DBA 702

DBA702chapter4.pdf

Home >Business & Finance homework help >Operations Management homework help >DBA 702

4 THE ANALYTICAL VIEW

This chapter will present the analytical view in detail. Concepts on which the view is based will be provided, such as models, causal relations, hypotheses, problem formulations, deduction, induction, verification and abduction, operational definitions and ceteris paribus. The relation of the analytical view to its paradigm and some typical results from this view will also be provided.

SOME BASIC CONCEPTS

There is good reason to say that a methodological view is its concepts. Major parts of this chapter are therefore devoted to presenting concepts that are often used by proponents of the analytical view: reality and models, causal relations, hypotheses, deduction, induction, abduction and verification, operational definitions, analysis and ceteris paribus (“other things being equal”). We will finish by discussing how the analytical view relates to its paradigm and look at various ambitions associated with being an analytical creator of knowledge.

REALITY AND MODELS

The analytical view is based on the assumption that reality is factive. There are objective facts and subjective facts. Both are looked at as true. Objective facts are circumstances, which, in principle, are indisputable, not questionable and not influenced by somebody’s opinion. This could be the age of a company, which product is selling best or the telephone number to the present general manager of the firm. There are also subjective facts. These are true opinions that people may hold. To the extent that they are facts, they are, from a methodological point of departure, treated in the same way as objective facts. When looking for facts, especially in the subjective case, the question of whether a finding is true or not often arises. In the case of the analytical view we are faced with the task of discovering elements (things,

events, opinions) that are invariant in spite of changes in the environment and variations in perceptions among different individuals. These have always been the requirements of the analytical view. That which is invariant is seen as more “true”. Two questions are consistently asked:

Which are the facts? Arbnor, I., & Bjerke, B. (2008). Methodology for creating business knowledge. ProQuest Ebook Central <a onclick=window.open('http://ebookcentral.proquest.com','_blank') href='http://ebookcentral.proquest.com' target='_blank' style='cursor: pointer;'>http://ebookcentral.proquest.com</a> Created from indwes on 2021-03-16 10:02:01.

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

How can we explain these facts?

Being so concerned with invariance, it is natural that logic and mathematics have a dominant position in the analytical view. The following arguments are considered valid. a. Logical and mathematical knowledge is not subject to sensory illusions. b. Results from logical and mathematical analyses are universal and valid; they are not subject to change.

c. The only results that are precise and not subjective are those that can be formulated in mathematical terms resulting from calculations and measurements.

The ambitions in the analytical view are to work up pictures of factive reality (ambitions that the analytical view shares with the systems view). Some of these pictures can be called models. An analytical approach (we remind the reader that analytical “approach” means analytical view “in application”) is to look for representative models (many systems creators of knowledge have the same ambition). Models in the analytical view tend to contain quantitative elements. Models that are

quantitative, but do not adhere to laws of mathematics, must meet strict requirements as to their origin and form to qualify as models. In the analytical view, the researcher/consultant/investigator therefore has some guarantee that a model – if it is quantitatively based – actually reflects the invariant phenomena in question. Models with at least a certain minimum of generality can be part of the theoretical concepts

of the analytical view (see Box 4.1).

Box 4.1

Theory

Theory (from the Greek word theoria, reflection) is a term with various meanings, among others the following:

1. a hypothesis 2. a strictly verified assumption (vs. a hypothesis) 3. a purely intellectual explanation or understanding (vs. application or practice) (“theory

and practice”) 4. a set of systematically constructed theses.

In the analytical view we can say that theory contains models of factive reality, models that are valid for more than one case in real life. (The concept of theory is used in the systems and actors views as well. We will return to this.)

Arbnor, I., & Bjerke, B. (2008). Methodology for creating business knowledge. ProQuest Ebook Central <a onclick=window.open('http://ebookcentral.proquest.com','_blank') href='http://ebookcentral.proquest.com' target='_blank' style='cursor: pointer;'>http://ebookcentral.proquest.com</a> Created from indwes on 2021-03-16 10:02:01.

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

CAUSAL RELATIONS, EXPLANATIONS AND HYPOTHESES

A very important concept in the analytical view is hypothesis. In this view, a hypothesis is a suggestion of an explanation of certain facts and a guide in surveying others. Not all analytical studies start with the formulation of hypotheses. Studies for creating

knowledge might be explorative; in other words, their objective can be to formulate hypotheses. Furthermore, there are a large number of studies that are of a purely descriptive character. The goal of these studies is to present facts without trying to establish the “logical consequences” of these facts; that is, without attempting to explain any described invariance. Hypotheses may include possible factive patterns – guides or templates that give structure to

purely descriptive studies. But because the highest ambition of the analytical view is to find explanations, we will reserve the concept of hypotheses to apply to what might be called “statements (confirmed or not) about causal relations”, or the assumption that one factor (or group of factors) determines the emergence of another. Traditional psychoanalytical theory provides an example of a hypothesis of this type: the hypothesis that asserts that a person’s early childhood experiences are an important factor in determining the person’s adult personality. “Hypothesis” is a concept that can also appear in the systems and actors views, but in those

instances it is usually given a less precise meaning – namely, a general assumption or an idea about how something is constructed or that a phenomenon will occur. What’s more, it is not related to causal thinking and determinism, and the close connection to theory is missing. Hypotheses can originate from various sources. A hypothesis may simply be based on a

guess. It can be the outcome of results from various studies and an expectation that a similar relation among two or more variables will be valid in the current study as well. A hypothesis can also come from existing theory that allows a creator of knowledge to predict that if specific circumstances are present, certain results should follow. No matter where a hypothesis comes from, it has the following important purposes right

from the start for those analytical studies whose aim it is to find explanations. It is the lodestar for: a. the type of data to be collected in order to confirm or reject the question formulated in the study

b. the extra data to be collected to test the veracity of the question formulated in the study c. the best way to organize these data for the analysis.

