Inquiry question: The specific questions to be answered by social research institutions can be answered through research, which is similar to but different from the "questions" in daily life. Usually more specific, focused and clear. Main sources: real social life, personal experience, various documents, etc. Such as life practice, social problems, new social phenomena, other people's research, social policy needs and so on.
Research topic: a phenomenon field or problem field involved in social research. Compared with the problem studied, it is more extensive and universal. Such as: marriage, family, interpersonal relationship, social mobility, etc.
Clear research questions: through clear research questions and clear statements, we can turn the vague ideas in our minds into clear research questions; Turn the initial broad research scope or field into a specific phenomenon or a specific problem in a specific field. Usually it is to turn the topic into a problem.
Literature review: Literature review or literature review refers to the process of systematically consulting and analyzing all kinds of literature related to a research problem so far to understand the research status in this field. In other words, it is a process of systematically identifying, searching, examining and summarizing the documents related to our research.
Post-analysis is a special technique used by researchers in comprehensive review or more often in methodological review. Researchers collect detailed information of a large number of research projects, and then make statistical analysis of these information.
Verb (abbreviation of verb) research and design (chapter 4, chapter 5 of Feng, chapter 6 of Newman)
Research Design: Plan the whole research work, formulate specific strategies to explore specific social phenomena or things, determine the best research methods, and choose appropriate research methods. At the same time, it also includes detailed operation steps and research scheme. According to the research purpose, it can be divided into exploratory research, descriptive research and interpretive approach. Most studies are descriptive and explanatory, but the emphasis is different.
Exploratory research: it is a kind of research type that makes a preliminary understanding of the research object or problem, so as to gain a preliminary impression and perceptual knowledge, and at the same time provide the foundation and direction for more thorough and in-depth research in the future. Exploratory research usually uses participatory observation, unstructured interviews and other methods to collect data, and its research object is usually relatively small. The data and results of exploratory research are mainly used to "detect" the basic scope, content or characteristics of a phenomenon or problem, and "hint" the possible ways to study this phenomenon in depth.
Descriptive research: a type of research that describes the distribution of the whole research or some phenomena in some characteristics. Its focus is usually not on why there is such a distribution, but on answering what this distribution is like. In other words, its main purpose is to collect information, understand the situation and provide information. The description of social phenomena should pay attention to two aspects: ① accuracy; ② Generality. In descriptive research, data collection mainly adopts closed questions, and questionnaires are conducted by self-filling, mailing or structured interviews. The obtained data must be statistically processed, and various results mainly in quantitative form are obtained, and these results and conclusions are inferred into the whole.
Explanatory method; It refers to the type of social research that explores the reasons behind the phenomenon, reveals the inherent law of the occurrence or change of the phenomenon, and answers various "why". The goal of explanation method is to answer "why" and explain the reason and relationship, so its theoretical color is often stronger. It is usually based on theoretical assumptions, through in-depth field collection of empirical materials, and through the analysis of data to test assumptions, and ultimately achieve the purpose of theoretical explanation of social phenomena. Pay special attention to the applicability and pertinence of the research content. Binary and multivariate statistical analysis is often needed in analytical methods.
Theoretical research: refers to the basic knowledge that focuses on the social world related to development, especially the empirical research that focuses on establishing or testing various theoretical hypotheses. Its focus is to explore the causal relationship between phenomena, and its main goal is to increase people's understanding of the inherent laws of social phenomena and social things.
Applied research: focus on understanding, describing and discussing some realistic social problems or problems aimed at some specific social phenomena. Its focus is usually to quickly understand the actual situation, analyze the causes of phenomena or social problems, and try to provide targeted policy suggestions on this basis to help formulate social policies, solve social problems and evaluate social consequences.
Analysis unit: the research object of social research, that is, the object (person or thing) to be analyzed and described in the research. Five types: individuals, groups, organizations, communities and social products.
Hierarchical fallacy/cluster fallacy: refers to the phenomenon that in social research, researchers use a clustered analysis unit to conduct research and use a non-clustered analysis unit to draw conclusions. In other words, researchers use cluster analysis units to collect data, but use non-cluster analysis units to draw conclusions. For example, when a researcher collects the information of a certain cluster (city factory company) and then draws the conclusion of personal behavior from this information, he commits the cluster fallacy. To avoid making the mistake of block fallacy, the key is to ensure that the analytical unit you use when making a conclusion is the same as the analytical unit you use when using evidence.
Simplification theory: means that researchers use individual data to explain macro-level phenomena. For example, it emphasizes the relationship between personal achievement and national development (ignoring macro factors). Causing factors: social research is easy to obtain specific information about individuals, while objective unit operation is abstract and vague. The key to avoid these two kinds of mistakes is to ensure that the analytical unit used in drawing conclusions is the analytical unit used in using evidence.
