Stages of statistical research. Statistical observation is the first stage of statistical research, which is a scientifically organized collection of data on the studied phenomena and processes of social life.

Statistical research requires special training to conduct it at a high scientific level.

Statistical study- this is a scientific and organizational process in which a single program is monitored for certain phenomena and processes, the collection, registration of primary data, their processing and analysis.

Any research begins with taking into account the facts and collecting primary material, which, depending on the purpose and task of the work, can be versatile in meaning and methods of obtaining. For example, population censuses are needed to study the size and composition of the population. To study the spread of diseases, it is necessary to record and register individual diseases in medical institutions. It is possible to obtain systematic information about the activities of medical institutions only if they organize the proper type of relevant data. Therefore, the task of statistical research is the collection of objective, reliable and complete basic information.

The process of statistical research can be divided into stages:

Drawing up a plan for statistical research, developing its program;

Registration and collection of statistical material;

Development and summary of data;

Statistical analysis;

Implementation of research results into practice.

Plan and program of statistical research

Statistical research is always carried out according to a specific plan, which includes both programs and organizational issues, and is determined by the task of statistical observation, which should provide a complete and versatile description of the phenomenon under study. Thus, the preparation of a research plan involves the solution of a number of organizational issues that lie in the formation of the goal, objectives of the study, the choice of the object and unit of observation, the place and time of the study, the source of information, the form of practical implementation, as well as methods of statistical research.

Target statistical research answers the question "why study?".

It foresees the determination of the patterns inherent in the phenomenon and the connections of this phenomenon with others, the development of measures to reduce the impact of negative factors on health, the implementation of the results of work in the practice of health protection and measures aimed at improving the quality of medical care.

Task answers the question "what to do?".

So, for example, the task of a statistical study may be to study the level and structure of the phenomenon (morbidity, mortality) in certain population groups, the frequency of the phenomenon in groups that are influenced by various factors (environment, biological and social), the volume and quality of medical care for certain groups population.

When preparing an observation, in addition to the goal, it is necessary to determine what exactly is to be examined - to establish it an object, namely, a statistical set of persons or phenomena, which consists of units, facts that are to be studied. So, for example, it can be a set of individuals (sick, deceased), functional units (beds in a hospital, hospital), contingents that are characterized by certain phenomena (incapacitated workers), etc.

The object of statistical observation must have the boundaries of the population designated for study. For example, before conducting a statistical study, the activities of medical institutions must be determined, the activities of which institutions will be studied. They are regulated by the objectives of the study.

When studying the spread of diseases and mortality of the population, it is also necessary to outline the boundaries of this population, among which groups of the population this phenomenon should be studied. If the object and boundaries of the study are not precisely defined, then the data obtained will not give a complete understanding of the level and composition of the phenomenon.

When conducting a population census, the object of observation will be the totality of persons who permanently reside in a certain territory. At the same time, it is important to know who to enumerate: the population that actually lives in a given territory at the time of the census, or that lives permanently. Thus, it is important to know data on the size of the actual population for the organization of various types of services, including medical, and the population that lives permanently - to determine the composition of different contingents (for example, children of preschool or school age to determine the availability of their schools and childcare facilities) . Thus, the choice and purpose of the object depends on the purpose and objectives of the statistical study.

Simultaneously with the definition of the object, it is necessary to assign a unit of observation. An observation unit (accounting unit) is an integral part of the statistical population (an individual, an individual phenomenon), an integral element of an object that has characteristics that are subject to registration and study (gender, age, birth weight, length of service, result of treatment, time spent in hospital). hospital, etc.). It must be clearly defined: so in the study of diseases, the unit of observation can be like a sick person. So is a certain disease, depending on the tasks and objectives of the study.

When studying diseases according to the data of applications to outpatient clinics, only the initial visit is taken as the unit of observation. When determining the number of newborns, only living ones are taken into account.

However, sometimes there are special instructions for choosing units of observation. So, for example, the concept of dead birth is determined by special rules that define the terms "born alive and dead", or "born dead". The quality of the obtained materials and the possibility of their use for analysis depend on the correct choice of the research unit.

When creating plans for statistical research, not only the forms of accounting documents and the rules for filling them out are worked out, but also the questions of who will fill them out, control the correctness and completeness of the collected data, as well as other organizational and methodological issues related to the collection of statistical materials. Thus, at the first stage, executors are appointed, and the budget is approved.

Research methods (types).

Depending on the nature of the observation in time, there are current, periodic and one-time observations.

If the collection of material is carried out systematically, with constant registration of facts as they appear, then this will current observation.

If it is done regularly, but not constantly, then it will periodical observation.

Current statistical study- this is the identification of phenomena that change rapidly over time and are a continuous process that requires ongoing registration. This method determines the incidence of individual groups, the birth rate, the death rate of the population, etc.

Simultaneous Observations reflect the state of the phenomenon at a certain point in time, which is called the critical moment of observation. An example would be a population census or a census of people who went to a polyclinic at a certain point in time, a census of places, health care institutions, the timing of the work of doctors or paramedical workers, etc. Such observations show the statics of phenomena, the change of which over time is relatively free, If necessary, a combination of both forms of statistical research is used. Thus, data on the number and structure of health care institutions are collected using a one-time method, and on their activities - through current accounting.

From the point of view of the sufficiency (completeness) of taking into account the facts of observation, statistical studies are divided into: continuous (solid) and discontinuous (not continuous) (partial).

Continuous (continuous) research cover all units of observation that are part of the population under study (the main population). This is necessary if it is necessary to establish the absolute dimensions of phenomena (population, number of places with AIDS, etc.). Conducting such a study is a very cumbersome, economically unprofitable method that requires significant costs. The development of the material, of course, will require a lot of time, although, at first glance, the method is the most likely,

If a continuous observation is impossible or incomplete, then it is necessary to carry out discontinuous. It will not require a full account of all units of the population, but will be content with a certain part. When studying this part, it is also possible for the material to obtain generalizing conclusions, which, with sufficient probability, can be extended to the entire set.

Discontinuous research can be monographic, main array, selective.

monographic the description is used for a detailed, in-depth characterization of typical units of the population, for studying the development of an institution, the reasons that contribute to its success or cause shortcomings. A detailed description of the work of some typical or advanced medical institutions is important for the socialization and formation of elements of excellence and its dissemination.

Method usage main array allows you to study objects that concentrate more units of observation. For example, if it is known that the bulk of tuberculosis patients (80-90%) are treated at two specialized clinics in the city, then studies of the organization of medical care for these contingents are carried out in these hospitals. The shortcomings of the method are that some of the patients remain unexplored, and the results may differ from those obtained from the main array.