The background of a specific hypothesis is of great importance to the contribution the study can bring to the state of general knowledge (to existing theory). A hypothesis that originates in intuition or a mere guess can eventually offer an important contribution to the state in an effort to create knowledge. Yet if it has been tested in only one study, it has two limitations to its usefulness. First, there is no guarantee that the relations that were discovered between two variables in the study will appear in other studies. Second, a hypothesis based on a mere guess is probably not directly related to other knowledge or to existing theory. In consequence,

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

results based on it have no clear connection to the great amount of social science knowledge that is now available. The results may well pose interesting questions or stimulate further studies and eventually be incorporated into existing theory. But until that happens, the results provide only some isolated information. A hypothesis that emerges from the results of other studies escapes the first of these two

limitations, at least to some extent. If the hypothesis is based on what other studies have stated and the subsequent study supports the hypothesis, the results verify that some kind of regularity in factive reality has been discovered. The possibility that the relations that have emerged depend on specific circumstances in a given situation is reduced, if not completely eliminated. A hypothesis that originates not only from the results of previous studies, but also from a

theory that is formulated in more general terms, is free of the second limitation as well – that is, of isolation from general knowledge. If connections are satisfactorily established between general theory and specific data from a study, and if the results from the study confirm the idea on which the hypothesis is based, the creator of knowledge has provided a two-way contribution. First, the results of the study assist in confirming that the theory holds in this new case as well as in previous ones. Second, by applying general theory to the specific case, the creator of knowledge is assisted in interpreting the results, that is, gets an answer to why he/she has come up with the results in question. Whether the character of an expected relation can be expressed explicitly from the start (i.e.

be formulated as hypotheses at the beginning of the study) depends primarily on the state of knowledge of the study area being focused on. Studies for creating knowledge can either be based on well-formulated hypotheses or they can produce them. From this it follows that formulating and verifying (or falsifying) hypotheses is one goal of studies that proponents of the analytical view want to call creation of knowledge. This goal cannot be reached without effort. In several business areas significant hypotheses are missing, which means that many explorative studies must be undertaken before hypotheses can be formulated. Such explorative work is an inevitable part of any effort intended to develop better knowledge about companies and related social phenomena. It does not make sense to say that studies that start by formulating hypotheses are more

knowledge-creative than others. Just when it is appropriate to formulate such hypotheses varies with the nature of the problem and the extent to which knowledge of an area already exists. Formulating and reformulating questions to create knowledge is a continuous process. A hypothesis implies that a special characteristic or event (X) is one of the factors that determine another characteristic (Y). Studies designed to test such hypotheses must result in data that allow creators of knowledge to conclude that X does or does not help determine Y. This leads us to causality. The concept of causality was discussed to some extent in Chapter 3. A deeper discussion

than that is beyond the scope of this book. Causality is, incidentally, a very complicated philosophical concept and has long been an issue on the agenda in many philosophical discourses. “Common sense”, in the context of causality, usually follows the idea that one circumstance (characteristic or event = cause) always leads to another circumstance (characteristic or event = effect). In analytical research, however, it is more usual to search for a number of determining factors (not just one) that, taken together, make possible the

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

appearance of a characteristic or the occurrence of an event. Furthermore, basic and more advanced theoretical thinking may be focused on finding necessary and sufficient conditions for a characteristic or an event. However, the methodological issues are the same whether limited to one characteristic or event or considering several simultaneously, which means that our discussion can be based on the premise of one cause → one effect. A necessary condition (circumstance or event) is, by definition, a condition that must exist if

the phenomenon of which it is a cause is to appear or happen. If X is a necessary condition for Y, Y cannot appear or happen unless X exists or happens. A sufficient circumstance is one that always will be followed (contemporarily or later) by the phenomenon of which it is a cause. If X is a sufficient condition for Y, Y always appears or happens when X appears or happens. According to the analytical view in general, a condition can be both necessary and

sufficient for a phenomenon to appear or happen. If that is the case, Y will never appear or happen if X does not appear or happen. Furthermore, Y will always appear or happen if X appears or happens. Taken together, this means that Y appears or happens if, and only if, X exists or happens. However, in social sciences it is in practice impossible to find necessary and/or sufficient causal (extrinsic) relations. Social sciences usually go no further than presenting intrinsic or logical relations (compare the discussion in Chapter 3). The reason why the social science creator of knowledge cannot find causal necessary or sufficient relations in his/her reality is, according to the analytical view, understood to be that the social science creator of knowledge is not able, like the natural science creator of knowledge, to control his/her environment enough during various studies to be able to isolate what such causal relations might possibly be. Example of an intrinsic necessary relation: More than one company must exist in a market in order for that market to be competitive.

Example of a logical sufficient relation: If a machine breaks down, manufacturing done on that machine must stop.