Cross-sectional study: Cross-sectional study. Refers to the collection of research data at a certain time point, and is used to describe the situation of the research object at that time point, or to explore the relationship between different variables at that time point. For example, opinion polls of various contents, national census. This is the most common form of social research. Mainly exploratory research and descriptive research.
Longitudinal study: collect data at several different time points, describe the development and changes of phenomena, and explain the relationship between different phenomena. ① Types of longitudinal research A. Trend research; B. group study; C. follow-up study. (2) Characteristics of longitudinal research A. We can understand the development process of the phenomenon and compare the changes in different periods; B. Because the time sequence of various variables is clear, it is easy to make logical causal judgments; C. the investigation scope of longitudinal research is small, and different types are difficult.
Trend research: Trend research refers to the study of the changes of research objects over time. The purpose of trend research is to reveal and discover the changing trends and laws of social phenomena by comparing the attitudes, behaviors or conditions of the general population in different periods. Studying the trend of a crowd is actually equivalent to analyzing and exploring its development and change law by using the results of several horizontal studies on this crowd.
Cohort study: Cohort study is a longitudinal study. Cohort study, also known as population characteristic group or * * * identical characteristic group study, refers to the study of population characteristics.
Study on the change of special population with time. In this kind of study, the samples of each study are not the same, that is, the specific objects of each study can be different, but they must all belong to this special population. This special population is usually related to time or age.
Cohort study: also known as population study or follow-up study, is a study of the changes of the same group of people over time. Different from the above, it is always the same sample. The same group study is mainly used to explore the changing patterns and processes of people's behaviors, attitudes or intentions, and analyze various factors that affect this change. Because the same group of studies use the same samples every time, the changes over time are sometimes unpredictable. The difficulty is that it is difficult to obtain these samples in the follow-up study.
Triangulation: By observing things from different angles or viewpoints, they can determine their actual positions. This process is called triangulation. Including the following: ① triangulation of measurement, that is, researchers measure the same phenomenon many times; (2) Observer triangulation, that is, multiple observers or researchers will add multiple perspectives, personal backgrounds and social characteristics; (3) Theoretical triangulation, that is, using various theoretical perspectives to conceive research schemes and interpret data; Methods Triangulation, that is, mixing qualitative and quantitative types of research and data.
Double load hypothesis: the combination of two independent relationships shows that researchers are unclear and confused. Researchers should avoid using it. For example, regional poverty and high concentration of teenagers will lead to theft.
Tautology: A circular argument that occurs when someone seems to be saying something new, but in fact he is repeating what he said, or describing something that is "correct by definition". Tautology cannot be tested by empirical evidence. For example, someone is conservative because he thinks that government control should be reduced, but these attitudes are not the reason why someone is conservative.
Teleology: It comes from the ultimate goal and purpose and has many forms. An event initiated by God or God predetermines that it will happen. Teleology can not be verified by experience, which violates the temporary time series requirement of causality and lacks a real independent variable, because the "trigger" is extremely unclear. Many people confuse teleology with goal motivation. First of all, a person's state of consciousness can be measured by experience. Second, it currently exists. Finally, it can be compared with events that may or may not happen in the future.
False correlation: it means that the relationship between variables is wrong and an illusion. When there is a correlation between two variables, there will be a false correlation, but it is not a causal correlation, because there is actually a third factor that has not been detected, which may be the real reason for this relationship. /The third variable is both an independent variable and a dependent variable, which can explain the observed correlation. As far as causality is concerned, the undetected third factor represents a stronger explanation. Such as hair length and TV programs, and hidden variables (gender).
Six, measurement (Fiona Fang chapter 5, chapter 60, Newman chapter 7)
Measurement: Effective observation and measurement of the determined research contents or survey indicators. Specifically, measurement refers to the measurement according to A.
Fixed rules assign numbers or symbols to the characteristics of the research object (that is, the research variable), thus quantifying or classifying social phenomena.
Classified measurement: also known as category measurement or named measurement, with the lowest measurement level. Essence is a classification system, which distinguishes different attributes or characteristics of research objects, marks them with different names or symbols, and determines their categories. The mathematical feature is (not) equal to. It is symmetrical and transitive. Such as gender: male, female.
Ranking measure: also known as rank measure or order measure, its value can arrange the research objects in a certain logical order to determine their rank and order. In other words, sequence measurement is to divide objects into different intensity, degree or grade sequences according to certain characteristics or standards. The mathematical characteristics are greater than and less than. It has the symmetry, asymmetry and transitivity to distinguish categories. Such as city size: megacities, big cities, medium-sized cities, small cities, etc.