Selective called a study in which the characteristics of the entire set of facts are given according to some part of it, which is selected randomly or by certain criteria.

Sampling method, as one of the types discontinuous research is possible provided that the sample is representative of the main one in quantitative and qualitative terms, i.e., that the sufficiency of the number of cases to be taken into account is determined, and all the versatility of the phenomenon under study is created in the sample. In such a case, the results can be extended to the main population.

Representativeness sample group is achieved by the correct selection of units of observation. It is important that each unit of the entire population had the same opportunity to get into the sample population. In addition, its qualitative characteristics are important, which can be ensured typological choice method. Its essence lies in the fact that the entire population is divided into several groups of the same type, from which units of observation are selected. So, for example, when studying diseases of the urban population, it is necessary to single out territorial units (districts). In typologically displayed groups, the choice of units of observation can be made proportionally or disproportionately, according to the size of each group.

The choice of units of observation can be carried out by methods:

Random selection- draw, lottery, random selection, etc.;

Mechanical selection- according to a certain size of the population by a reliable principle (every fifth, tenth, etc.);

nesting- nests (groups) are formed from all sets, the most typical objects that are studied by a continuous or selective method;

Directed Choice, which consists in the fact that persons with the same experience, age or gender are selected, etc.

Often in sample statistical studies, complex methods of selection are used that provide a high probability of results.

Selective studies require less time, personnel, funds, they can have an in-depth program, which is an advantage over continuous research. The sample set will always differ from the main (general, exhaustive). However, there are methods that allow you to establish the degree of discrepancy between their quantitative characteristics and the boundaries of possible fluctuations in indicators for a given number of observations.

Sample size, i.e. the probability of the number of observation units for different selection methods is calculated differently. The main formulas are shown in Table 1.

Table 1.

Required sample size for some array generation methods

Legend:

n - required sample size;

σ - standard deviation (feature variability);

N- the size of the general population;

T - reliability criterion;

W- evaluation of the part;

∆ - marginal error.

The number of observations in this case plays an important role, the larger the number of observations, the more accurately the main population is displayed and the smaller the size of the probabilistic error. The methods presented allow us to select the necessary volume of observations for study with a sufficient degree of probability.

The repeated or non-repeated choice is determined by the possibility of multiple or single participation in the formation of sample groups of each of the observation units.

Thus, the selective method, with its proper organization and conduct, is the most perfect form intermittent observation.

Methods for recording and collecting medical and statistical information

In a statistical study, various methods can be used:

Direct registration;

documentary accounting;

copying;

questioning;

At direct accounting facts, the necessary statistical data are obtained by special accounting - inspection, measurement, weighing and recording on an individual observation card.

Documentary accounting as primary is based on the systematic registration of facts, for example, in medical institutions. Such data from various official documents are copied into the map for study.

copying data in the developed statistical document can be used, for example, to obtain information on the composition of persons seeking medical care, on the medical institutions themselves, their activities, personnel and other issues, respectively, of development programs.

The use of technical methods of accounting for medical information, its centralization optimizes the mechanism for its further processing and analysis.

The collection of medical and statistical information through a survey is carried out by expeditionary or correspondent methods, self-registration.

At forwarding method the researcher interrogates the patient and, from his words, independently fills out the research card, which ensures control over the correctness of the answers.

At self-registration the examined patient independently fills out the card.

At correspondent method the researcher sends out cards for examination with appropriate instructions for filling them out. Having completed cards (with answers to questions), the respondent sends them to the address of the researcher.

Questionnaire method used when it is impossible to directly observe the phenomenon under study. Questionnaires are sent to specific individuals, but their answers are incomplete and inaccurate. The disadvantage of this method is that the correctness of filling out the questionnaires depends on the understanding of the formulated questions.

So questionnaire method used as an auxiliary or in the absence of more reliable methods of obtaining data. Often it is useful in sociological research.

The choice of survey methods is determined by the tasks and the monitoring program. The expedition method is the most reliable, but it requires the most expenses. The self-registration method is less expensive, therefore it is used when it is possible to fill out the cards by the examined persons. This method is often used in censuses. The correspondent method requires the least cost, but the data obtained with its help is not always reliable. It can be used as an auxiliary, taking into account its subjectivity, inaccuracy.

Simultaneously with the development of methods for collecting material, preparations are being made for grouping and combining data,

Grouping in statistics is the division of population units into homogeneous parts with their inherent characteristics. Its tasks are to separate the studied facts into separate qualitative homogeneous parts, which is a necessary condition for determining generalizing indicators.

The plan of statistical research should provide for the groups into which the phenomenon should be divided. The significance of such a division of the population into qualitatively homogeneous groups lies in the need to show their peculiarity, connection with others, and mutual dependence. So, when studying the incidence of nosological forms, patients in these groups are qualitatively heterogeneous: children, youth, elderly people, therefore, each group of diseases must be divided into even more homogeneous ones - by sex, age, etc.

The principle of grouping statistical material should be determined by a doctor who is well aware of its methodological basis. The features of the population units that underlie the grouping are called grouped. They are variable (quantitative) and are quantified. Variable grouping is carried out according to the numerical values of the signs (grouping of patients by age, time of illness, stay in bed, children by body weight, height, etc.).

Qualitatively defined features are called attributive: division of patients by disease groups, population by sex, professions, etc.

When grouping by attributive features that do not have a quantitative expression, the number of groups is determined by the feature itself (gender, profession, disease).

When conducting statistical grouping, it is possible to divide a qualitatively homogeneous group (men) into age groups (according to variable characteristics) - this will be a combined grouping.

The choice of group features is based on three basic rules:

the grouping should be based on the most significant features that meet the objectives of the study;

when choosing group characteristics, it is necessary to take into account the specific conditions in which this phenomenon is realized;

when studying a phenomenon that is influenced by several different factors, grouping must be carried out not according to one, but several signs (combined).

Grouping is the basis for combining statistical material and, given the understanding of all the rules, allows you to draw the right conclusions and identify reliable patterns that are inherent in the population under study.

Grouping must be distinguished from classification, which is based on the division of phenomena and objects into certain groups, classes based on their typicality and difference. A qualitative sign is the basis of classification. Classifications are standard and unchanged over a long period of time, are determined and adjusted by state and international statistics authorities. Classifications are the same for any study and often form the basis of groupings.

At the first stage, programs for statistical observation, development and integration of statistical material, and data analysis are developed.

The observation program is a list of signs recorded in the accounting document that characterize each unit of observation. It must meet the following requirements: it must contain a list of only essential features that reflect the phenomenon under study, its type, features and property; wording accuracy and logical order.