A strict creator of knowledge would probably say that neither of the above examples is an explanation! However, to clarify, nothing in the analytical view in social sciences prevents them from generating causal relations. It happens all the time. However, they are never of a necessary or sufficient kind! As the reader has noticed, cause as well as effect can indicate a state (a characteristic) or a

course (an event). It is often arbitrary, or more the matter of how to interpret the meaning of a word, how we express it. As mentioned before, in the normal case a single cause and a single effect are not enough to

explain social events or phenomena. Creators of knowledge, therefore, try to find a group of causes that, combined, makes it more probable that an effect will take place or emerge. Conversely, most creators of knowledge search for several effects from a given cause. A usual additional complicating factor in business is that causes in a given situation interact

with each other. The cause–effect relation is rarely (if ever) one-dimensional, even though that is the ideal situation in the analytical view. Many analytical operative paradigms are

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

nevertheless set up as if this ideal were possible. Creators of knowledge then consciously get a bundle of causes (or effects) as the outcome of the study (see Figure 4.1).

Figure 4.1 The Normal Hypothesis

Most hypotheses in business (and in the social sciences in general) are associated with contributing or alternative conditions and prerequisites under which the hypothesis is valid. It is therefore usually not possible to demonstrate directly that a given characteristic or event (X), either by itself or in combination with other characteristics or events (A, B, C, etc.) will automatically lead to an Effect Y. Instead, creators of knowledge find themselves, from the data observed, having to draw the conclusion that the condition X is a condition of Y is (or is not) defensible with a specified degree of certainty (if the study is done in quantitative terms, this degree may be possible to state in statistical numbers). Three stipulations are, as mentioned in Chapter 3, necessary to draw a certain and safe conclusion that the causal connection proposed between factor X and Y in a hypothesis is valid.

1. Relation between the Factors. The first stipulation is that factor X and factor Y co-vary in the way stated in the hypothesis. Let us assume that we want to test the hypothesis that X is a contributing cause of Y. If we find that the number of cases that contain both X and Y are fewer than the number of cases that do not contain X but do contain Y, we draw the conclusion that the hypothesis is not defensible. Furthermore, if our hypothesis also specifies the amount of Y that is determined by X, we should also find that cases showing a higher amount of X also show a higher amount of Y. (Instead of a positive co-variation between X and Y it is possible, of course, to consider hypotheses that suggest that the existence of X prevents the existence of Y or, if the creator of knowledge wants to express it more quantitatively, that Y increases as X decreases – or vice versa. In this case, our discussion would be analogous.) Suppose we want to test the following hypothesis: large market share for a product results in

high profit from the same market. We reject the hypothesis if our data do not show that in more cases where a product has had a large market share, the profit is higher than among those cases where a product has had a small market share. Two alternative relational forms of X causing Y

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

may, however, be valid under these conditions: a. What we thought was a cause might actually be an effect (X could be an effect instead of a cause), and vice versa (Y could be a cause instead of an effect).

b. External or specific circumstances other than X could be the “real” causes of X being related to Y.

To return to our example, it might well be that high profit brings large market share instead of the other way around (see (a) above). It could also be the case that the cause of high profit as well as a large market share can be found in good senior management or that the study has looked only at industries that are particularly conducive to high profit no matter what. Other industries might have shown other results (see (b) above).

2. Chronological Order of the Factors. The possible alternative hypothesis – that high profit will cause large market share – leads us to the second stipulation necessary for saying that a hypothesis has been verified. A characteristic or an event cannot be considered a cause of another if it exists or happens later in time than the other one. A cause must always exist or happen before – or possibly at the same time as – an effect. Moreover, two factors can be each other’s causes and effects. The relation is then said to be symmetrical. In our example, it might be possible that large market share normally brings high profit. Yet this profit may be necessary for maintaining the large market share, by enabling the use of new resources. Even so, the chronological order itself, even if combined with an existing relation between

two variables (factors), is not enough for us to say that a causal relation has been found. A third prerequisite is necessary.

3. Elimination of Other Possible Causal Factors. In order to be able to say that X is a cause of Y, it is necessary:

that there is no background factor Z that can explain X as well as Y that there is no intervening variable W that gives a better picture of the causal relation.

The most serious of these is, of course, the background factor. If an intervening factor is present, it might simply improve our explanation. Examples of what having to deal with background and intervening factors could mean are illustrated in Figure 4.2 (compare Figure 3.5).

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

Figure 4.2 Importance of Background and Intervening Factors

Let us summarize the three requirements for drawing the conclusion that a hypothesis is an expression of a particular established causal relation:

1. An existing covariation. X (the causing or independent variable) and Y (the affected or dependent variable) shall vary in a way that is predicted in the hypothesis; X and Y shall tend either to assist or counteract each other.

2. Y shall not have existed or taken place prior to X. 3. It shall be possible to eliminate other factors determining Y.

It must be pointed out that even if these three requirements are met, they are not proof (in practice) that X actually causes Y. It is possible to have overlooked a decisive cause other than X; alternatively, X might cause Y only under certain circumstances that are not determined in the study. In other words, we draw the conclusion that it is probable that X causes Y, but we can never be absolutely certain. Further studies will have to disprove or support our conclusion; in practice, it will never be absolutely proven.

