2 Education statistics are divided into descriptive statistics and inferential statistics.
Descriptive statistics: sorting, classifying, summarizing and expressing the data obtained from the survey to quantitatively describe the characteristics of the sample or the population. The content includes data grouping, data calculation and simplification, description of data concentration number and difference number, and correlation analysis.
Inferential statistic: Study how to infer the overall situation from the information provided by local data. The main contents include parameter estimation, hypothesis testing, variance analysis and chi-square analysis.
Gao Erdun: It was British anthropologist and biologist Gao Erdun who first applied statistics to psychology and education.
Thorndike: Introduction to Psychometrics is the 1 monograph on educational statistics in the world.
7 measurement: it is a process of quantitatively describing the characteristics of things according to certain laws.
The two basic elements of measurement are the unit of measurement and the reference point. The unit of measurement needs two conditions, one is clear meaning, and the other is equal quantity.
The reference point of measurement is the starting point of a quantity. There are two kinds of reference points, one is absolute reference point and the other is relative reference point.
10 educational measurement: in a narrow sense, it refers to the process of quantitatively describing students' academic performance and psychological characteristics through tests, and in a broad sense, it refers to the process of describing the characteristics of various things or phenomena in the field of education with a specified amount.
1 1 Characteristics of educational measurement: purposiveness, indirectness and uncertainty (specifically randomness and fuzziness).
12 China is the hometown of educational measurement, which originated in the Western Zhou Dynasty. The real rise of educational measurement was after the 20th century, represented by Thorndike.
13 the significance of statistics and metrology in learning education: 1, scientifically evaluate students' learning progress and provide a basis for the improvement of education and teaching, 2, quantitatively analyze the factors affecting students' learning and find effective improvement strategies, 3, strengthen quantitative analysis and promote scientific education research,
14 data classification: according to the source of data, data can be divided into three types: counting data, measurement evaluation data and manual coding data; According to the measurement level, the data can be divided into four types: nominal data, sequential data, equidistant data and proportional data. According to the form of data distribution, data can be divided into discrete data and continuous data.
The counting data of 15 is obtained by counting the number or times, which are mostly expressed as integers, such as the number of pulse classes.
16 measurement evaluation data is the data obtained by evaluating some properties of things with the help of measurement tools or evaluation methods, such as test scores.
17 manually coded data is data formed by people assigning corresponding numbers to different kinds of things according to certain rules, such as job numbers.
The naming data of 18 only shows the differences between one thing and other things in name, category or attribute, but does not show the size, order and quality of the differences between things, such as one male and two female.
19 sequential data refers to variables that can arrange things in order according to the number or size of a certain attribute, which has the characteristics of hierarchy and persistence. Like ranking.
Isometric data not only represent different categories and their sequential relationships, but also have isometric measurement units, such as temperature.
2 1 equidistant data not only has nominal order and equidistant, but also has absolute zero, which means that the digital zero in equidistant data indicates the lack in practical sense, such as height and weight.
Characteristics of data: discreteness, variability and regularity,
23. Principles of data collation: 1, classification marks should be based on the research purpose, 2, each classification mark should be unidirectional,
24. Frequency distribution: it is a statistical process of data occurrence times, which refers to the times when different values appear in a batch of data or the times when different values appear in a batch of data.
25. Simple frequency distribution table, frequency distribution table, reflects the frequency distribution structure of a batch of data on each equidistant block.
26. The main steps of compiling a simple frequency distribution table are: 1. Find the total distance r; 2. Determine the number of groups K=? , three group intervals (5 or multiples of 5), four group limits, five group median values, six group classification marks, seven registration times,
27. Relative frequency distribution table: it is the ratio of each group frequency f to the total frequency n, which is represented by the symbol Rf.
28, cumulative percentage distribution table, cumulative relative frequency distribution table, cumulative percentage distribution table.
29. Double histogram, that is, there are some patterns formed by closely arranging straight bars with different heights and equal widths on the coordinate axis, which can present the distribution characteristics and structural forms of data more intuitively and make people clear at a glance.
30. The degree polygon diagram is a graphic method to form polygons with closed broken lines and the change of reaction time (similar to the imaginary point extending forward and backward by one unit on the basis of dotted statistical diagram).
3 1. relative frequency histogram and relative frequency polygon, cumulative percentage distribution, cumulative relative frequency curve and cumulative percentage curve,
32. A scatter chart, also known as a scatter chart, represents the correlation and contact between two things by the dispersion degree of points in a plane rectangular coordinate system, with the horizontal axis representing independent variables and the vertical axis representing dependent variables.
33. Line chart: A statistical chart showing the development, change and evolution trend of things with undulating broken lines, which is mostly used for continuous data, such as the evolution trend in time series, or to describe the trend of one thing changing with another.
34. Bar chart: A bar chart, also known as pareto chart (only used for nominal variables), uses the length or height of bars with the same width to represent the quantitative relationship among various statistical items, and arranges them in order of times.
35. The prototype diagram, also called C diagram, is a graph that represents statistical matters by the percentage of each sector area in the unit circle to the whole circle area, and accounts for a corresponding proportion in the whole.