Questions about the studied characteristics are resolved after the appointment of the unit, taking into account the objectives of the study. So, in the study of diseases, program features can be gender, age, bad habits, date of seeking medical help, work experience, place of work, etc.

Of great importance is the formulation of program issues, their clarity and unambiguous interpretation. They can be presented in the form of closed questions - alternative (yes, no), or with a choice of three or more answers. To an open question (“Tell me your comments about the work of the department?”), the respondent can give any answer.

In order to ensure unambiguous data that is recorded for each unit of observation, the observation program is issued in the form of a record document. When conducting a statistical study, the sources of information may be official accounting or specially developed accounting documents.

If the research program does not go beyond the boundaries of existing official reporting and accounting documents (statistical coupon for registering the final diagnosis, medical death certificate, outpatient coupon, etc.), then after being developed for writing a report of a medical institution, they can be used for applied statistical research.

If the research program requires the receipt of materials that are not in the official accounting documents, a special accounting document is developed. It may take the form of a form, questionnaire, card, or be recorded in a computer database. Signs are entered on a map or in a computer database, which are recorded for each unit of observation: data of one newborn or deceased, one patient, etc. The list documents (journal, statement, account book) contain data from two or more units of observation, which are located in separate his ranks. Individual account documents may have more questions than list documents. Therefore, with card or computer forms of accumulation of material, its integration is facilitated, and the development is carried out with a more in-depth program.

Development program (associations) - addition of table layouts.

The association can be centralized - all primary materials are sent for processing to one analytical center, decentralized - processing is carried out locally.

The aggregation is carried out in the form of statistical tables, which are filled with the data of the combined statistical materials. The received statistical data should be checked beforehand.

Statistical tables- this is a form of systematic, rational and visual presentation of digital material that characterizes the phenomena and processes under study.

The table has a common title at the top. It briefly indicates its essence, time and place of obtaining data. The statistical table should also contain data on the numerical measurement of the phenomenon under study (%, abs. numbers, etc.) and the calculated totals of the studied features.

The statistical table has a subject and a predicate. The object of study is called the subject. This might be unit of the statistical population, or their groups (diagnoses, types of diseases of the population by age groups, etc.). The predicate of the statistical table can be a list of quantitative indicators that characterize the object of study, that is, the subject of the table. The names of units or groups (subject) are made on the left of the table, and the name of the predicate in the headings is a column. In the upper part, above the table heading, their numbering is given (table 1,2,3...).

The statistical subject is divided by horizontal lines into glades, the statistical predicate - by vertical lines into graphs. The intersections of horizontal and vertical lines form cells in which digital data is recorded. Horizontal rows and vertical columns of numbers, and their result must have the same number in the cell in the rows of predicates. In the names of tables, rows and columns indicate the unit of measurement.

Table layouts can be developed, when data are given separately for each feature. Then, based on them, they are formed analytical tables, which generally presents data on group characteristics.

There are the following types of statistical tables: simple, group, combined.

simple table- numerical distribution of data on one attribute,

There are no groupings in such a table; it does not characterize the relationship between features. A simple table provides little information, although it is clear and fast for analysis. An example of a simple table can be Table 2.

To get an idea about a particular phenomenon, to draw conclusions, it is necessary to conduct a statistical study. The subject of statistical research in health care and medicine can be the health of the population, the organization of medical care, various sections of the activities of medical institutions, environmental factors that affect the state of health.

The methodical sequence of performing a statistical study consists of certain stages.

Stage 1. Drawing up a plan and program of research.

Stage 2. Collection of material (statistical observation).

Stage 3. Material development, statistical grouping and summary

Stage 4. Statistical analysis of the phenomenon under study, formulation of conclusions.

Stage 5 Literary processing and presentation of the results.

Upon completion of the statistical study, recommendations and management decisions are developed, the results of the study are put into practice, and efficiency is evaluated.

In conducting a statistical study, the most important element is the observance of a strict sequence in the implementation of these stages.

First stage statistical research - drawing up a plan and program - is preparatory, at which the purpose and objectives of the study are determined, a plan and research program are drawn up, a program for summarizing statistical material is developed and organizational issues are resolved.

When starting a statistical study, it is necessary to accurately and clearly formulate the purpose and objectives of the study, to study the literature on this topic.

The goal determines the main direction of research and is, as a rule, not only theoretical, but also practical. The goal is formulated clearly, clearly, unambiguously.

To disclose the goal, the research tasks are defined.

An important aspect of the preparatory phase is the development of an organizational plan. The organizational plan of the study provides for the definition of the place (administrative-territorial boundaries of the observation), time (specific terms for the implementation of the observation, development and analysis of the material) and the subject of the study (organizers, performers, methodological and organizational leadership, research funding sources).

Pl a n research d ov a nia includes:

Definition of the object of study (statistical population);

The scope of the study (continuous, non-continuous);

Types (current, one-time);

Ways to collect statistical information. Research program includes:

Definition of the unit of observation;

List of questions (accounting signs) to be registered in relation to each unit of observation*

Development of an individual accounting (registration) form with a list of questions and features to be recorded;

Development of table layouts, in which the results of the study are then entered.

As accounting forms, accounting medical forms used in the practice of medical institutions can be used.

Other medical documents (case histories, and individual cards of an outpatient, the history of the development of the child, the history of childbirth), reporting forms of medical institutions, etc. can serve as sources of information.

To enable the statistical development of data from these documents, the information is copied onto specially designed accounting forms, the content of which is determined in each individual case in accordance with the objectives of the study.

At present, in connection with the machine processing of the results of observation using a computer, program questions can be formalized , when questions in the accounting document are put in the form of alternatives (yes, no) , or ready-made answers are offered, from which a specific answer should be selected.

At the first stage of the statistical study, along with the observation program, a program * of the summary of the data obtained is compiled, which includes the establishment of the principles of grouping, the selection of grouping characteristics , determination of combinations of these signs, drawing up layouts of statistical tables.

Second phase- collection of statistical material (statistical observation) - consists in the registration of individual cases of the phenomenon under study and the accounting signs characterizing them in registration forms. Before and during the performance of this work, instruction (oral or written) of the observers is carried out, and they are provided with registration forms.

In terms of time, statistical observation can be current and one-time.

At current observation Yu denia the phenomenon is studied for some separate period of time (week, quarter , year, etc.) by daily recording of the phenomenon as each case occurs. An example of a current observation is accounting for the number of births , dead, sick , discharged from the hospital, etc. This takes into account rapidly changing phenomena.