THE CREATION OF KNOWLEDGE PROBLEM

As the reader has probably understood, the analytical view is rather formalistic and full of rules. There are rules for what should be seen as reality, there are rules for what is science and what is not, there are rules for what should be counted as an explanation, etc. The reader will later discover that there are rules in the analytical view for how to incorporate, develop, and/or modify previous techniques, results and/or theories, that is, methodical procedures and rules for how an analytical study is designed, that is, methodics. One aspect of the relatively tight discipline which is associated with the analytical view, is

its attitude to what is a creation of knowledge problem. Generally, when possible, the analytical view recommends that already at the beginning of the process of creating knowledge – even before the project starts – the creator of knowledge should state the research problem, and, if applicable, also the practical problem, which is to be studied.

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

One may ask, of course, whether the creation of knowledge problem can always be described at the start of a study and, furthermore, whether it is always connected to a practical problem? Of course not! There are projects, which could be called explorative, which are designed to clarify a problem more than anything else. However, according to the analytical view, such studies should also be governed by a problem formulated in the beginning, even if a very preliminary one. As far the second part of the question is concerned, the creator of knowledge has the right, of course, to formulate problems, which have not yet been recognized in practice and/or concentrate on theoretical issues, which eventually will have a practical application in the future. Whatever type of analytical study is at hand, a problem should be guiding it, so the view in

question says. Also, there are several types of analytical studies in terms of what they aim for (this will be discussed further at the end of this chapter). Each of those should be introduced by a clear definition of the problem according to the analytical view. In this respect and with this view, there are basically three types of studies, that is, explorative, descriptive and explanatory.

The problem for the explorative study is to come up with a more precise statement of the problem, which can be brought forward in further efforts of creating knowledge. The problem of a descriptive study is to come up with as detailed a map as possible of some factive reality. The problem of the explanatory or hypothesis-testing study is to explain the connections between causes and effects of some characteristics or events.

The details of what is to be contained in these different formulations of a problem differ, but one thing is certainly common to all. Definitions of key terms and concepts should be given as early as possible in the study, according to the analytical view. The three studies above are of a more scientific type. The consultant and/or an investigator

outside the academic arena have, on top of these, almost always an ambition to suggest solutions to specific problems. If this is the case, in the ideal case, a formulation of a problem should answer the following questions:

What? Whom? How? When? Why? With what effect(s)?

For consulting and similar studies, it is possible to see a problem hierarchy in the analytical view. At the most general level, there is a management problem. The management problem represents a decision that a manager must make and is the problem prompting the research. At a more specific level, there is a research problem. A research problem is the single question

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

or hypothesis that best states the objective of the research study. A study, which answers these questions or attacks these hypotheses is supposed to provide the manager with the desired information. On occasion, it may be more than one question, but often, it is just one. At the most specific level, there is an investigative problem. The investigative problem contains those more specific questions that the creator of knowledge must answer in order to satisfactorily answer the general research questions. There is, of course, nothing preventing an original problem formulation in an analytical

approach to be modified as the study goes on, depending, perhaps, on the fact that new information is received or new experiences are gained. New directions in the original design of the study or even restarts may be necessary.

DEDUCTION, INDUCTION, VERIFICATION – AND ABDUCTION

The analytical view is always focused on facts. From this it follows that creators of knowledge in this view are first of all collectors of facts. Second, they attempt to describe, in general terms, both what facts they see and what facts they hear people talk about in terms of what they have seen and what they think. Third, they make a prediction based on their theories and explanations, which – again – can be verified against facts. The most characteristic aspect of the analytical view and its operative paradigms is that they

are cyclical. They start with facts and ends with facts, and the facts of one cycle are the facts that begin the next. Creators of knowledge regard their theories, however, as only probable and are therefore always prepared to give them up if the data that emerge are not consistent with the predictions based on the theories. If a series of observations developed to verify certain predictions forces them to abandon a theory, they look for new and better theories. Because in the analytical view they expect the progress of knowledge (gained through scientific or other means) to go on in a chain of successes, they expect that this cyclical process will continue. See Figure 4.3 opposite for an illustration. The horizontal dashed line in Figure 4.3 separates the empirical world (the factive world)

from the theoretical world, which, in accordance with the analytical view, is more or less quantitatively constructed. In the world below the line we see creators of knowledge who, for instance, look into microscopes or observe behaviour or interview people. In the world above the line, we see a chain of mathematical formulae, logical sequences and systematically arranged data. We also notice three steps in the figure. The first step involves moving from original observations towards theories. This is known as induction – a way of creating knowledge whereby creators of knowledge conclude general laws from individual cases, that is, construct theories using factive knowledge. This can mean, for instance, that a mathematical formula is found that is perceived to suit the facts that a creator of knowledge is trying to incorporate in the theory. Then the creator of knowledge asks, “Is this really what I want?” The creator of knowledge is then forced to go back to the world of facts (factive reality) to verify his/her construction.

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

A creator of knowledge cannot, however, check on a general law without first asking what it says about specific cases. The course of the stock market, for instance, cannot be observed indefinitely. The creator of knowledge may perhaps observe that average prices rise one day, that they rise the following day as well, and then, after one more day, fall, and so on. Any limited number of cases can be observed. Creators of knowledge using the analytical view must therefore decide what their general theory says about tomorrow’s development, for example. This step is reached through deduction – a way of creating knowledge whereby the creator of knowledge infers single cases from general laws; that is, a logical analysis of what general theory says about a specific event tomorrow. Not until then will creators of knowledge be ready to return to facts and find out whether or not their predictions were correct. This third and last step is the verification of the theory.

