Sociologists make efforts to cognize social reality – an object which can be hardly formalized – which distinguishes itself by complexity, variety and diversification and as a whole uncertainty. Uncertainty – is a quality of an object, first of all, complicating the fulfillment of scientific formal procedures. At the same time uncertainty is the quality that is inevitable and even necessary for science, though not absolutely obligatory. Uncertainty in Sociology is eliminated by statistical accuracy and thoroughness of collecting of the observation materials. But it should be noted that accuracy is not always the most important quality for the sociological research, sometimes in the research there is too much numeric data.

Scientific observation of social reality is a sequence of logically combined successions, everything that sociologists define in social continuum has infinite and uncertain borders. Max Veber's ideal types contain "imperfect" signs of social phenomena, which they explain. It’s impossible to define clearly the bounds of social position or status, the characters of social roles, etc.

We don't mean that it is the uncertainty of a social fact as a part of social reality itself that makes the meaning of the term uncertain. Everything that we can find in the sphere of empirical facts we can reflect in definitions with which we define social facts. We can always define the bounds of uncertainty where statistics doesn't give solid bases. The thing is that social facts are infinitely uncertain in themselves. No matter how clear is the measurement unit which is used, the more subtle distinctions appear through the conceptual network. The more differences we find the more possibilities we create for appearance of border cases. As a result the researcher has to limit the scale of distinctions' differentiation. This is connected with the researcher's subjective understanding of the research goal, but that contradicts the objectivity of existence of social reality. That's why sociologists will always be unsatisfied with researches of predecessors and even contemporaries.

One would think that uncertainty will get away if sociologists use the term or definition which verbalizes a social fact. But the most captious question in sociological disputes is: how we define and what is behind the scientific term. The fact that all sociological terms are uncertain to some extent and that they have to be such is evident from the way by which we learn to understand and use the language. To do all that effectively we must understand the similarity and draw conclusions from similarities. Understanding of similarities is always a question of extent. The history of use of any term in Sociology sooner or later leads to a situation when there appear doubts in accuracy of its usage. These doubts appear not because we are uncertain about the facts which empirically "support" the term, but because there appeared and became stronger the factors which had been poorly and not very clearly expressed before – the uncertainty factors. You can day after day watch the colleague who loses hair but it is always remains uncertain when he becomes bold.

Uncertainty of sociological terms is not the result of the fact that we fail to solve the problem of "where to draw the line". We can't solve this problem in reality a priority and forever. Such "bounds" we can imagine (though with difficulty and not forever) for reality in nature, for social reality it's unacceptable. The fact is that here the bounds of terms are drawn conventionally, they are not drawn from the objective specific features of reality. These bounds appear due to solution of some real task, specific in each social situation, if the goal is achieved these bounds can be reviewed. Achievement of social goals, social decisions can never be good or satisfactory, this is always a balance, as for example function – disfunction of the social institution of Merton. More than that, neither decision can anticipate the needs and tasks in future. This is another reason why there are no strict bounds in Sociology. Moreover, sociological terms which are not clearly defined can be combined together in such a way that they acquire the precise meaning. But the change in rendering of one term mixes all the conceptual system.

Sociological terms even taken without any procedure of proof and substantiation have a quality of openness and incompleteness as a result of inevitability of boundary cases. In any textbook we can find an attribution judgment that social matter is multilevel and multidimensional. But this peculiarity of work with empirical facts doesn't suppose that our thinking should remain uncertain.

Here the law of Aristotle should come into operation, which says that the best position is the middle position when the extremes are cut out. It is also important to remember about the constant danger for sociologists – the danger of mixing the reconstructive logic of a scientist with practical logic of social practice. More than that, sociologists often project their own logic to the world of objective social phenomena. Evolutionary sampling of every unit of social life shows what it is, to which class or type, which we can define, it belongs. The problem is how we make a decision about the classification. Social facts are: "know well to which class they belong". But the signal which they send us about their essence comes to us distorted. Each term is a ray of light turned to a screen of experience, but whatever we would like to enlighten something should remain in shadow. Uncertainties of sociological terms characterize the existence of multiplicity of meanings best of all.

It wouldn't be enough to say that uncertainty is characteristic for our understanding of terms' meanings, which form scientific conception. Conception consists of a couple of judgments based on understanding, and each of them brings some extent of uncertainty into conception. It is unlikely that someone can point out to the terms, even those tightly connected with the empirical facts in natural sciences, in which you could not find some extent of uncertainty.