At one-time observation Yu denia statistical data are collected at a certain (critical) point in time. One-time observations are: a population census, a study of the physical development of children, accounting for hospital beds for horses of the year, certification of medical institutions, etc. Preventive examinations of the population also belong to this type. One-time registration reflects the state of the phenomenon at the time of study. This type of observation is used to study slowly changing phenomena.

The choice of the type of observation over time is determined by the purpose and objectives of the study. For example, the characteristics of hospitalized patients can be obtained as a result of the current registration of those who left the hospital (current observation) or by a one-day census of patients in the hospital (one-time observation).

Depending on the completeness of the coverage of the phenomenon under study, a continuous and non-continuous study is distinguished.

At continuous The study studies all the units of observation included in the population, i.e. the general population. A continuous study is carried out in order to establish the absolute size of the phenomenon, for example, the total population, the total number of births or deaths, the total number of cases of a particular disease, etc. The continuous method is also used in cases where information is necessary for operational work (accounting for infectious diseases , workload of doctors, etc.)

At discontinuous The study examines only part of the general population. It is divided into several types: questionnaire, monographic, main array, selective. The most common method in medical research is the sampling method.

Monographic method- gives a detailed description of individual units of the population, characteristic in any respect, and a deep, comprehensive description of objects.

Main Array Method- involves the study of those objects in which the vast majority of units of observation are concentrated. The disadvantage of this method is that a part of the population remains uncovered by the study, although small in size, but which can differ significantly from the main array.

Questionnaire method- this is the collection of statistical data using specially designed questionnaires addressed to a certain circle of people. This study is based on the principle of voluntariness, so the return of questionnaires is often incomplete. Often the answers to the questions posed bear the imprint of subjectivity and chance. This method is used to obtain an approximate description of the phenomenon under study.

Sampling method- is reduced to the study of some specially selected part of the units of observation to characterize the entire general population. This method has the advantage of obtaining results with a high degree of reliability as well as a significantly lower cost. The study employs a smaller number of performers , moreover, it requires less time.

In medical statistics, the role and place of the sampling method is especially great, since medical workers usually deal with only a part of the phenomenon under study: they study a group of patients with a particular disease, analyze the work of individual departments and medical institutions , evaluate the quality of certain events, etc.

According to the method of obtaining information in the course of statistical observation and the nature of its implementation, several types are distinguished:

1) direct observation(clinical examination of patients , conducting laboratory , instrumental research , anthropometric measurements, etc.)

2) sociological methods: interview method (face-to-face survey), questioning (correspondence survey - anonymous or non-anonymous), etc .;

3) documentary research a nie(copy of information from accounting and reporting medical documents, information from official statistics of institutions and organizations.)

Third stage- grouping and summary of material - begins with checking and clarifying the number of observations , completeness and correctness of the information received , identifying and eliminating errors, duplicate records, etc.

For the correct development of the material, encryption of primary accounting documents is used. , those. designation of each feature and its group with a sign - alphabetic or numeric. Encryption is a technique , facilitating and accelerating material development , improving the quality, accuracy of development. Ciphers - symbols - are developed arbitrarily. When coding diagnoses, it is recommended to use the international nomenclature and classification of diseases; when coding professions - a dictionary of professions.

The advantage of encryption is that, if necessary, after the end of the main development, you can return to the material for development in order to clarify new relationships and dependencies. Encrypted accounting material makes it easier and faster , than unencrypted. After checking, the features are grouped.

grouping- division of the totality of the studied data into homogeneous , typical groups according to the most significant features. Grouping can be carried out on qualitative and quantitative grounds. The choice of a grouping feature depends on the nature of the studied population and the objectives of the study.

Typological grouping is carried out according to qualitative (descriptive, attributive) features, for example, by gender , profession, disease groups, severity of the course of the disease, postoperative complications, etc.

Grouping by quantitative (variation) features is carried out on the basis of the numerical size of the feature , For example , by age , the duration of the disease, the duration of treatment, etc. Quantitative grouping requires a solution to the question of the size of the grouping interval: the interval can be equal, and in some cases - unequal, even include the so-called open groups.

for example , when grouping by age, open groups can be determined: up to 1 year . 50 years and older.

When determining the number of groups proceed from the purpose and objectives of the study. It is necessary that groupings could reveal the patterns of the phenomenon under study. A large number of groups can lead to excessive crushing of the material, unnecessary detailing. A small number of groups leads to obscuring of characteristic features.

Having finished grouping the material, proceed to the summary.

With vodka- generalization of isolated cases , received as a result of a statistical study, into certain groups, their calculation and inclusion in the layout tables.

A summary of the statistical material is carried out using statistical tables. Table , not filled with numbers , called a layout.

Statistical tables are list , chronological, territorial.

The table has a subject and a predicate. The statistical subject is usually placed on horizontal lines on the left side of the table and reflects the main, main feature. The statistical predicate is placed from left to right along the vertical columns and reflects additional accounting features.

Statistical tables are divided into simple , group and combination.

AT simple tables the numerical distribution of the material according to one attribute is presented , its constituent parts (Table 1). A simple table usually contains a simple list or summary of the totality of the phenomenon under study.

Table 1

Distribution of the dead in the N. hospital by age

AT group tables presents a combination of two signs in connection with each other (Table 2).

table 2

Distribution of the dead in N. hospital by sex and age

AT combin a qi about these tables the distribution of the material according to three or more interrelated features is given (Table 3).

Table 3

Distribution of deaths in N. hospital with different diseases by age and sex

Diagnosis of the underlying disease	Age
0-14	15-19	20-39	40-59	60 and >	Total
m	well	m	well	m	well	m	well	m	well	m	well	m+f
Diseases of the circulatory system.	-	-	-	-
Injury and poisoning		-		-					-
malignancy. neoplasms.	-	-	-	-	-	-
Other zab.	-	-	-	-
Everyone got sick.		-		-

When compiling tables, certain requirements must be met:

Each table should have a heading that reflects its content;

Within the table, all columns should also have clear, concise titles;

When filling out the table, all cells of the table must contain the corresponding numerical data. The cells of the table remaining blank due to the absence of this combination are crossed out ("-"), and in the absence of information in the cell, "n.s." or "...";

After filling the table in the bottom horizontal row and in the last vertical column on the right, the results of vertical columns and horizontal lines are summed up.

Tables must have a single sequential numbering.

Currently, computers are widely used in sorting and summarizing material. . which allow not only to sort the material according to the studied characteristics , but do the calculations.

