Types of observation by the time of data recording. Non-repeating and re-selection

Organizational forms of statistical observation	Types of statistical observation	Methods of statistical observation
by the time of registration of facts	by coverage of population units
1. Statistical reporting 2. Specially organized surveillance 3. Registers	1. Current or continuous 2. Discontinuous: a) periodic; b) one-time	1. Solid 2. Discontinuous: a) selective; b) main, array; c) monographic	1. Immediate 2. Documentary 3. Survey: a) forwarding; b) self-registration; c) correspondent; d) questionnaire; e) attendant

Statistical observation is a process that, from the point of view of its organization, can have a variety of methods, forms and types of conduct. The task of the general theory of statistics is to determine the essence of the methods, forms and types of observation in order to decide where, when and what methods of observation will be applied.

Statistical observations have two main groups:

1) coverage of population units;
2) time of registration of facts.

According to the level of coverage of the studied population, statistical observation is divided into two types: solid and non-solid.

Continuous (complete) observation refers to the coverage of all units of the studied population.

Under non-continuous observation is understood only the coverage of a certain part of the studied population.

There are several types of non-continuous observation: selective; observation of the main array; monographic.

Selective observation is understood as a part of the units of the studied population, selected by the method of random selection. The method of momentary observations is called selective observation, which involves the selection of not only the units of the population under study, but also the points in time at which the registration of signs is carried out.

Observation of the main array is the coverage of the survey of certain, most significant features of the units of the population.

Monographic observation is characterized by a comprehensive and complete study of only some units of the population that have some special characteristics or represent some new phenomenon. The purpose of such observation is to identify existing or only emerging trends in the development of a given process or phenomenon. Monographic observation is closely related to continuous and selective observations.

According to the time of registration of facts, observation can be continuous and discontinuous. Discontinuous monitoring, in turn, includes periodic and one-time.

Continuous (current) observation is realized by continuous registration of facts as they become available.

Discontinuous observation is carried out either systematically at fixed intervals, or once and irregularly as needed.

The specificity of a one-time observation is that facts are recorded not in connection with their occurrence, but according to their state or presence at a certain moment or over a period of time. Quantitative measurement of signs of a phenomenon or process occurs at the time of the survey, and re-registration of signs may not be carried out at all or the timing of its implementation is not predetermined.

The methods for obtaining statistical information are documentary method of observation; method of direct observation: survey.

Documentary observation is based on the use of data from various documents as a source of information. Considering that, as a rule, high requirements are imposed on filling out such documents, the data reflected in them are of the most reliable nature and can serve as a high-quality source material for analysis.

Direct observation is carried out by registering the facts personally established by the registrars as a result of inspection, measurement, and counting the signs of the phenomenon under study. In this way, prices for goods and services are recorded, measurements of working hours are made, an inventory of stock balances, etc.

The survey is based on obtaining data from the respondents. The survey is used in cases where observation by other methods cannot be carried out. This type of observation is typical for conducting various sociological surveys and public opinion polls.

Statistical information can be obtained by different types of surveys : forwarding, correspondent, personal, private.

Expeditionary survey is carried out by specially trained workers who record the answers of respondents in the observation forms. The form is a form of a document in which it is necessary to fill in the fields for answers.

The correspondent method assumes that, on a voluntary basis, the respondent staff reports information directly to the monitoring body. The disadvantage of this method is that it is difficult to verify the correctness of the received information.

In the questionnaire method, respondents fill out questionnaires voluntarily and mostly anonymously. Since this method of obtaining information is not reliable, it is used in those studies where high accuracy of the results is not required. In some situations, approximate results are sufficient, which capture only the trend and record the emergence of new facts and phenomena.

The secret method involves the submission of information to the bodies conducting monitoring, in secret. In this way, acts of civil status are registered: marriage, divorce, death, birth, etc.

The theory of statistics also considers the forms of statistical observation: reporting; specially organized statistical observation; registers.

Statistical reporting is the main form of statistical observation, which is characterized by the fact that statistical authorities receive information about the phenomena under study in the form of special documents submitted by enterprises and organizations within a certain time frame and in the prescribed form. The forms of statistical reporting themselves, the methods of collecting and processing statistical data, the methodology of statistical indicators established by the State Statistics Committee of Russia are the official statistical standards of the Russian Federation and are mandatory for all subjects of public relations.