Figure 4.3 Cyclical Nature of Creating Knowledge in the Analytical View

So, in principle, the deductive mode means that the creator of knowledge goes from theory to facts and the inductive mode means that the process goes the other way around. The most common criticism directed at the pure deductive approach is precisely its dependence on theory, and the influence on sampling and operationalization of variables in the models being constructed that follow. There is a risk that those observations which are made will not fit that reality which they are meant to describe. This is a criticism of the deductive approach which is directed by Glaser and Strauss (1967), who with their so-called constant comparative method, based on an analytical view, recommend induction. However, in their version of methodology, they claim that the objective is not verification but to generate a grounded theory (more about grounded theory can be read in the Appendix): “Generating a theory from data means that most hypotheses and concepts not only come from data, but are systematically worked out in relation to the data during the course of research” (Glaser and Strauss, 1967: 6). Apart from induction and deduction, sometimes a third method is mentioned in this context,

that is, abduction (Peirce et al., 1998). Abduction means that a single (often surprising) case is placed in a general hypothetical pattern, which, if it is true, will explain the case, in question. The explanation should then be confirmed by new observations (new cases). The method then becomes a kind of combination of induction and deduction, but it also brings on new steps.

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

Another way to put it is to say that induction and deduction can be seen as two alternative research strategies, while abduction is more of a research tactic. Induction starts from facts and deduction from theory. Abduction starts from facts like induction, but it does not turn away theoretical knowledge and is, to that extent, closer to deduction. The analysis of facts may very well be combined with, or preceded by, studies of existing theory in the literature, not as a mechanical application to single cases, but as a source of inspiration to discover patterns for further explanations. This means that the research process consists of alternations between (existing) theory and facts, where both are seen in the light of each other. The single models of explanations, that is, deduction and induction, are then seen as too one-dimensional and unrealistic, compared with how research is done in practice.

OPERATIONAL DEFINITIONS

All creators of knowledge must use a set of concepts when they organize their data so as to discover relations. A concept is an abstraction of observed events or characteristics, a short verbal representation of them. Some concepts may be relatively easy to relate to the phenomena they attempt to represent – such concepts are price tag, factory or machine. It then becomes possible to point out the corresponding phenomenon in factive reality. Phenomena represented by other concepts, however, can be more problematic to point out, observe or measure. These concepts are more abstract, more distant from the empirical world. Examples of such concepts might include creativity, image and organization. These more abstract concepts are sometimes called constructs. The closer a concept is to factive reality, the more operational the concept is said to be.

Because the analytical view is, to a large extent, occupied with creating knowledge of events, characteristics and behaviour in this reality, the creator of knowledge tries as much as possible here to formulate operational definitions, that is, concepts that in themselves contain a description of how a specific phenomenon is to be discovered in factive reality. Using the rules the definition provides, anyone should be able to discover the specific phenomenon in the reality represented by the definition. An operational definition should contain the following parts:

1. A statement of which object(s) or subject(s) are to be observed or questioned. 2. A description of the situation in which the observation or questioning is to take place. 3. A determination of the type of measuring scale to be applied to the observations made or

answers obtained. 4. Rules for how to handle the data obtained through the observation or questioning.

Here are two examples of operational definitions:

Level of stock: Average book value of a company’s stock during a year, calculated as the arithmetical mean of the public figures that can be read in the company’s accounts in the beginning and at the end of a given financial year. Improved product quality: More yes-

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

answers than no-answers to the question: “Do you think the quality of Product X has improved since the last time you used it?” The answers should come from a random sample of at least 100 customers among those who have actually bought the product in question at least twice. The latest purchase must have taken place within the past 2 months. Each customer is to answer individually and independently from other customers. The survey can be conducted by phone or by mail, and the questions are to be posed by an impartial third party with no connection to the company that manufactures and markets the product in question, such as an employee of an independent marketing research firm.

THE CONCEPT OF ANALYSIS

We have given the name “analytical view” to the methodological view described in this chapter. It is therefore reasonable to justify the label of the view and to clarify the meaning of the concept of analysis. In our everyday language we make a distinction between analysis and construction. Without

going into a detailed and sophisticated discussion of the demarcation of different schools of philosophy, we can say that this distinction exists there as well. Analysis, then, means a thorough investigation of an existing situation. Construction has to do with putting together a new (and, perhaps, better) situation. (There is, in fact, a philosophical school, which is called the analytical school, a school, which the analytical view leans heavily on and which is, therefore, another reason for our name “the analytical view”. We will return to this shortly.) The analytical view, like the systems view, contains the concept of “analysis”. The concept

is, in fact, central to both. However, the analytical view, in its analytical concept, is oriented primarily toward the techniques used by the creator of knowledge, whereas the systems view is oriented mainly toward the results that can be employed by the user (the one who may use the results provided by the creator of knowledge). Analysis is therefore of primary (general) interest to the creator of knowledge representing the analytical view, whereas it is of only secondary (or rather of starting and finishing) interest to the creator of knowledge representing the systems view. The concept of analysis also has different meanings in the two views. In both cases,

however, analysis means a specific treatment of factive reality (or a model thereof), the existence of which is acknowledged both by those who represent the analytical view and those who represent the systems view. Factive reality, according to the analytical view, is best clarified by what can be characterized as disintegration or decomposition, that is, by studying its smallest components by themselves. The sum of these partial results becomes knowledge of the whole. Analysis as seen by the systems view (systems analysis), however, means that any part of the

factive reality is best clarified by studying components in relation to the whole. Yet the sum of the study of these individual components in relation to the whole does not constitute knowledge of the whole. In addition, it is possible to talk about the totality as such (the system) – for example, in relation to its environment (placing the system in its larger context). We return to the systems view in the next chapter.