Fourth stage- statistical analysis - is a crucial stage of the study. At this stage, the calculation of statistical indicators (frequency , structures , the average size of the phenomenon under study), their graphic representation is given , dynamics , trends, connections between phenomena are established . forecasts are given, etc. The analysis involves the interpretation of the data obtained, the assessment of the reliability of the results of the study. In conclusion, conclusions are drawn.

Fifth stage- Literary processing is final. It involves the finalization of the results of a statistical study. The results can be presented in the form of an article, report, report , dissertations, etc. There are certain requirements for each type of design , which must be observed in the literary processing of the results of a statistical study.

The results of medical and statistical research are being introduced into healthcare practice. Various options for using the results of the study are possible: familiarization with the results of a wide audience of medical and scientific workers; preparation of instructive and methodological documents; formulation of a rationalization proposal and others.

Introduction

1. Methodology for obtaining initial data

2. Statistical summary and grouping of primary data

2.1 Grouping

2.2 Determination of arithmetic and structural averages

2.3 Histogram and cumulate

2.4 Cost of fixed assets

2.5 Production volume

3. Correlation analysis

3.1 Study of the relationship between factor and performance characteristics. Building a correlation table

3.2 Determination of the degree of closeness of communication

4. Regression analysis

4.1 Simulation

4.2 Prediction

Conclusion

Used literature and programs

Introduction

Statistical study of the phenomena of social life begins with the stage of statistical observation, during which, in accordance with cognitive goals and objectives, an array of initial data about the object under study is formed, i.e. the information base of the study is formed, on which accounting and control, planning, statistical analysis and management are carried out. At this stage, methods of mass observation based on the "law of large numbers" are used, since quantitative patterns of mass phenomena are clearly manifested in the study of only a sufficiently large number of socio-economic phenomena and processes.

Any statistical observation should be prepared and carried out according to a clearly developed plan, which includes issues of methodology, organization and technique of data collection, control of its quality and reliability. Thus, statistical observation must have a program and an organizational plan for conducting it. At the same time, it is necessary to resolve questions about the method, form, type, means, timing, place of organization and conduct of observation, etc., which, in turn, determines its plannedness.

Statistical observation should not be carried out spontaneously, from time to time, but systematically: either continuously or periodically - at regular intervals. This is due to the spatio-temporal variation of the studied socio-economic phenomena and processes.

Statistical observation can be carried out by state statistics bodies, research institutes, economic and analytical services of various organizational structures.

The second stage of statistical research is a statistical summary and grouping of statistical observation data. As a result of statistical observation, information is obtained about each unit of the population, which has numerous features that change in time and space. Under these conditions, there is a need to systematize and generalize the results of statistical observation and obtain, on this basis, a summary of the characteristics of the entire object using generalizing indicators, in order to make it possible to identify the characteristic features, specific features of the statistical population as a whole and its individual components and to discover patterns of socially studied - economic phenomena and processes. From what has been said, it follows that a summary of the primary statistical material is necessary.

The statistical summary is carried out according to a specially developed program that ensures the completeness and reliability of the results obtained. This program contains a list of groups into which the set of observation units can be divided according to individual characteristics, as well as a system of indicators characterizing the studied set of phenomena as a whole and its individual parts.

The third stage of statistical research is the analysis of statistical information. At this stage, based on the results of a statistical study, conclusions are obtained that are useful for practical actions, and the phenomenon or process under study is predicted.

1. Methodology for obtaining initial data

In order to study the dependence of the volume of production on the value of fixed assets for the period 2006-2007. the territorial body of state statistics for the Chelyabinsk region organized a statistical study of instrument-making enterprises.

Produced 20 percent typical sample.

The object of statistical observation is a set of instrument-making enterprises in the city of Chelyabinsk and the Chelyabinsk region. The reporting unit of statistical observation is an instrument-making enterprise.

In order to improve the system of sample surveys of instrument-making enterprises, the Goskomstat of the Russian Federation has developed a target Program.

According to the Program, in order to save resources, 20% of the total number of enterprises in the Chelyabinsk region operating on the date of work will be examined. The program activities include a number of organizational, methodological, software and technological works that ensure the preparation and conduct of sample surveys of instrument-making enterprises, the subject of which covers such an issue as the dependence of the volume of production on the value of fixed assets. In order to ensure proper procedures for the preparation of sample surveys of enterprises, the activities of the Program also include the training of personnel in surveys and outreach. The Program is expected to be implemented during 2008-2009. The results of a sample observation of instrument-making enterprises in the city of Chelyabinsk and the Chelyabinsk region for two indicators (production volume and cost of fixed assets) are shown in table 1.

Table 1 . The main performance indicators of instrument-making enterprises in the city of Chelyabinsk and the Chelyabinsk region for the period 2006 - 2007.

Factory No.	The cost of fixed assets	Production volume, million rubles	Factory No.	The cost of fixed assets	Production volume, million rubles

2. Statistical summary and grouping of primary data

2.1 Grouping

According to the statistical observation data, it can be seen that the variation of signs manifests itself in relatively narrow boundaries and the distribution is uniform. In this case, a grouping is built at equal intervals. The number of groups depends primarily on the degree of fluctuation of the trait: the greater the fluctuation of the trait (range of variation), the more groups can be formed. Below are the formulas for building a statistical grouping.

Since the sample size is not large, we use the formula to determine the number of groups:

Interval value h according to the formula:

The value obtained by formula (1.2), which will be the interval step, is rounded off (rounding should not differ from the original value by more than 10-15%). In this case, for the first interval, the lower boundary will be , and the upper - (+ h) etc. Thus, the lower limit of the i -th interval is equal to the upper limit of the (i -1) -th interval. Abstract >> Philosophy

... stages. 1.Main stages development of sociology 1.1First stage ... , "Main questions of Marxism", "Art and public a life", "K... naturalistic interpretations public phenomena. The essence of ... credibility statistical information in sociological research". ...

Marketing research in tourism

Cheat sheet >> Physical education and sports

Tourism of the region, scientific and statistical research in the field of tourism, training ... (member of the commission). 3. Basic stage: organization of certification work ... etc.). Method is a way of knowing, research phenomena public life, reception or system of receptions in...

Methods statistical research (2)

Test work >> Economics

Costs life. Index... statistical research statistical study consists of three major stages: statistical ... stage research; an organizational plan for its implementation is drawn up; object is defined (set public phenomena ...

Statistical methods of analysis of macroeconomic indicators (1)

Abstract >> Marketing

... statistical study…………………………………………………………………………..4 2.2.System statistical macroeconomic indicators………………..6 2.3. Main... level life population... stage statistical research apply the method statistical ... public phenomena reflected...