Distance measurement: equidistant measurement or interval measurement can not only distinguish social phenomena or things into different categories and grades, but also determine the distance and quantitative differences between them. The results can be added or subtracted. Such as temperature measurement.
Isometric measurement: also known as proportional measurement or proportional measurement, it has all the properties of the above three kinds of measurement, and has an absolute zero point with practical significance, and its data can be added, subtracted, multiplied and divided. Like the birth rate. Whether there is a non-intermediate zero point with practical significance is the only difference between fixed ratio measurement and fixed distance measurement.
Concept: it is an abstraction of phenomena and a subjective reflection of the attributes of a class of things. Concept is a brick of structural theory and a general expression of similar phenomena within the research scope. Concept is the product of the interaction between the objective world and subjective thinking. Concepts can be divided into entity concepts and non-entity concepts. The concept of entity refers to objects, things or phenomena that can be directly observed; The concept of non-entity is a concept that cannot be directly observed.
Variable: refers to a concept with multiple values or a concept with multiple categories. Variables are conceptual.
A type transformed by conceptual concretization. Variables have different states or attributes in different situations, reflecting the variability of concepts and explaining the changes of phenomena in scale, weight, density and speed. , or the process of phenomenon.
Poor degree of change. Judging from the influence relationship between variables, it can be divided into independent variables and dependent variables.
Indicator: refers to a set of observable things that indicate the meaning of a concept or variable, which is called a set of indicators of this concept or variable. Indicators are categories that reflect the changing characteristics of social phenomena. It is to concretize the stipulation of social category changes. It closely combines the qualitative and quantitative phenomena to describe a certain feature of social phenomena in detail.
Operationalization: replacing the internal facts about social structure, system or process and people's behavior, thoughts and characteristics with external facts representing them, so as to study the former through the latter. In other words, operability is the process of transforming abstract concepts into observable concrete indicators. It is a detailed description of the procedures, steps, methods and means used to measure those concepts with higher abstract level. This is one of the most difficult and critical steps in the process of social research.
Concretization: from abstraction to experience, concretizing concepts into observable things. The formation of the concept of 1. should refine the abstract concept; 2. Definition of concepts, including reducing the dimensions referred to by concepts, classifying basic units, and giving a descriptive definition of concepts; 3. Select the measurement index, that is, the specific measurement scheme of the concept; 4. Prepare comprehensive indicators.
Index: A simple cumulative score consisting of multiple different answers. It consists of a set of statements about attitudes or opinions about things. Respondents express their opinions on these statements respectively, and then add up the respondents' scores on all statements according to some standard to get the quantitative result of respondents' attitude towards this thing.
Scale: a quantitative tool to measure people's attitudes, opinions, opinions and personalities. It is a kind of composite measurement with structural strength order, that is, all statements or items are arranged in a certain structural order to reflect the different degrees of the concept or attitude to be measured.
Sum Scale: It consists of a set of statements that reflect people's attitudes or views on things. Respondents expressed their opinions, agreed or disagreed with these statements respectively, and scored them. Then add up the respondents' scores on all statements to get his attitude score.
Likert scale: a special form of total addition scale. It was improved by American social psychologist Hitt on the basis of the original total price scale. Richter scale is also composed of a group of statements about attitudes or views on things, but different from the total scale, respondents' answers to these statements are divided into five categories: "very agree, agree, don't know, disagree, very disagree" or "agree, compare agree, don't care, compare oppose and oppose". With the increase of the types of answers, the differences of people's attitudes can be more clearly reflected.
Social distance scale: Bogdas scale, also known as social distance scale. It came into being in the 1920s and was founded by American social psychologist Bogdas in 1925. A quantitative tool to quantitatively measure people's psychological distance, the degree of mutual relationship or people's attitude towards a certain group and the distance they keep. It consists of a series of statements with certain logical structure in content, and different statements represent different degrees of people's attitudes.
Semantic difference scale: used to study the different meanings of concepts to different people. Composed of two groups of adjectives with opposite meanings, researchers study people's views or attitudes towards a concept through the scores represented by the marks marked by the respondents in appropriate positions, or make comparative analysis between individuals or groups. Used for cultural comparative analysis, differences between individuals and groups, and people's attitudes towards the environment and things.
Sociometrics: Sociometrics, also known as social interaction measurement, is "a tool to evaluate interpersonal attraction or exclusion in a group or organization", or further, "a tool to determine the acceptance of individuals in a group or organization, discover the relationships among people in a group, and reveal the structural characteristics of the organization itself." Sociometrics was invented by American psychologist Moreno and others in 1930, which is mainly used to study the interpersonal relationship and group structure of small groups.