Statistical reporting is divided into specialized and standard. The composition of standard reporting indicators is the same for all enterprises and organizations, while the composition of specialized reporting indicators depends on the specifics of individual sectors of the economy and areas of activity.

According to the deadlines for submission, statistical reporting is daily, weekly, ten-day, fortnightly, monthly, quarterly, semi-annual and annual.

Statistical reporting can be transmitted by telephone, via communication channels, on electronic media with mandatory subsequent submission on paper, certified by the signature of responsible persons.

Specially organized statistical observation is a collection of information organized by statistical authorities, either to study phenomena that are not covered by reporting, or to study reporting data in more depth, verify and refine them. Various kinds of censuses, one-time surveys are specially organized observations.

Registers are a form of observation in which the facts of the state of individual units of the population are continuously recorded. Observing a unit of the population, it is assumed that the processes occurring there have a beginning, a long-term continuation and an end. In the register, each unit of observation is characterized by a set of indicators. All indicators are stored until the observation unit is in the register and has not ended its existence. Some indicators remain the same as long as the unit of observation is in the register, others may change from time to time. An example of such a register is the unified state register of enterprises and organizations (USRE). All work on its maintenance is carried out by the State Statistics Committee of Russia.

statistical observation information error

and ways of observing. We are talking about their allocation in the statistics. We propose to first consider the types of observation that are used in this branch of knowledge. The need to choose a data collection option in it is determined by the fact that there are several types of observation. They differ mainly in the way in which facts are taken into account over time. From this point of view, the following types of observation are distinguished: systematic, periodic and one-time.

Systematic, periodic and one-time observation

Systematic observation, which is carried out continuously and as signs of the phenomenon of interest appear, is usually called current. It is carried out on the basis of primary documents containing the necessary information for a fairly complete characterization of the phenomenon.

Periodic monitoring is carried out at regular intervals. An example is

If the observation is carried out from time to time, there is no strict periodicity, or it has a one-time character, we are talking about a one-time observation.

Non-continuous and continuous observation

Types of observation in statistics are distinguished taking into account the difference in information in terms of completeness of coverage of the population. Distinguish in connection with this discontinuous and continuous. The latter is called one that takes into account all units of the studied population, without exception. However, it is not always expedient and possible to organize it, especially when it comes to product quality control. Continuous observation in this case leads to the fact that the mass of products of enterprises is excluded from the sphere of use. Therefore, it is necessary to carry out partial (non-continuous) observation. It takes into account only a part of the units of the population and gives an idea of the phenomenon as a whole, its characteristic features.

Continuing to consider the forms, types and methods of observation, we note that non-continuous observation has the following advantages:

1) much less communication and labor costs are required compared to continuous, since the number of surveyed units decreases;

2) it is possible to collect data according to a wider program and in a shorter period of time in order to comprehensively disclose within the given limits the features of the population of interest to us, to carry out its deep study;

3) non-continuous observation data are used to control materials obtained from continuous;

4) this species must be representative (representative).

Selection of units in non-continuous observation

Non-continuous observation is deliberately oriented towards taking into account a specific part of the units, which makes it possible to obtain generalizing stable characteristics of the population as a whole. In the practice of statistics, various types of observation methods are used. At the same time, the quality of a non-continuous one, of course, is inferior to the results obtained with a continuous one. However, in some cases only partial observation is possible.

The units to be studied are selected in such a way that, based on the data obtained from them, a correct idea of the phenomenon of interest as a whole is formed. Therefore, one of the main features of non-continuous observation is that the selection of population units is organized in the following ways:

monographic;

main array;

Selective;

Questionnaire.

Main array way

The selection of units of a certain population, which prevail according to the trait under study, involves the method of the main array. However, it is not so often used when a non-continuous view is used, and this method of observation does not ensure the selection of precisely those units that would represent the totality as a whole, all its parts. Selection with the help of the main array is made when the most significant, largest populations are taken, which prevail according to the trait under study in their total mass.

Selective observation

In order to obtain a characteristic of the population as a whole in terms of its units, which is used based on the principles of sampling. In this variant, the random nature of the selection guarantees the safety of the results obtained and prevents their bias.

Monographic description

Let us supplement the types of observation with a monographic description. It is a specific kind of observation in statistics. This is a detailed study of a single typical object that is interesting from the point of view of the population as a whole.

These are the main types of non-continuous observation.