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

The analytical view is therefore based on a very important theoretical assumption: a. That the elements of an object or subject under study can be regarded as relatively independent of each other.

And a practical assumption: b. That the intrinsic properties of the object or subject under study does not change rapidly or in such a way that it is not possible to explain this object or subject by explaining its separate elements.

The systems view denies assumption (a) and sometimes assumption (b) as well. But both the analytical view and the systems view make the following assumption: c. That the analysis of an object or a subject can be done without influencing it in a manner that cannot be controlled.

Because analysis plays such a central role in the analytical view (plus the fact that “its” philosophy is the analytical philosophy), but only constitutes one part of the systems view, we see the name of the former view as totally adequate. Besides, the term analytical view is not our own. It was used more than fifty years ago by those who “launched” the systems view, and it is still used today in the sense of “the analytical view” versus “the systems view”.

CETERIS PARIBUS

The assumption in the analytical view that elements of the objects and subjects being studied are relatively independent of each other makes it possible for results that are developed from various operative paradigms of the analytical view to be formulated “ceteris paribus” (as if “all other things remain equal”, i.e. unchanged). Among other activities, sampling, hypothesis testing and theory construction in the analytical view are based on this assumption. In consequence, causal relations that have developed are considered to be valid provided “nothing else has changed”. Study the following three statements:

1. If the price of a product is falling, the demand for that product will tend to increase. 2. If the price of a product is falling and nothing else changes, the demand for that product

will increase. 3. If the price of a product is falling and the product is sold in a situation of perfect

competition, the demand for that product will increase.

Case 2 is what we mean by ceteris paribus. This is the usual case for business theories in the analytical view (although Case 3 could also exist). The systems view might accept Case 1.

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

THE PHILOSOPHICAL FOUNDATION OF THE ANALYTICAL VIEW

The philosophical foundation of the analytical view is definitely analytical philosophy. Just like any part of philosophy, explaining what analytical philosophy is isn’t straightforward. However, providing such as explanation is not the intention here. Rather, we want to bring up some points that have been raised by analytical philosophers over the years, points, which by no means have been left unopposed and some which are even not valid today, but they provide a kind of panorama which still flavours the analytical view today:

About philosophy: Can philosophy be done the way that science is? Can more rigor be introduced into philosophy by proceeding in a logical fashion? Can philosophy be reduced to logic?

The structure of reality is essentially the same as that of mathematical logic, that is, atomistic. Studying mathematics can tell us more about reality. Science is to be advocated as the only way to truly learn about the world (this is sometimes referred to as positivism, a stream which gave a start to analytical philosophy, but is no longer seen as of stand-alone interest within this philosophy) The principle of verification: only those propositions that can be verified are meaningful. An interesting statement can always be reduced to observation statements.

Analytical philosophy is largely a twentieth-century story. Famous analytical philosophers during the last century include Russell (1872–1970), Moore (1873–1958), Carnap (1891– 1970), Wittgenstein (1889–1951), Ryle (1900–1976) and Austin (1911–1960). Analytical philosophy is active today in many areas, particularly the philosophy of mind.

What we can call meaningful, how we find out about the factive world and what kind of facts it is composed of are central questions for analytical philosophers today. More about analytical philosophy (and positivism) can be found in the Appendix.

THE RELATION OF THE ANALYTICAL VIEW TO ITS PARADIGM

What the analytical view stands for today can be summarized in terms of the description of paradigm provided earlier in the book:

1. Conception of reality 2. Conception of science 3. Scientific ideals 4. Ethical/aesthetical aspects

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

Let us look at business through “the lense of the analytical view”.

Conception of reality, or What Does Business Reality Actually Look Like? Business reality exists objectively in itself, independent of us as theorists and/or creators of knowledge and it exists subjectively as relatively stable opinions of, and beliefs in, this objective world. Such objective and subjective parts of reality can together be classified as factive. Even if we, through interference or through change of mind can make this reality appear in various forms, it is basically a stable construction. This basic construction has a summative character consisting of explanatory links that can be studied relatively independently of each other.

Conception of science, or How Do We Look at Business as a Science? Business as a science (and as it is applied by researchers, consultants and other kinds of investigators) is a picture of business reality. This picture consists of more or less definite facts that can be discussed in terms of “all other things remaining equal”. This picture will improve as time goes on and as more knowledge is created. Theories (as language) can be discussed with regard to how they are constructed (syntax),

in relation to what they stand for (semantics) and in relation to their users (pragmatics). Because theories should be based on facts, their terms (e.g. cost accounting, budget, matrix organization, long-term planning and advertising) should be as syntactically and semantically correct as possible, yet – as far as possible – be made independent of pragmatic controversies. It is therefore natural that the terms are, to a large extent, defined operationally, based on logics and mathematics. Business knowledge is logically separated from moral matters and ethics. This means, for

instance, that business can argue about which consequences different alternative actions will have in terms of formulated objectives, but deciding which objectives should actually be chosen is beyond its domain. According to analytical creators of knowledge, science (this pertains to business (practice) as well) should not take a definite position on which business objectives to choose, and related matters. There is also a sharp boundary between philosophy and creation of knowledge. Philosophy

can be used as a basis for discussion, but it can never provide the truth. That is the task of the creator of knowledge as researchers, consultants or other kinds of investigators.

