Measurement is the process of expressing the attributes or characteristics of a certain object or phenomenon with numbers or symbols according to certain rules.

(1) Measurement object, that is, the object of measurement. It is a thing or phenomenon that exists in the objective world, and it is an object that we want to express, explain and explain using numbers or symbols.

(2) Measurement content, that is, certain attributes or characteristics of the measurement object

(3) Measurement rules, that is, operating rules that use numbers and symbols to express various attributes or characteristics of things.

(4) Numbers and symbols, i.e. tools used to express measurement results

Categorical measurement is to distinguish different attributes or characteristics of the survey objects and label them with different names or symbols to determine their categories.

For example, gender, occupation, marital status, religious beliefs, etc.

Classified measurements have two properties: symmetry and transitivity.

The values ​​of ordinal measurement can rank the survey objects in a certain logical order and determine their level and order.

For example, education level, city size, social status, living standards, housing conditions, and work ability.

Ordinal measurements have three properties: symmetry, asymmetry, and transitivity.

Interval measurement can not only distinguish social phenomena or things into different categories and levels, but also determine the distance and quantity difference between them.

For example, IQ, temperature, etc.

In addition to all the properties of the above three levels of measurement, constant ratio measurement also has an absolute 0 point. Therefore, the data it measures can be added, subtracted, multiplied and divided.

For example, income, age, birth rate, sex ratio, etc.

Book P87 Table 5-1

In social measurement, there is an important rule: measure them as high a level as possible. Because high-level measurement contains a lot of information, and the results of high-level measurement can easily be converted into low-level measurement results, but not vice versa.

(1) Concepts

(2) variable ( variable )

(3) indicators

Concepts or variables are abstract, indicators are concrete

operationalization

From a broad perspective, the operationalization process mainly includes the following two aspects of work: first, defining concepts, and second, developing indicators. Specifically

Example 1. Parental investment.

Example 2. Survey on the social status of Chinese women.

Example 3. Quality of life of urban residents.

Operationalize "spoiling children" in relation to actual situations.

The summation scale consists of a set of statements that reflect people's attitudes or opinions about things. Respondents express their opinions on these statements respectively, and give different scores according to the degree of agreement or disagreement of the respondents. Then the respondents are divided into all statements. The scores above are added up to obtain the respondent's attitude score towards this thing or phenomenon. This score is a quantitative result of the attitude, and its level represents the individual's position on the attitude scale.

Notice

Likert scale is a specific form of summed scale, which consists of a set of statements about attitudes or opinions about something. Respondents' responses to these statements are divided into "strongly agree, agree, don't know, disagree" , Strongly Disagree", or "Agree, Somewhat Agree, Don't Care, Somewhat Disagree, Disagree". As the number of answer types increases, differences in people's attitudes are reflected more clearly.

It is mainly used to study the different meanings that concepts have for different people. This scale was originally used by American psychologist C. Osgood and others in their research.

The form of the semantic difference scale consists of two groups of adjectives with opposite meanings at both ends. Each pair of antonym adjectives is divided into 7 levels in the middle. The scores of each level from left to right are 7, 6, 5, 4, 3, 2, 1 can also be counted as +3, +2, +1, 0, -1, -2, -3.

Reliability refers to the degree to which the results are consistent when the same object is measured repeatedly using the same method. In other words, reliability refers to the consistency or stability of the measurement results, that is, whether the measurement tool can stably measure the thing or variable being measured.

type

Measurement validity refers to the degree to which a tool or measurement method can accurately measure the variables to be measured, or the degree to which the attributes of things can be accurately and truly measured.

type

three conditions

Generally, a measurement that lacks reliability must be an invalid measurement.

A measure with high reliability does not mean that it is also a highly valid measure; that is, it may be valid and still be invalid.

When researchers pursue measurement reliability, they often damage or reduce the validity of the measurement to a certain extent, and vice versa.

Case