General population and sample

The generalizing indicators of the population in the sampling method are established on the basis of some part of it (rather small - about 5-10%). At the same time, the set from which the selection of this part of the units is carried out is usually called the general set. The part of the units that was selected is called the sampling set (in other words, the sample). When using the sampling method, research is carried out with minimal costs of funds and labor and in a shorter time. This reduces registration errors and improves responsiveness.

Application of the sampling method in practice

Describing the main types of observation, one cannot but dwell in more detail on the selective one, which is very popular. It is only possible when only destructive can be realized. This type is common in departmental and state statistics (the study of the budget of the families of employees, peasants, workers, as well as housing conditions). It is also popular in trade (the effectiveness of new forms of its conduct, the demand for goods from the population), etc.

The sampling method is, in fact, a large one that differ significantly from each other. As a rule, they are based on the principle of random selection from the general population.

Examples of using the sampling method

Examples of types of observation allow you to visually demonstrate their use. Here are some examples of sampling, and you will better understand its features. It is he who is most theoretically developed today from discontinuous ones, since he is based on the principle of random selection. Each unit in the population at random selection has the same probability of being included in the sample. In a lottery draw, for example, this principle applies because there is an equal chance of winning for all tickets. The draw also uses random selection. If 1,000 out of 10,000 schoolchildren are selected for the purpose of studying their performance, then this can be done as follows: write the names of schoolchildren on separate sheets of paper and blindly pull out 1000.

Non-repeating and re-selection

Both non-repeated and repeated can be random selection. In practice, non-repetitive is most often used, that is, the unit that fell into the sample is not returned back to the general population, which means that the number of the latter is constantly decreasing. Lottery draws follow this pattern. The selected unit, when re-selected, is returned back to the general population. Consequently, the number of the latter remains unchanged during the sampling process. If we turn to our example with schoolchildren, we can note the following: in this case, if a sheet with a surname was included in the number of randomly selected ones, it would again come back and again could fall into the sample.

Selection methods by experts

It is very important that no factors, such as the commission organizing the survey or individuals, be able to influence. In other words, it is necessary that the principle of random selection be respected. However, in practice, its implementation is often difficult. There are areas of statistics in which expert selection methods predominate. This situation is due to various circumstances. For example, it takes place when selecting goods for the calculation of price indices or when forming the composition of "baskets" for assessing the cost of living. In such cases, the rejection of the random selection method can significantly increase the accuracy. However, in this case, the objectivity of the study is lost, and sometimes various types of observation errors occur, since everything depends in this case on the qualifications of the expert.

Mechanical (systematic) selection

Often in practice, mechanical (systematic) selection is used. For example, out of 10,000 schoolchildren, a thousand should be selected. In this case, they do this: all the guys are arranged in alphabetical order, and then every tenth of them is selected.

Since the interval in this case is 10, a 10% selection is made (10000 divided by 1000). If the third student is in the top ten (you can choose him by drawing lots), in this case the 13th, 23rd, 33rd ... 9993rd will be selected. With systematic selection, as we see, the general population is divided mechanically into a number of groups, and one unit is taken from each (in our example, one student). It should be noted that mechanical (systematic) selection is always non-repetitive. It should also be emphasized that the selected units with it are distributed evenly throughout the entire population.

Methods of Observation in Statistics

It is necessary to distinguish between methods and types. The latter we have just considered, we now turn to the study of methods. The fact is that the varieties of observation can also be distinguished regardless of the ways and sources of obtaining primary information. From this point of view, a distinction is made between documentary observation, questioning and direct observation.

Direct observation is such an observation that is carried out by counting, measuring the values of certain signs, by taking instrument readings by the persons who carry it out (they are called registrars).

Due to the fact that it is impossible to apply other methods and types of statistical observation, quite often it is carried out using a survey on a specific list of questions. Answers are recorded in a special form. There are, depending on how they are received, correspondent and forwarding, as well as a method of self-registration. Let's briefly describe each of them.

Forwarding is carried out by a special person (forwarder, counter) orally. This person completes the survey form or form.

The correspondent method is organized by sending survey forms to a certain circle of persons prepared in an appropriate way (they are called correspondents). These people, according to the agreement, must fill out the form, and then return it to the organization. The self-registration survey checks whether the forms have been completed correctly. As with the correspondent method, the questionnaires are filled out by the respondents themselves, however, the collection and distribution of them, as well as the control of the correctness of filling and instruction, are carried out by counters.