Scientific ideals, or What Do We Want From Business as a Science? Gradually, the study of Business will fill in as many “empty squares” in our knowledge of business reality as possible. This knowledge will consist of verified assumptions (hypotheses) in factive reality, assumptions which follow certain formal rules. Knowledge emerges most surely with the use of well-tested techniques, many of which have been picked up from the natural sciences. The results will be an increasing number of, and ever more refined, logical models and representative, generalizable cases. Even if we aim for explanations that are as simple as possible (but nevertheless true), we

know that generally the more causes we find, the more complete the explanation of a given effect will be. We will continue to get increasingly more deterministic relations, but in the meantime we will have to be satisfied with results that are sometimes stochastic.

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

Business knowledge is to be used to make better and better predictions of the consequences of various alternative actions that could be taken, and consequently to steer BUSINESS reality in a desirable direction.

Ethical and aesthetical aspects, or What Can and Should Business Creators of Knowledge Do, and What Attitude Should They Take Toward What is Done? These are relatively uninteresting questions to proponents of the analytical view. It is instead seen as important to regard creation of knowledge as progress; as business creators of knowledge we do not have to take responsibility for how people in business reality use the knowledge that will be presented as we go on.

We present further thinking associated with the analytical view under the headings “Discovery and explanation”, “General theses” and “Difficulties in relation to explanation”. Before we do so, however, we remind the reader of the outline of the book. Our intent, as far as possible, is to present each of the methodological views on its own terms. If you believe in the analytical view, you accept Chapter 4, and so on. This also means that the descriptions in Chapters 4, 5 and 6 are in several basic respects often not comparable. It is important to remember that the different views are based on their own paradigms.

DISCOVERY AND EXPLANATION

From a methodological point of departure, the analytical view treats all social sciences as one. The reason is that all of them relate to humans and that they all use the same explanatory principles (if a creator of knowledge in that science has an analytical orientation). The view also often sees classical natural sciences as an ideal to emulate, because the

natural and social sciences are seen as having similar purposes, if not the same results. If an objective of creation of knowledge is to establish various general relations among phenomena in factive reality, if the test of veracity of relations relies ultimately on data, and if these data are collected in factive reality, then, according to the analytical view, you are doing science. All social sciences, basically, meet these expectations, according to this view. The differences between the classical natural sciences and the social sciences (to which

business belongs) are then more a matter of degree than of kind. Being engaged in the work of creating knowledge means, according to the analytical view, that you always face the same basic problem: to discover and to explain. In certain respects, the social sciences have not reached as far as the more exact sciences.

Furthermore, if some of the social sciences have established less reliable answers than others, this is not because the creators of knowledge were less ambitious or less intelligent. Nor is the reason, essentially – according to the analytical view – that less successful subjects are younger. Nor could the situation be explained by different methods being used in different areas, because the methods in analytical operative paradigms are the same. So, the differences are not caused by level of ambition, intellect, the age of the subject or the methods employed; they are caused by the built-up (complexity) of the problem being approached.

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

According to the analytical view, every science has two basic levels of ambition: to discover and to explain. To discover can mean either to explore or to describe. Descriptions and explanations, in turn, may be used to forecast and to guide. The first (discover) determines whether the study area is to be called an area of interest to science, the second (explain) the extent to which we are entitled to call the science successful. To discover is either the same as developing problems of creation of knowledge or, if they already exist, to show the characteristics and the behaviour of objects and subjects in reality. This can be called discovery because the problems or the characteristics or behaviour might be unknown (at least in this format) until the creator of knowledge reveals them. To explain is to present more or less general relations among characteristics, behaviours, or both. Explanations in this sense are far more common in natural sciences than in social sciences, which some claim is a striking difference between the two. Descriptions and explanations are invaluable results of all creation of knowledge efforts

conducted according to the analytical view. Such results, that is, empirical theses, are more or less generalizable. So that they are not misunderstood, they should contain operational definitions as far as possible. Otherwise there is a risk that, for instance, explanations might be based on concepts that are not related to factive reality. Over the years, the social sciences have, through their use of the analytical view, put

together a huge number of empirical theses, some examples of which are given at the end of this chapter. If the first task of a science is to establish generalizations, the social sciences have established quite a few. Yet these generalizations are neither so general nor so exact as those in the biological or technical sciences, for instance. Among other things, the results are valid primarily for the industrialized part of the world, and they are stated with a number of provisos. It is difficult to find strict explanations in business. This is partly because we cannot easily

manipulate the objects and subjects being studied or the circumstances around them. For instance, it is difficult (even unethical) to experiment with people, compared with experimenting with things. Furthermore, people cannot be exposed to the same treatments or conditions as when testing new situations among animals and physical objects. This is also the reason why the creator of knowledge using the analytical view in a business context applies a number of statistical techniques as a control mechanism, even though they are considered to be less reliable than physical control. The most important task for business when creating knowledge with the analytical view is

to find explanations. Explanations are based on discoveries (formulations of problems and descriptions). To explain means to answer the question “Why?” Explanations should not only be directly supported empirically, but also indirectly follow as a natural consequence of existing theory (be deducted). This is so whether interest is directed towards a single phenomenon or a more generalizable thesis. It is in the area of finding explanations that the social sciences face extra problems.

