the real experiment through reasoning that consists in the change of values

of each variables and check out the consequences in relations to the rest of variables So, the correctly constructed model will enable managers not only to orientate in the current business reality, but it also help to predict changes related with the considered impact on certain fragments of the reality

Reference

1 W Flakiewicz: Taking business decisions, PWE publisher, Warsaw 1973

2 A Rappaport (red): Information for decision making New Jersey 1970

3 R L Ackoff: Optimal decisions in applied researches PWN publisher, Warsaw

1969

Purpose of modelling

Model category Model structure Identification

Algorithms Computings Forecasts

Verification

Data

Verified model

Hypotheses

Theories

Laws

Empirical

knowledge

4 A Sulek: Experiment in social research PWN publisher, Warsaw 1979

5 P Sztompka: About modelling in sociology “Sociological studies”, 1968, No 1

6 J O’Shaughnessy: Methodology of decision making PWE publisher, Warsaw

1975

7 H A Simon: Administration’s activities PWN publisher Warsaw 1976

8 S Beer: Cybernetics and management PWN publisher, Warsaw 1966