Forms of observation in statistics

Considering the forms, methods, types of statistical observation, we did not talk only about the forms. There are three of them: a register, specially organized monitoring and reporting. As you can see, the types and forms of statistical observation are not the same thing. You should understand the difference between them.

Reporting is the main form of observation. With its help, state statistics bodies receive information from organizations and enterprises in the form of reporting documents signed by responsible persons.

Specially organized observation is the collection of information organized by the statistical authorities to study phenomena not covered by reporting or for a deeper study of reporting data, their clarification and verification. It is carried out in the form of various surveys and censuses.

We have described almost all the main methods, types and forms of statistical observation. Only the last form remains - registers. It takes place in the case of continuous monitoring of processes that take place for a long time, which have a certain beginning, development and end. The facts of the state of the units of the population are continuously fixed. In statistical practice, business registers and population registers are distinguished. The latter represent a regularly updated and named list of the country's inhabitants. The register of enterprises contains enterprises with various types of economic activity and the values of certain characteristics for each unit.

So, we have considered the forms, methods, types of statistical observation. Of course, we touched on them only briefly, but we noted the most important.

There are two main forms of statistical observation(according to the method of organization): reporting and specially organized supervision.

Reporting- a form of observation in which enterprises and organizations provide statistical and higher authorities with permanent information characterizing their activities. Reporting is provided according to a predetermined program within strictly defined terms and contains the most important indicators necessary in the course of daily work.

Specially organized surveillance- an observation that is carried out for the specific purpose on a certain date in order to obtain data that are not collected by statistical reporting, as well as to verify the data of statistical reporting.

Types of statistical observation:

1) By the time of registration of facts statistical observation can be continuous, periodic and one-time.

Continuous surveillance (current)- carried out systematically, that is, all cases are recorded. Facts are recorded as they occur. An example is the registry office (births, marriages, divorces, deaths, traffic accidents, illegal acts).

Periodic observation- repeats at regular intervals. An example is the population census.

Single observation- is carried out as needed in the collection of data on any phenomenon or process without observing a certain periodicity. An example is the valuation and revaluation of fixed assets, a sample survey of schools, theaters, etc.

2) By coverage of population units distinguish continuous and non-continuous observation.

Continuous observation- observation in which all units of the studied population are subjected to research.

Discontinuous observation- observation, in which only a part of the units of the studied population, selected in a certain way, is subjected to research.

Types of non-continuous observation:

- Main Array Method- the largest units of the phenomenon under study are investigated.

- Sampling method- provides for special methods of selection and formation of the studied part of the population.

- Monographic method- with it, individual units of the population are subjected to the most thorough study, usually representatives of new species, types, or the best or worst units. The results are carried over to the entire population. This method allows you to identify trends in the development of any phenomena.

Methods of statistical observation.

The basis for the registration of facts can be either documents, or an opinion expressed, or chronometric data. In this regard, the following methods of observation are distinguished:

1. Direct observation (registration of signs or facts is carried out personally by counting, weighing, measuring, etc.).

2. Documentary (based on documentation).

3. Poll (the necessary information is obtained from the words of the respondent).

The following types of surveys are used in statistics: oral (expeditionary), correspondent, questionnaire, self-registration and private.

At oral questioning specially trained workers receive the necessary information on the basis of a survey of relevant persons and record the answers themselves in the observation form.

Correspondent poll- information to the monitoring bodies is provided by the staff of voluntary correspondents.

Questionnaire method involves the collection of information in the form of questionnaires.

At self-registration the forms are filled in by the respondents themselves, and the enumerators distribute the forms of the questionnaire to them, explain the rules for filling out, and then collect them.

Private method provides for the submission of information to the bodies conducting monitoring, in person. For example, registration of marriages, divorces, births, etc.

Statistical observations are divided into types according to the following criteria:

by the time of data registration;

by completeness of coverage of population units;

Types of statistical observation by registration time:

Current (continuous) observation - is carried out to study current phenomena and processes. Registration of facts is carried out as they are completed. (registration of marriages and divorces)

Discontinuous observation - is carried out as needed, while temporary gaps in data recording are allowed:

 Periodic observation - carried out at relatively equal time intervals (population census).

 One-time observation - carried out without observing the strict frequency of its implementation.

According to the completeness of coverage of population units, the following types of statistical observation are distinguished:

Continuous observation - is the collection and receipt of information about all units of the studied population. It is characterized by high material and labor costs, insufficient information efficiency. It is used in the population census, when collecting data in the reporting form, covering large and medium-sized enterprises of various forms of ownership.