GENERAL THESES

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

The longer a science has been on “the road of explanations”, the greater the number of general theses it tends to contain. It is unfortunately true that business contains a relatively small number of such theses under the analytical view. It is also true that what can be claimed, as belonging to the subject of business is rather

vague. What should (or should not) be included (or omitted) from business is primarily determined by the problem: to occupy oneself with business. Often, therefore, results from such related subjects as Economics, Sociology and Psychology are used. From this point of view, the situation is considerably brighter. Acknowledging that they contain the same principles would, according to the analytical view, promote the development of knowledge. Business would, in this case, be further stimulated by the larger number of theories, concepts and methods.

DIFFICULTIES IN RELATION TO EXPLANATIONS

So, the difficulties of business are related more to finding explanations than to making discoveries. Business, in this context, operates (like other social sciences) at two levels at the same time.

It contains empirical theses about individuals (often collected from psychology), but also of aggregates of individuals – groups, organizations, classes and societies. The relationships between these two levels are not simple. In both the analytical and the systems view we treat organizations as if they were objective entities. This, of course, is only a simplification of factive reality. If we want to explain in greater detail why an organization behaves the way it does, why it behaves differently from other seemingly similar organizations, we would probably have to consider its individuals (although external circumstances could also explain different behaviour, of course). But this does not mean leaving the analytical view for the actors view. Instead, this can take place through what in the analytical view is called a reduction, in this case a reduction of organizational theory theses to psychological ones. There are difficulties associated with such reductions in business. There are certainly many

relatively rested theses in psychology, but they are only statistically true and they usually have a very limited scope. In business, for instance, we often find that a thesis that is valid under one set of circumstances is not valid under another set. When this happens, it is more likely that another thesis, sometimes the opposite, is valid. According to the analytical view, business does progress in its search for explanations.

Even so, there is still a gap between general business theses and the numerous empirical results that exist. This distance will most likely remain into the foreseeable future. One problem is that new (general) explanations are more difficult to find, because the new hypotheses (suggested explanations) that are formulated do not (yet) have a solid theoretical support on which they can be based.

EXAMPLES OF THEORETICAL RESULTS Arbnor, I., & Bjerke, B. (2008). Methodology for creating business knowledge. ProQuest Ebook Central <a onclick=window.open('http://ebookcentral.proquest.com','_blank') href='http://ebookcentral.proquest.com' target='_blank' style='cursor: pointer;'>http://ebookcentral.proquest.com</a> Created from indwes on 2021-03-16 10:02:01.

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

Let us point out a few examples of theoretical results from the analytical view. The results presented here are not meant to be in any way verified as absolutely true according to the basic idea behind the analytical view, even less to be inclusive. The results are presented only to give the reader an idea of what kinds of theoretical results can be developed by starting from the analytical view, how they are usually formulated and at what level of general applicability they are normally kept. The examples are from an area that we could call “Organizations”.

Example 1.

The larger, the more complex, and the more heterogeneous the society, the greater the number of organizations and associations that exist within it. 1.1. Organizations tend to call forth organizations. If people organize on one side of an issue, their opponents will organize on the other side.

1.2. There is a tendency for voluntary associations to become more formal.

Example 2. The larger an organization becomes, the more levels of personnel it will tend to have.

Example 3.

A period of innovation and change affecting an organization is likely to produce a heightened amount of communication among the members, communication oriented both toward the task and toward mutual emotional support.

Example 4.

Other things remaining equal, the more friendly and helpful the boss,

the lower the absenteeism the higher the productivity the more likely that subordinates feel that the organization’s requirements are reasonable, and the more willing they are to accept changes in organizational practices the better liked the leader is the more strongly the subordinates associate with the organization the less tension there is within the organizational unit and the greater the internal cohesion the higher the subordinates’ morale

Example 5. The leader’s style of leadership tends to be influenced by the style in which he/she was led.

Example 6. The more closely a member holds to the organization’s professed values, the more likely he/she is to be promoted within the organization.

This chapter has only discussed the analytical view, the oldest of our three methodological views. As with all methodological views, it is greatly concerned with concepts. The concepts of the analytical view have consequently been presented in this chapter, including its fundamental concept of “ceteris paribus”. Also presented here is some basic thinking in

C o p yr

ig h t ©

2 0 0 8 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

the analytical view, like having definitions as operational as possible and regarding steps in creating knowledge, like deduction, induction, verification and abduction, as distinctly different from each other. At the end, the relation of the analytical view to its paradigm and some typical results of knowledge being created from this view have been provided.

POINTS OF REFLECTION

1. The analytical view talks about invariance. Explain this concept for your fellow students!

2. What is determinism? 3. Formulate what you consider to be three well-formulated hypotheses. 4. Think about and describe a cause/effect relation consisting of a main cause and at least

two contributing causes! 5. What does it mean to conduct an explorative study? 6. In the analytical view deduction, induction, verification and abduction are some

central concepts. Without any problem you may surely explain the first three concepts, but what is abduction?

7. To operationalize something means ......... What? 8. Ceteris paribus is Latin, but what does it mean? 9. What would your story look like if somebody asked you to present the relation of the

analytical view to its paradigm?