But we don't support the intentional uncertainty which can be often seen in the modern sociological works. Sociologists are the victims of reconstructive logic which makes the research uncertain. We won't be able to control this quality till we lose the illusion to reach the exactness of calculations and certainty of terms meanings, characteristic for natural sciences.

For current state of Sociology the problem of a serious analysis of methods of the social practice itself and not only methods of research of social behavior of people becomes more and more essential. The main reason for that is that scientific language becomes more and more specialized and gets away from the natural language of life of an individual. That leads to absurd existence of two different words – the world of theory and the world of life.

In these conditions the development of methods of Sociology of practice may become the main mechanism of negotiation of alienation between people and the main form of existence in controversial world. By means of methods of Sociology of practice people learn about the practice of life of a holistic person (acting and feeling ), and knowledge is "humanized", because it is being represented as people's experience with their goals, needs and fate. Principles and analysis logics of these methods haven't been researched enough yet.

L.V. Yatsenko made an attempt to present a special theory of practice. She insisted that genuineness of a general theory which is built on the basis of methodical knowledge is realized through its effective practical use in the process of organization of practical activity [8, p. 44].

Representatives of Praxiology deal with the same problems. This is a complex discipline which synthesizes data from various sciences (organization theory, management theory, psychology of labor, ergonomics, innovation theory, etc.), arts and practical experience, which belong to the forms of labor organization and effectiveness of activity. From the point of view of Praxiology (this is an opinion of T. Kotarbinsky, its founder), method is "a way of fulfillment of some complex action by means of certain choice and placement of its components, which can be planned and used repeatedly" [4, p. 82].

Looking deeper into the essence of the problem it turns out that such generalization of practical methods is just external, and has a formal character. In fact, it reduces all practical methods to purely organizational, opposing methods of cognition to methods of transformation of reality, and as a result – of knowledge and prescript. Opposing the categories of "knowledge" and "prescript", "act of logic" and "logic of act", Praxiology overemphasizes them, it doesn't presume that knowledge can act in function of prescript as well as prescript in the function of knowledge, that these functions can overpass each other.

We can agree that the basis of dialectics of theoretical and practical is first of all an interconnection of descriptive and prescriptive (normative). Though descriptive knowledge finalized in theory, describing connections, relationships, laws, thereby prescribes a certain way of action with an object. But if knowledge is formulated in the form of method it's prescriptive function becomes clearly shaped, basic and specific.

In any kind of activity (scientific, technological, organizational, practical, etc.) method is always a special kind of knowledge about the means, ways, procedures, norms of activities, prescriptions and demands to a thinking and acting person, guided by which he acts and verifies genuineness of the rules he uses and correspondingly of the knowledge which is in the basis of them.

In reality: 1) scientific method is nothing but the theoretical generalization of practical methods and the latter is the specification of scientific methods; 2) methodology doesn't limit itself by studying either only methods of cognition or methods of practical activity, not to turn neither into natural philosophy nor into pure methodical instruction, which has only practical meaning; 3) one cannot join or oppose these methods of cognition and practice without a risk of mixing methodology and methods.

So method is always in the basis of interaction between theory and practice. Being a subjective alternative to objective regularities of social phenomena method servers not only to explain but also to change the reality in practice. Methods of Sociology of practice find in that their content and function, they express at the same time logic of cognition, logic of action, logic of creation, ability to "create the future". But it is not the same – to reveal their content according to the materials of sociological science or to show them as the forms of real, practical, creative activity of people. The latter is practically not studied in Sociology. So the task is not only to make our knowledge and theories adequate for the essence of subjects, but to make social reality adequate for people's needs and goals.

In modern research, autonomation and differentiation of practical and theoretical judgments is based on rejection of real meanings of estimates, imperatives, etc. But then it turns out that they can't be analyzed from the point of view of logic. The situation is paradoxical: the practice of use of norms and estimates in Sociology and social administration shows that there is a logical connection between prerequisites and consequences, but it is denied by the representatives of phenomenological Sociology on the basis of logic.

Using our methodological approach to practical social action we can speak about it as a synthesis of knowledge taken as a principle of "practical usefulness", "potential realization" and "physical realization", that is as a program of rational and effective practical activity. Transformation model of social activity practically coincides with the following types of social administration: 1) normative (as a correspondence of chosen goals to norms), 2) cost-based (as a relation of expenses to a supposed or real result), 3) resultant (as a relation of the result to the goal), 4) real (as a relation of need of possible norms to real possibilities and resources).