Statistical Research (SI) allows you to get an idea about a particular phenomenon, to study its size, level, to identify patterns. The subject of SI can be the health of the population, the organization of medical care, environmental factors affecting health, etc.

When conducting SI can be used 2 methodological approaches:

1) the study of the intensity of the phenomenon in the environment, the prevalence of the phenomenon, the identification of trends in the state of health of the population - are carried out on general populations or sample populations large enough in number to obtain intensive indicators and reasonably transfer the data obtained to the entire population

2) carrying out strictly planned studies on the study of individual factors without revealing the intensity of the phenomenon in the environment - they are carried out, as a rule, on small populations in order to identify new factors, study unknown or little-known cause-and-effect relationships

Stages of statistical research:

Stage 1. Drawing up a plan and research program- is preparatory, it determines the purpose and objectives of the study, draws up a plan and program of the study, develops a program for summarizing statistical material and resolves organizational issues.

A) the purpose and objectives of the study should be clearly formulated; the goal determines the main direction of research and is, as a rule, not only theoretical, but also practical, it is formulated clearly, clearly, unambiguously; to disclose the goal, the research tasks are determined.

B) it is necessary to study the literature on this topic.

B) it is necessary to develop organizational plan - provides for the determination of 1) the place (administrative-territorial boundaries of the observation), 2) the time (specific terms for the implementation of the observation, the development and analysis of the material) and 3) the subject of the study (organizers, performers, methodological and organizational leadership, sources of research funding).

D) development Research plan – includes the definition:

- the object of study (statistical population);

- the scope of the study (continuous, non-continuous);

– types (current, one-time);

– ways to collect statistical information.

D) need to make Research (observation) program - includes:

– determination of the unit of observation;

- a list of questions (accounting signs) to be registered in relation to each unit of observation

- development of an individual accounting (registration) form with a list of questions and signs to be recorded;

- development of layout tables, in which the results of the study are then entered.

For each unit of observation, a separate form is filled out, it contains a passport part, clearly formulated questions of the program, put in a certain sequence, and the date of filling out the document. As accounting forms, accounting medical forms used in the practice of medical institutions can be used.

At present, in connection with the machine processing of the results of observation using a computer, program questions can be formalized , When questions in the accounting document are put in the form of alternatives (yes, no) , Or ready-made answers are offered, from which a specific answer should be selected.

E) it is necessary to draw up a program for summarizing the data obtained, which includes establishing the principles of grouping, highlighting grouping characteristics , Determination of combinations of these signs, drawing up layouts of statistical tables.

Stage 2. Collection of material (statistical observation)- - consists in the registration of individual cases of the phenomenon under study and the accounting signs characterizing them in registration forms. Before and during the performance of this work, instruction (oral or written) of the observers is carried out, and they are provided with registration forms.

Statistical observation can be:

BUT ) by time:

1) Current- the phenomenon is studied for some separate period of time (week, quarter , Year, etc.) by daily recording the phenomenon as each case occurs (counting the number of births , The dead, sick , discharged from the hospital). This takes into account rapidly changing phenomena.

2) one-time- statistical data are collected at a certain (critical) point in time (population census, study of the physical development of children, preventive examinations of the population). One-time registration reflects the state of the phenomenon at the time of study, is used to study slowly changing phenomena.

The choice of the type of observation in time is determined by the purpose and objectives of the study (characteristics of hospitalized patients can be obtained as a result of the current registration of those who left the hospital - current observation or by a one-day census of patients in the hospital - one-time observation).

B) depending on the completeness of the coverage of the phenomenon under study:

1) continuous- all the units of observation included in the population, i.e. the general population, are studied. Carried out in order to establish the absolute size of the phenomenon (total population, total number of births or deaths). It is also used in cases where information is necessary for operational work (accounting for infectious diseases, the workload of doctors, etc.)

2) discontinuous- only part of the general population is studied, it is divided into several types:

1. Monographic method- gives a detailed description of individual units of the population, characteristic in any respect, and a deep, comprehensive description of objects.

2. Main Array Method- involves the study of those objects in which the vast majority of units of observation are concentrated. The disadvantage of this method is that a part of the population remains uncovered by the study, although small in size, but which can differ significantly from the main array.

3. Questionnaire method- this is the collection of statistical data using specially designed questionnaires addressed to a certain circle of people. This study is based on the principle of voluntariness, so the return of questionnaires is often incomplete. Often the answers to the questions posed bear the imprint of subjectivity and chance. This method is used to obtain an approximate description of the phenomenon under study.

4. Sampling method- the most common method, is reduced to the study of some specially selected part of the units of observation to characterize the entire general population. This method has the advantage of obtaining results with a high degree of reliability as well as a significantly lower cost. The study employs a smaller number of performers , In addition, it requires less time. In medical statistics, the role and place of the sampling method is especially great, since medical workers usually deal only with a part of the phenomenon being studied (they study a group of patients with a particular disease, analyze the work of individual units).

C) according to the method of obtaining information during the conduct and the nature of its implementation

1. Direct observation(clinical examination of patients , Conducting laboratory , instrumental research , Anthropometric measurements, etc.)

2. sociological methods: interview method (face-to-face survey), questioning (remote survey - anonymous or non-anonymous), etc .;

3. Documentary research(copy of information from accounting and reporting medical documents, information from official statistics of institutions and organizations.)

Stage 3. Material development, statistical grouping and summary– starts with checking and refining the number of observations , Completeness and correctness of the information received , Identification and elimination of errors, duplicate entries, etc.

For the correct development of the material is used Encryption of primary accounting documents, That is, the designation of each feature and its group with a sign - alphabetic or numeric. Encryption is a technique , Facilitating and accelerating material development , Increasing quality, development accuracy. Ciphers - symbols - are developed arbitrarily. When coding diagnoses, it is recommended to use the international nomenclature and classification of diseases; when coding professions - a dictionary of professions.

The advantage of encryption is that, if necessary, after the end of the main development, you can return to the material for development in order to clarify new relationships and dependencies. Encrypted accounting material makes it easier and faster , Than unencrypted. After checking, the features are grouped.

Grouping - division of the set of studied data into homogeneous , Typical groups according to the most significant features. Grouping can be carried out on qualitative and quantitative grounds. The choice of a grouping feature depends on the nature of the studied population and the objectives of the study.

BUT) Typological grouping produced according to qualitative (descriptive, attributive) features (sex , Occupation, disease groups)

B) Variational grouping(by quantitative traits) is carried out on the basis of the numerical dimensions of the trait (age , Duration of the disease, duration of treatment, etc.). Quantitative grouping requires a solution to the issue of the size of the grouping interval: the interval can be equal, and in some cases unequal, even include the so-called open groups (when grouping by age, open groups can be defined: up to 1 year, 50 years and older).