Non-continuous observation - is based on the principle of random selection of units of the studied population, while in the sample population all types of units in the population must be represented. It has a number of advantages over continuous observation: reduction of time and money costs.

Non-continuous observation is divided into:

Sample observation - based on a random selection of units that are observed.

Monographic observation - consists in the examination of individual units of the population, characterized by rare qualitative properties. An example of a monographic observation: a description of the work of individual enterprises, to identify shortcomings in the work or development trends.

The method of the main array - consists in the study of the most significant, largest units of the population, which, according to the main feature, have the largest share in the population under study.

Momentary observation method - consists in conducting observations at random or constant time intervals with marks on the state of the object under study at one time or another.

Methods of statistical observation

Direct statistical observation - observation, in which the registrars themselves, by direct measurement, weighing, counting, establish the fact to be recorded.

Documentary observation - is based on the use of various kinds of accounting documents. It includes a reporting method of observation - in which enterprises submit statistical reports on their activities in a strictly mandatory manner.

Survey - is to obtain the necessary information directly from the respondent.

There are the following types of survey:

Expeditionary - registrars receive the necessary information from the interviewees and record it themselves in the forms.

Method of self-registration - the forms are filled in by the respondents themselves, the registrars only distribute the forms and explain the rules for filling them out.

Correspondent - information to the relevant authorities is reported by the staff of voluntary correspondents.

Questionnaire - the collection of information is carried out in the form of questionnaires, which are special questionnaires, it is convenient in cases where high accuracy of the results is not required.

Unsolicited - is to provide information to the relevant authorities on an unsolicited basis.

Selective observation

The concept of selective observation

Selective observation refers to a variety of discontinuous observation. It covers a selected part of the units of the general population. The purpose of selective observation is to characterize the entire population of units based on the selected part of the units. In order for the selected part to be representative (that is, to represent the entire population of units), sample observation must be specially organized. Therefore, unlike the general population, which represents the entire population of the units under study, the sample population represents that part of the units of the general population that is the object of direct observation.

For obvious reasons, the sampling method can be widely used by state statistics authorities. It allows, with significant savings in funds and costs, to obtain the necessary reliable information. The guarantee of representativeness is ensured by the use of scientifically based methods for selecting units that are to be surveyed.

It should immediately be borne in mind that when comparing indicators from the results of a sample study with characteristics for the entire general population, deviations may occur. The magnitude of these deviations is called observational error, which can be either registration error (imperfection of specifications) or representativeness error (accidental or systematic violation of the rules in the selection of units).

The following conventions are used in statistics:

N is the volume of the general population;

n is the sample size;

Average in the general population;

Average in the sample;

p is the proportion of units in the general population;

w is the proportion of units in the sample;

General dispersion;

S 2 - sample variance;

Standard deviation of a feature in the general population;

S is the standard deviation of a feature in the sample population.

Types of sampling, methods of selection and sampling errors

According to the method of selection (method of formation) of a sample of units from the general population, the following types of sample observation are common:

simple random sampling (properly random);

typical (stratified);

serial (nested);

mechanical;

combined;

stepped.

Simple random sampling (actually random) is the selection of units from the general population by random selection, but subject to the probability of choosing any unit from the general population. The selection is carried out by drawing lots or by a table of random numbers.

A typical (stratified) sample involves the division of a heterogeneous general population into typological or regionalized groups according to some significant feature, after which a random selection of units is made from each group.

For a serial (nested) sample, it is characteristic that the general population is initially divided into certain equal-sized or non-equal-sized series (the units within the series are connected according to a certain attribute), from which series are selected by random selection and then continuous observation is carried out within the selected series.

Mechanical sampling is the selection of units at regular intervals (alphabetical, temporal, spatial, etc.). During mechanical selection, the general population is divided into groups of equal size, from which one unit is then selected.

Combined sampling is based on a combination of several sampling methods.

A multi-stage sample is the formation within the general population of large groups of units at first, from which groups are formed that are smaller in size, and so on until those groups or individual units that need to be studied are selected.

Selective selection can be repeated and non-repeated. When reselecting, the probability of choosing any unit is not limited. In non-repetitive selection, the selected unit is not returned to the original population.

For the selected units, generalized indicators (average or relative) are calculated, and in the future, the results of the sample study are extended to the entire population.

