Statistical data is a general term for the digital data and other data related to the national economy and social phenomena obtained in the process of statistical work activities. Statistics are the results of measuring a phenomenon. For example, a measure of total economic activity can yield data on the gross domestic product (GDP); a measure of stock price changes can yield data on a stock price index, and a measure of population gender can yield data such as male or female. The classification of statistical data is explained from different perspectives below. Statistical data is the product of the statistical work day. From the perspective of the statistical work process, the quality of statistical data is “standard”, and it must accurately reflect the objective reality.
Introduction to Statistics
Statistical data are data that express the characteristics, scale, structure, level, and other indicators of natural economic elements in a geographical area. It is the basic data for qualitative, positioning and quantitative statistical analysis. Such as what we usually call the statistical yearbook.
Data Classification
Statistical data can be expressed in two forms: statistical tables and statistical maps.
The representation method, it is divided into:
i. District statistics. That is, the area of ​​the graph or the number of the same graph represents the sum of all similar phenomena in the zoning unit; for example, the 2008 American Community Survey one-year data sample file population.
ii. Grading statistics. That is, it is classified according to administrative divisions or economic zones in the form of statistical charts, and the difference between the relative indicators of the phenomenon is represented by different shades of color or halo lines and halo points of varying density;
iii. Location statistics. The special phenomenon and change law at a certain point are represented in the form of statistical charts.
According to statistical indicators, statistics are divided into macroeconomic indicators statistics and industry economic indicators statistics.
Common macroeconomic indicators are GDP, CPI, PPI, [2] PMI and cash in circulation.
Industry economic indicators such as the coal industry, analysis of the prosperity of the oil industry, etc.
Statistical data is the result of measuring things using a certain measurement scale, and different types of statistical data will be obtained by using different measurement scales. From the results of the above four measurement scales, the statistical data can be divided into the following four types:
Categorical data – expressed as categories, but without distinguishing the order, it is formed by the classification scale measurement.
Ordinal data – presented as categories, but in order, formed by ordinal scale measurements.
Fixed-distance data – expressed as numerical values, which can be added and subtracted, and are formed by the measurement of fixed-distance scales.
Ratio data – expressed as a numerical value, which can be added, subtracted, multiplied, and divided, and is formed by the measurement of a fixed ratio scale.
The first two types of data describe the quality characteristics of things, which cannot be represented by data, and the results are expressed as categories, also known as qualitative data or quality data (Oualitative data); the latter two types of data describe the quantitative characteristics of phenomena, which can be It is represented by numerical values, so it is also called quantitative data or quantitative data (Quantitative data). Since the fixed-distance scale and the fixed-ratio scale belong to the same measurement level, the latter two kinds of data can be regarded as the same type of data, collectively referred to as quantitative data or numerical data.
It is important to distinguish between the level of measurement and the type of data, because different types of data will be processed and analyzed using different statistical methods. For example, for categorical data, the frequency or frequency of each group is usually calculated, the mode and difference ratio are calculated, and [3] contingency table analysis and x2 test are performed; for ordinal data, the median and Interquartile range, calculation of rank correlation coefficient and other non-parametric analysis; more statistical methods can be used to deal with interval or ratio data, such as calculation of various statistics, parameter estimation and testing, etc. Most of what we deal with is quantitative data.
It should be pointed out here that statistical methods applicable to low-level measurement data are also applicable to higher-level measurement data, because the latter has the mathematical properties of the former. For example, when describing the central tendency of data, the mode is usually calculated for categorical data, and the median is usually calculated for ordinal data, but the mode and median can also be calculated for interval and ratio data. Conversely, statistical methods that are suitable for high-level measurement data cannot be used for lower-level measurement data, because low-level data do not have the mathematical properties of high-level measurement data. For example, averages can be calculated for interval and ratio data, but not for categorical and ordinal data. Understanding this is useful for choosing a statistical analysis method.
Data Collection
Statistical data mainly come from two channels: one is from direct investigation and scientific experiment, which is the direct source of statistical data, called first-hand or direct statistical data; , which is an indirect source of statistics, called secondary or indirect statistics.
The organizational form of statistical data collection
The organizational forms of statistical data collection include censuses, sample surveys, statistical reports, key surveys, and typical surveys.
Census: A census is a one-time comprehensive survey specially organized for a specific purpose to collect comprehensive information on important national conditions, national strength and resources, and provide a basis for the government to formulate plans, guidelines and policies.
Sampling survey: Sampling survey is the most widely used survey method in practice. It randomly selects a part of the unit seat sample from the population of the survey object for survey, and infers the overall quantitative characteristics according to the sample survey results. Comprehensive survey method.
Statistical report: Statistical report is an investigation method mainly based on comprehensive investigation. It is arranged by the competent government department in accordance with statistical regulations, in the form of statistical tables and administrative means from top to bottom, and then by enterprises and institutions from the bottom. An investigation method that provides basic statistical data level by level, and the upper level summarizes and reports it.
Key investigation: Key investigation is a kind of non-comprehensive investigation specially organized. It selects individual or some key units in the whole to conduct investigation to understand the basic situation of the whole.
Typical survey: Typical survey is also a kind of non-comprehensive survey specially organized. It is based on the purpose and requirements of the survey and research, on the basis of a comprehensive analysis of the whole, and consciously selects representative typical units to conduct in-depth and detailed investigations. In order to understand the essential characteristics of things, causal relationships and the laws of development and change.
How to collect statistical data
In the actual survey, the specific methods of collecting data mainly include an interview survey, mail survey, telephone survey, symposium, individual in-depth interviews, and an online survey.
Interview survey: An interview survey, also known as a dispatch survey, is a survey method in which the investigator and the respondent obtain the required information through face-to-face conversation.
Postal survey: Postal survey is a survey method in which the questionnaire is sent to the respondents by mail or other means, the respondents fill in, and then send the questionnaire back or put it at the designated collection point.
Telephone survey: Telephone survey is a survey method in which investigators use the telephone to communicate with the respondents in language to obtain information. The advantages of telephone surveys are fast timeliness and low cost; the disadvantages are that the number of survey questions should not be too large.
Symposium: Symposium, also known as the collective interview method, is a method of gathering a group of respondents at the survey site and asking them to express their opinions on the subject of the survey, thereby obtaining survey data. This method is suitable for collecting the tendencies and opinions of a small number of people closely related to the research topic.
Individual in-depth interview: An individual in-depth interview is a special qualitative study involving only one respondent at a time. Often used in motivational research to uncover the deep-seated opinions of respondents that are not superficial. This method is most suitable for studying more secret issues, such as personal privacy; more sensitive issues and so on.
Online survey: Online survey mainly includes three methods: E-mail, interactive CATI system and Internet CGI program.