Having finished grouping the material, proceed to summary– generalization of isolated cases , Received as a result of a statistical study, into certain groups, their calculation and inclusion in the table layouts.

A summary of the statistical material is carried out using statistical tables. Table , not filled with numbers , called layout.

Statistical tables are list , Chronological, territorial.

Statistical tables are divided into:

BUT) Simple- the numerical distribution of the material according to one attribute is presented , constituent parts of it. A simple table usually contains a simple list or summary of the totality of the phenomenon under study.

B) Group- a combination of two features in connection with each other

AT) Combination- the distribution of the material according to three or more interrelated features is given

When compiling tables, certain requirements must be met.:

- each table should have a heading that reflects its content;

- inside the table, all columns should also have clear short names;

– when filling in the table, all cells of the table must contain the corresponding numerical data. The cells of the table that remain blank due to the absence of this combination are crossed out ("-"), and in the absence of information in the cell, "n. with." or "…";

- after filling in the table in the bottom horizontal row and in the last vertical column on the right, the results of vertical columns and horizontal lines are summed up.

– Tables must have a single sequential numbering.

In studies with a small number of observations, summarization is done manually. All accounting documents are decomposed into groups in accordance with the sign code. Next, the data is calculated and recorded in the corresponding cell of the table. Currently, computers are widely used in sorting and summarizing material. . Which allow not only to sort the material according to the studied characteristics , But do the calculations.

Stage 4. Statistical analysis of the phenomenon under study, formulation of conclusions- a critical stage of the study, at which the calculation of statistical indicators (frequencies , structures , The average size of the phenomenon under study), their graphic representation is given , Dynamics being studied , Trends, connections between phenomena are established . Forecasts are given, etc. The analysis involves the interpretation of the data obtained, the assessment of the reliability of the research results. In conclusion, conclusions are drawn.

Stage 5 Literary processing and presentation of the results- is final, involves the final registration of the results of a statistical study. The results can be presented in the form of an article, report, report , Dissertations, etc. There are certain requirements for each type of design , Which must be observed in the literary processing of the results of a statistical study.

The results of medical and statistical research are being introduced into healthcare practice. Various options for using the results of the study are possible: familiarization with the results of a wide audience of medical and scientific workers; preparation of instructive and methodological documents; preparation of a rationalization proposal and others

Upon completion of the statistical study, recommendations and management decisions are developed, the results of the study are put into practice, and efficiency is evaluated.

In conducting a statistical study, the most important element is the observance of a strict sequence in the implementation of these stages.

2.1 Scheme for conducting a statistical study

Statistical data analysis systems are a modern and effective tool for statistical research. Wide opportunities for processing statistical data have special systems of statistical analysis, as well as universal tools - Excel, Matlab, Mathcad, etc.

But even the most perfect tool cannot replace the researcher, who must formulate the purpose of the study, collect data, select methods, approaches, models and tools for data processing and analysis, and interpret the results.

Figure 2.1 shows the scheme for conducting a statistical study.

Fig.2.1 - Schematic diagram of a statistical study

The starting point of statistical research is the formulation of the problem. When determining it, the purpose of the study is taken into account, it is determined what information is needed and how it will be used in making a decision.

The statistical study itself begins with a preparatory stage. During the preparatory phase, analysts study technical task- a document compiled by the customer of the study. The terms of reference should clearly state the objectives of the study:

the object of study is defined;

lists the assumptions and hypotheses that must be confirmed or refuted during the study;

describes how the results of the study will be used;

the timeframe in which the study is to be conducted and the budget for the study.

Based on the terms of reference, a analytical report structure- then, in any form the results of the research should be presented, as well as statistical observation program. The program is a list of features to be recorded during the observation process (or questions to which reliable answers must be obtained for each surveyed unit of observation). The content of the program is determined both by the characteristics of the observed object and the objectives of the study, and by the methods chosen by analysts for further processing the collected information.

The main stage of statistical research includes the collection of the necessary data and their analysis.

The final stage of the study is the preparation of an analytical report and its provision to the customer.

On fig. 2.2 is a diagram of statistical data analysis.

Fig.2.2 - The main stages of statistical analysis

2.2 Collection of statistical information

The collection of materials involves the analysis of the terms of reference of the study, the identification of sources of necessary information and (if necessary) the development of questionnaires. In the study of information sources, all the required data is divided into primary(data not available and to be collected directly for this study), and secondary(previously collected for other purposes).

The collection of secondary data is often referred to as "desk" or "library" research.

Examples of primary data collection: observations of store visitors, surveys of hospital patients, discussion of a problem at a meeting.

Secondary data is divided into internal and external.

Examples of internal secondary data sources:

information system of the organization (including the accounting subsystem, the sales management subsystem, CRM (CRM-system, short for Customer Relationship Management) - application software for organizations designed to automate customer interaction strategies) and others);

previous studies;

written reports from employees.

Examples of external secondary data sources:

reports of statistical bodies and other state institutions;

reports from marketing agencies, professional associations, etc.;

electronic databases (address directories, GIS, etc.);

libraries;

mass media.

The main outputs of the data collection phase are:

planned sample size;

sample structure (presence and size of quotas);

type of statistical observation (data collection survey, questioning, measurement, experiment, examination, etc.);

information about the parameters of the survey (for example, the possibility of the fact of falsification of questionnaires);

coding scheme for variables in the database of the program selected for processing;

plan-scheme of data transformation;

plan-scheme of the statistical procedures used.

This stage also includes the questioning procedure itself. Of course, questionnaires are developed only to obtain primary information.

The received data should be appropriately edited and prepared. Each questionnaire or form of observation is checked and, if necessary, corrected. Each answer is assigned numeric or alphabetic codes - information is encoded. Data preparation includes editing, decryption and data validation, coding and necessary transformations.

2.3 Characterization of the sample

As a rule, the data collected as a result of statistical observation for statistical analysis is a sample. The sequence of data transformation in the process of statistical research can be schematically represented as follows (Fig. 2.3)

Figure 2.3 Statistical Data Conversion Scheme

Analyzing the sample, it is possible to draw conclusions about the general population represented by the sample.

Final determination of general sampling parameters produced when all questionnaires are collected. It includes:

determination of the real number of respondents,

determination of the sample structure,

distribution according to the place of the survey,

establishing a confidence level of the statistical reliability of the sample,

calculation of statistical error and determination of sample representativeness.