The main task in sampling research is to determine sampling errors. It is customary to distinguish between mean and marginal sampling errors. For illustration, we can propose the calculation of the sampling error using the example of a simple random selection.

12 Absolute and relative values in statistics. Absolute values are the results of statistical observations. In statistics, unlike mathematics, all absolute values have a dimension (a unit of measurement), and can also be positive and negative. Units absolute values reflect the properties of units of the statistical population and can be simple, reflecting 1 property (for example, the mass of cargo is measured in tons) or complex, reflecting several interrelated properties (for example, ton-kilometer or kilowatt-hour). Units absolute values can be of 3 types:

natural- are used to calculate quantities with homogeneous properties (for example, pieces, tons, meters, etc.). Their disadvantage is that they do not allow summing dissimilar quantities.

Conditionally natural- apply to absolute values with homogeneous properties, but exhibiting them in different ways. For example, the total mass of energy carriers (firewood, peat, coal, oil products, natural gas) is measured in toe. - tons of reference fuel, since each of its types has a different calorific value, and 29.3 mJ / kg is taken as the standard. Similarly, the total number of school notebooks is measured in US dollars. - conditional school notebooks with a size of 12 sheets. Similarly, canning products are measured in a.c.b. - conditional cans with a capacity of 1/3 liter. Similarly, the production of detergents is reduced to a conditional fat content of 40%. Cost units of measurement are expressed in rubles or in another currency, representing a measure of the value of an absolute value. They allow you to summarize even heterogeneous values, but their disadvantage is that in this case it is necessary to take into account the inflation factor, therefore statistics always recalculate cost values in comparable prices. Absolute values can be momentary or interval. Momentary absolute values show the level of the studied phenomenon or process at a certain point in time or date (for example, the amount of money in your pocket or the value of fixed assets on the first day of the month). Interval absolute values are the final accumulated result for a certain period (interval) of time (for example, salary for a month, quarter or year). Interval absolute values, unlike moment ones, allow subsequent summation. The absolute statistical value is denoted X, and their total number in the statistical population is N.The number of values with the same attribute value is denoted f and is called frequency (repeatability, occurrence). By themselves, absolute statistical values do not give a complete picture of the phenomenon under study, since they do not show its dynamics, structure, and the relationship between parts. For these purposes, relative statistical values are used. A relative statistical value is the result of the ratio of two absolute statistical values. If absolute values \u200b\u200bwith the same dimension are correlated, then the resulting relative value will be dimensionless (the dimension will be reduced) and is called the coefficient.

More often instead of the name relative statistic a shorter synonym term is used - index (from lat. index- indicator, coefficient).

Depending on the types of correlated absolute values, when calculating relative values, different types of indices are obtained: dynamics, plan task, plan fulfillment, structure, coordination, comparison, intensity.

13. Average values. General principles for their application The average value is a generalizing indicator that characterizes the typical level of the phenomenon. It expresses the value of the attribute, related to the unit of the population. The average always generalizes the quantitative variation of the attribute, i.e., in the average values, individual differences in the units of the population due to random circumstances are canceled out. In contrast to the average, the absolute value that characterizes the level of a feature of an individual unit of the population does not allow comparing the values of the feature for units belonging to different populations. So, if you need to compare the levels of remuneration of workers at two enterprises, then you cannot compare two employees of different enterprises on this basis. The wages of the workers selected for comparison may not be typical for these enterprises. If we compare the size of wage funds at the enterprises under consideration, then the number of employees is not taken into account and, therefore, it is impossible to determine where the level of wages is higher. Ultimately, only averages can be compared, that is, how much one worker in each enterprise earns on average. Thus, there is a need to calculate the average value as a generalizing characteristic of the population. Let us dwell on some general principles for the application of averages. 1. When determining the average value in each specific case, it is necessary to proceed from the qualitative content of the averaged trait, take into account the relationship of the studied traits, as well as the data available for calculation. 2. The average value must first of all be calculated on a homogeneous population. Qualitatively homogeneous aggregates allow us to obtain the grouping method, which always involves the calculation of a system of generalizing indicators. 3. General averages must be supported by group averages. For example, suppose that an analysis of the dynamics of the yield of a particular crop shows that the overall average yield in the republic is declining. However, it is known that the yield of this crop depends on soil, climatic and other conditions and is different in individual areas. 68 Grouping the districts according to the signs of difference and analyzing the dynamics of group averages, it can be found that in certain groups of districts the average yield either did not change or increases, and the decrease in the overall average for the republic as a whole is due to an increase in the share of districts with lower yields in general. production of this crop. It is obvious that the dynamics of group averages more fully reflects the patterns of changes in yield, and the dynamics of the overall average shows only the general result. 4. A reasonable choice of the population unit for which the average is calculated is required. Let us now consider the types of averages, the features of their calculation and areas of application. Average values are divided into two large classes: power averages, structural averages. Power-law averages include the most well-known and commonly used types, such as geometric mean, arithmetic mean, and mean square. The mode and median are considered as structural averages. Let us dwell on power averages. Power averages, depending on the presentation of the initial data, can be simple and weighted. A simple average is calculated from ungrouped data and has the following general form: where X, - variant (value) is averaged