Reliability: that is, reliability refers to the consistency of the results, the consistency and stability of the measurement results, and whether the measurement results can measure the measured things or variables stably when the same object is measured repeatedly by the same method. Including: retest reliability, replication reliability and semi-reliability.
Test-retest reliability: Take the same measurement for the same object at different times and calculate the correlation coefficient according to the results of the two measurements, which is the test-retest validity. This reliability can indicate whether the results of the two surveys have changed or not, reflecting the stability of the measurement, so it is also called the stability coefficient. When using this method, the methods and tools used in the two measurements are exactly the same. The disadvantage of retest reliability is that it is easily influenced by time factors, that is, some events and activities between two measurements will lead to objective changes in the results of the latter measurement, so that the correlation coefficient of the two results can not well reflect the actual situation of the two measurements.
Reliability of the copy: the copy is relative to the original, it is a copy of the original. For a survey question, the respondents are asked to accept a questionnaire survey and a copy of the questionnaire survey, and then the correlation coefficient is calculated according to the survey results, so as to obtain the reliability of the copy. Using duplicates to obtain reliability can avoid the shortcomings of retest reliability, but the duplicates used must be true copies, that is, they are consistent with the original in terms of questions, forms, contents, difficulty and discrimination, but different from the original in terms of questions and words. Replica surveys can be conducted continuously or at intervals. The replica reliability of continuous implementation is also called equivalent coefficient, and the replica reliability of interval implementation is also called stability and equivalent coefficient.
Semi-reliability: the semi-reliability method divides the survey results into two parts according to the odd and even numbers of the questions, and then calculates the correlation coefficient according to the total score of each person in these two parts, and then obtains semi-reliability, which can check whether the questions asked in the questionnaire or scale are all aimed at the same research content.
Reliability of raters: check the reliability of raters, and calculate the correlation coefficient between a group of ratings of one rater and the same group of ratings of another rater.
If the reliability is low, it means that the measurement tools are not standardized and objective, and are easily influenced by subjective factors of raters. The way to reduce this influence is to delete extreme raters or raters, score by more than one person and calculate the average.
Validity: the validity or accuracy of measurement refers to the degree to which measuring tools or measuring means can accurately measure the measured variables, or the degree to which the properties of things can be accurately and truly measured. That is, the measurement standards or indicators used can truly reflect the true meaning of a concept. Meaning is the unity of content nature and degree.
Surface (content) validity: Content validity, also known as face validity or logical validity, refers to the suitability and logical consistency between the measurement content or measurement index and the measurement goal. It can also be said that measuring whether the selected project "looks" meets the measurement purpose and requirements. To check the validity of the content is to check whether the empirical deduction from concept to index is logically valid. We can only rely on people's subjective judgment and the same definition, because the understanding of a concept varies from person to person, but in scientific research, it is necessary to take the concept definition accepted by most scientists as the standard.
Standard validity; Actual validity or predictive validity refers to comparing the results obtained by the new method or index with those obtained by the original criterion when different measurement methods or indicators are used to measure the same thing or variable, and the correlation coefficient between them reflects the effectiveness of the measurement tools and means. There are three forms of standard validity: predictive validity, variable validity and actual validity.
Structural validity: To examine the validity of current measurement tools or means by using existing theories or propositions, which involves the measurement of other concepts or variables in the relational structure of a theory. For example, whether two variables with the same direction show the same directionality in the measurement results.
Predictive validity: Predictive validity refers to the standard validity established by standards or guidelines according to the actual situation in the future. Predictive validity is to compare the measured results with the actual situation in the future to check their consistency. For example, designing a foresight.
A measure or test of students' academic performance, which is used to measure students' academic performance when they graduate. If the students' actual academic performance at graduation is consistent with the predicted value, then this scale or test has predictive validity.
* * * Variable Validity: It is called * * * Variable Validity if the curative effect index is proved to be effective in a certain measurement method. * * * Validity of variables is used to judge whether other measuring tools can replace measuring tools as standards.
Practical validity: if the actual experience is used as the criterion, it is called practical validity. Actual effectiveness is usually used to check the actual effect of measuring tools. Because there was no certain standard for measuring some phenomena in the past, we can only test the effectiveness of measuring tools according to actual experience.
Intrinsic validity and external validity: measurement is the observation of a specific respondent at a specific time and place. If no problems are found in the test of content validity, criterion validity and structure validity of a specific study, it can be considered that the study has internal validity and its data and conclusions can effectively answer the questions studied. It can be said that intrinsic validity refers to the validity of the data and conclusions of a study. External validity refers to the universal validity of the conclusion of this study. The investigation of external validity should not only consider the representativeness and particularity of the sample, but also consider the universal significance of the time, place, situation and content of the study.