Real quantity respondents may be more or less than planned. The first option is better for analysis, but disadvantageous for the customer of the study. The second one can adversely affect the quality of the study, and, therefore, is unprofitable for either analysts or customers.

Sample structure can be random or non-random (respondents were selected on the basis of a previously known criterion, for example, by the quota method). Random samples are a priori representative. Non-random samples may be intentionally unrepresentative of the general population, but provide important information for research. In this case, you should also carefully consider the filtering questions of the questionnaire, which are designed specifically to screen out unsuitable respondents.

For determination of estimation accuracy, first of all, it is necessary to establish the level of confidence (95% or 99%). Then the maximum statistical error sample is calculated as

or
,

where - sample size, - the probability of the occurrence of the event under study (the respondent getting into the sample), - the probability of the reverse event (the respondent not being included in the sample), - confidence coefficient,
is the variance of the feature.

Table 2.4 lists the most commonly used values of confidence probability and confidence coefficients.

Table 2.4

2.5 Computer data processing

Data analysis using a computer involves a number of necessary steps.

1. Determination of the structure of the initial data.

2. Entering data into a computer in accordance with their structure and program requirements. Editing and transformation of data.

3. Setting the method of data processing in accordance with the objectives of the study.

4. Obtaining the result of data processing. Editing and saving it in the desired format.

5. Interpretation of the processing result.

Steps 1 (preparatory) and 5 (final) cannot be performed by any computer program - the researcher does them himself. Steps 2-4 are performed by the researcher using the program, but it is the researcher who determines the necessary data editing and transformation procedures, data processing methods, and the format for presenting the processing results. The help of the computer (steps 2-4) is, ultimately, in the transition from a long sequence of numbers to a more compact one. At the “input” of the computer, the researcher submits an array of initial data that is inaccessible to comprehension, but suitable for computer processing (step 2). Then the researcher gives the program a command to process the data in accordance with the task and data structure (step 3). At the “output”, he receives the result of processing (step 4) - also an array of data, only a smaller one, accessible to comprehension and meaningful interpretation. At the same time, an exhaustive analysis of data usually requires their repeated processing using different methods.

2.6 Choosing a data analysis strategy

The choice of a strategy for analyzing the collected data is based on knowledge of the theoretical and practical aspects of the subject area under study, the specifics and known characteristics of information, the properties of specific statistical methods, as well as the experience and views of the researcher.

It must be remembered that data analysis is not the ultimate goal of the study. Its purpose is to obtain information that will help solve a specific problem and make adequate management decisions. The choice of analysis strategy should begin with an examination of the results of the previous stages of the process: defining the problem and developing a research plan. As a "draft", a preliminary data analysis plan is used, developed as one of the elements of the study plan. Then, as additional information becomes available at subsequent stages of the research process, certain changes may need to be made.

Statistical methods are divided into one- and multivariate. One-dimensional methods (univariatetechniques) are used when all elements of the sample are evaluated by one indicator, or if there are several of these indicators for each element, but each variable is analyzed separately from all the others.

Multivariate techniques are great for data analysis if two or more indicators are used to evaluate each sample item and these variables are analyzed simultaneously. Such methods are used to determine dependencies between phenomena.

Multivariate methods differ from univariate methods primarily in that they shift the focus from the levels (averages) and distributions (variances) of phenomena and focus on the degree of relationship (correlation or covariance) between these phenomena.

Univariate methods can be classified based on whether the data being analyzed is metric or non-metric (Figure 3). Metric data is measured on an interval scale or relative scale. Nonmetric data is evaluated on a nominal or ordinal scale

In addition, these methods are divided into classes based on how many samples - one, two or more - are analyzed during the studies.

The classification of one-dimensional statistical methods is presented in Figure 2.4.

Rice. 2.4 Classification of one-dimensional statistical methods depending on the analyzed data

The number of samples is determined by how the data is handled for a particular analysis, not by how the data was collected. For example, data on males and females can be obtained within the same sample, but if their analysis is aimed at revealing a difference in perception based on the difference in sex, the researcher will have to operate with two different samples. Samples are considered independent if they are not experimentally related to each other. Measurements made in one sample do not affect the values of variables in another. For analysis, data relating to different groups of respondents, such as those collected from females and males, are usually treated as independent samples.

On the other hand, if the data for two samples refer to the same group of respondents, the samples are considered to be paired - dependent.

If there is only one sample of metric data, the z- and t-test can be used. If there are two or more independent samples, in the first case, you can use the z- and t-test for two samples, in the second - the method of one-way ANOVA. For two related samples, a paired t-test is used. When it comes to non-metric data on a single sample, the researcher can use the frequency distribution tests, chi-square, Kolmogorov-Smirnov (K~S) test, series test and binomial test. For two independent samples with non-metric data, one can resort to the following analysis methods: chi-square, Mann-Whitney, median, K-S, one-way analysis of variance Kruskal-Wallis (DA K-U). In contrast, if there are two or more related samples, the sign, McNemar, and Wilcoxon tests should be used.

Multivariate statistical methods are aimed at identifying existing patterns: the interdependence of variables, the relationship or sequence of events, interobject similarity.

Quite conventionally, five standard types of patterns can be distinguished, the study of which is of significant interest: association, sequence, classification, clustering and forecasting.

An association occurs when several events are related to each other. For example, a study in a supermarket might show that 65% of those who buy corn chips also take Coca-Cola, and when there is a discount for such a set, they buy Coke in 85% of cases. Having information about such an association, it is easy for managers to assess how effective the discount provided is.

If there is a chain of events connected in time, then one speaks of a sequence. So, for example, after buying a house in 45% of cases, a new stove is also purchased within a month, and within two weeks, 60% of newcomers acquire a refrigerator.

With the help of classification, signs are revealed that characterize the group to which this or that object belongs. This is done by analyzing already classified objects and formulating a certain set of rules.

Clustering differs from classification in that the groups themselves are not predetermined. With the help of clustering, various homogeneous groups of data are distinguished.

The basis for all kinds of forecasting systems is historical information stored in the form of time series. If it is possible to find patterns that adequately reflect the dynamics of the behavior of target indicators, it is likely that with their help it is possible to predict the behavior of the system in the future.

Multivariate statistical methods can be divided into relationship analysis methods and classification analysis (Fig. 2.5).

Fig.2.5 - Classification of multivariate statistical methods

Portal for the student. Self-training

Introduction

1. Methodology for obtaining initial data

2. Statistical summary and grouping of primary data

2.1 Grouping

Marketing research in tourism

Methods statistical research (2)

Statistical methods of analysis of macroeconomic indicators (1)

RELATED ARTICLES