18The concept and classification of time series Dynamic series is a series of numerical data placed in chronological order. They are also called dynamic series, time series. A series of dynamics characterizes the magnitude of the phenomenon under study at a particular moment or for a certain period of time. There are two main elements in each series of dynamics: 1. indicator of time t; 2. corresponding levels of development of the studied phenomenon y. As indications of time in the series of dynamics, either certain dates (moments) or separate periods (years, quarters, months, days) are used. The levels of the time series reflect a quantitative assessment (measure) of the development of the phenomenon under study in time. They can be expressed as absolute, relative or average values. The series of dynamics differ in the following ways: 1) In time. Depending on the nature of the phenomenon under study, the levels of the series of dynamics can refer either to certain dates (moments) in time, or to individual periods. In accordance with this, the series of dynamics are divided into moment and interval. Moment series of dynamics reflect the state of the phenomena under study at certain dates (points) in time. An example of a moment series of dynamics is the following information on the payroll number of store employees in 2009 (Table 2): Table 2. Payroll number of store employees in 2009

A feature of the moment series of dynamics is that its levels can include the same units of the studied population. Although there are intervals in the moment series - gaps between adjacent dates in the series, the value of a particular level does not depend on the length of the period between two dates. Thus, the main part of the store's staff, which makes up the headcount as of 01.01.2009 and continues to work during this year, is displayed in the levels of subsequent periods. Therefore, when summing the levels of the moment series, repeated counting may occur. By means of moment series of dynamics in trade, commodity stocks, the state of personnel, the amount of equipment and other indicators that reflect the state of the studied phenomena at certain dates (points) in time are studied. Interval series of dynamics reflect the results of the development (functioning) of the studied phenomena for certain periods (intervals) of time. An example of an interval series is data on the retail turnover of a store in 2000-2004. (Table 3): Each level of the interval series is already the sum of levels for shorter periods of time. In this case, the unit of the population, which is part of one level, is not included in other levels. A feature of the interval series of dynamics is that each of its levels is made up of data for shorter intervals (sub-periods) of time. For example, summing up the turnover for the first three months of the year, you get its volume for the first quarter, and summing up the turnover for four quarters, you get its value for the year, etc. Ceteris paribus, the level of the interval series is the greater, the longer the interval, to which this level belongs. The property of summing levels for successive time intervals makes it possible to obtain series of dynamics of more enlarged periods. Through interval series, the dynamics in trade study changes in the time of receipt and sale of goods, the amount of distribution costs and other indicators that reflect the results of the functioning of the phenomenon under study for certain periods. Statistical display of the phenomenon under study in time can be represented by series of dynamics with progressive results. Their use is due to the need to display the results of the development of the studied indicators not only for a given reporting period, but also taking into account previous periods. When compiling such series, successive summation of adjacent levels is performed. This achieves a summary generalization of the result of the development of the indicator under study from the beginning of the reporting period (year, month, quarter, etc.). d.). Dynamic series with progressive results are built when determining the total volume of retail trade turnover. So, by summarizing commodity-money reports for the last operational periods (five days, weeks, decades, etc.). 2) By the form of presentation of levels. Dynamic series can also be built, the levels of which are relative and average values. They can also be either momentary or interval. In the interval series of the dynamics of relative and average values, the direct summation of the levels in itself is meaningless, since the relative and average values are derivatives and are calculated by dividing other values. 3) According to the distance between dates or time intervals, complete or incomplete series of dynamics are distinguished. Full series of dynamics take place when the dates of registration or the end of periods follow each other at equal intervals. These are equally spaced series of dynamics. Incomplete - when the principle of equal intervals is not respected. 4) According to the number of indicators, isolated and complex (multidimensional) time series can be distinguished. If one indicator is analyzed in time, we have an isolated series of dynamics. A complex series of dynamics is obtained when a system of indicators is given in chronological sequence, interconnected by the unity of a process or phenomenon.

Statistical observation can be divided into types according to two main features: the time of registration of facts and the coverage of population units (Fig. 2.4).

Rice. 2.4. Classification of types of statistical observation

According to the time of registration of facts, statistical observation is divided into continuous (current) and discontinuous. The latter, in turn, is divided into periodic and one-time.

Current observation- this is a type of statistical observation in which systematic registration of changes in relation to the studied phenomena is carried out as they occur, for example, in traffic accidents and illegal acts. Such an observation makes it possible to study in detail the dynamics of the desired phenomenon or process.

Periodic observation- this is a type of discontinuous observation, repeated at regular intervals. This type of observation includes a population census, which is repeated, as a rule, every 10 years.

One-time examination- this is a type of discontinuous observation, in which the collection of information about the quantitative characteristics of the phenomenon under study is carried out directly in the process of its study. A similar re-examination may be conducted after some predetermined period of time or may not be conducted again.

According to the coverage of population units, statistical observation can be continuous and non-continuous.

Continuous observation- this is a type of statistical observation in which all units of the studied population are subjected to examination. The main prerequisite for such an observation, on which the quality of the obtained statistical information depends, is the determination of the features to be registered. The statistical practice of using continuous observation is limited due to a number of significant shortcomings of this type of study. Firstly, these are the high costs of labor and financial resources for obtaining and processing the entire array of information. Secondly, the relatively low timeliness of the data received, since it takes a lot of time to collect and process them. Thirdly, continuous observation, as a rule, does not provide a complete examination of all units of the population without exception. In practice, for various reasons, a greater or lesser number of units remain unobserved. The percentage of unexamined units depends on a large number of factors: the method of observation, the list of questions asked, the professionalism of the registrar, the time of year and even the hour of the day of the study. Finally, for some populations, continuous observation is not possible, for example, when the survey leads to deterioration or destruction of the item under study, as in the study of food quality or testing the longevity of industrial parts. In such cases, non-continuous observation is resorted to.

Discontinuous observation- this is a type of statistical observation in which only a part of the units of the studied population is subjected to examination. The basis of such observation is the methodologically accurate choice of those units that will be subjected to the survey.

Selective observation- this is a type of non-continuous observation based on the principle of random selection of those units of the population under study that should be subjected to statistical observation. In the structure of the sample population, all characteristic types of units that make up the population under study should be represented. Only in this case, the sample population will be able to fully convey the dependencies characteristic of the desired population, and its results will be suitable for characterizing the entire population as a whole.

A type of sampling research is momentary method, in which statistical information is collected by registering the values of features from units of the sample population at some predetermined points in time. This method has received wide practical application in the analysis of the use of working time.

Main Array Method- this is a type of non-continuous observation, in which those units of the population that have the largest share of the studied trait are subjected to examination. Thus, the monitoring of price dynamics is carried out, as a rule, on the example of the largest cities in the country. The methodology of such a survey is based on the assumption that the exclusion of “insignificant” units from the observed set will not significantly affect the results of the study.

Monographic survey- this is a type of non-continuous observation, in which single objects of the studied population are subjected to a deep detailed examination, as a rule, representatives of characteristic or new types of social phenomena and processes. Accordingly, the purpose of a monographic survey is to characterize the already existing significant trends in the development of the phenomenon under study or to identify emerging changes. A deep study of one unit allows you to analyze both the positive and negative aspects of its development, which need to be eliminated in a timely manner. The detail of the results of a monographic examination is unattainable for other types of non-continuous observation. Often, the dependencies identified in the course of such a study turned out to have eluded earlier from the field of view in the analysis of mass data. Therefore, the results of a monographic survey are indispensable for refining the program of mass surveillance.

At present, it is becoming more and more common monitoring as a special systematic observation of the state of socio-economic phenomena and processes. Of particular relevance in modern society is the observation of social indicators characterizing such categories as, for example, the quality of life.

All types of statistical observation achieve the greatest efficiency in mutual combination and addition. A summary classification of forms, types and methods of statistical observation is presented in Table. 2.1.

Table 2.1

Classification of forms, methods and types of statistical observation