Current location - Recipe Complete Network - Complete recipe book - How to design experiments
How to design experiments
Question 1: How can we have a good experimental design? One fifth, determine a research question.

The question to be noted is:

1 to determine whether this problem is worth studying.

2. To determine the difficulty of this research depends on whether the existing technology and means can carry out this research;

3. Look at this problem that other researchers have not found or missed.

4, the problem should be clear, so that the problem can be easily transformed into experimental hypothesis, and an operational definition of the problem to be studied is given.

We can determine the problems to be studied from the following aspects:

1. The need of reality.

2. Theoretical sources

(1) Choose the topic according to the contradiction between different theoretical viewpoints.

(2) Choose the topic by questioning the existing theoretical viewpoints.

(3) Propose relevant topics to prove a certain theoretical viewpoint of others or yourself.

3. Research literature-contradictions, neglect, negligence, or problems that could not be solved before, but can be solved now.

4. Select the topic from the new progress of theory, method and interdisciplinary.

Second, form an experimental hypothesis and express it in clear language.

Assumptions are guesses based on existing facts and principles. Scientific hypothesis is the product of theoretical thinking based on certain professional knowledge and experience. It is embodied in a knowledge system. The process of defining independent variables, dependent variables and their relationships is the operational process of the research topic. After the hypothesis is formed, the specific experimental design and experiment can be implemented.

(A) the type of hypothesis

1. Predictive hypothesis: refers to making speculative judgments on some situations of objective things, especially differences.

2. Relevance hypothesis: refers to the speculation and judgment on the specific direction and closeness of the mutual connection of objective things.

3. Causality hypothesis: refers to the speculative judgment of causality between objective things.

(2) Assumption standard

(1) Assumption should be scientific. Assumptions must be guided by certain scientific theories, based on certain scientific experiments, based on certain scientific facts, and after certain scientific argumentation, which is by no means groundless subjective speculation.

(2) Assumptions should be predictable. The hypothesis that the relationship between two or more variables should be inferred remains to be tested.

(3) Assumptions should be feasible. Assumption should be highly operable, and the concepts contained in it should be manipulated by independent variables and dependent variables.

(4) Hypothetical language should be simple and clear, and should be clearly explained by declarative sentences. Hypothesis generally adopts the sentence pattern of "if-then-"or directly states the possible research results according to the relationship between independent variables and dependent variables.

(5) Hypothesis can include the potential relationship between the variables to be studied, and it must be based on facts, which can well explain the existing assumptions and predict future events.

Third, determine the respective variables in the experiment, including the determination of independent variables, dependent variables and irrelevant variables.

The question to be noted is:

1, determination of independent variables:

(1), the operational definition of independent variable, accurately describes the phenomenon with observable and measurable events, things or phenomena. Such independent variable is objective and is conducive to others' reference and repeated verification of experimental results.

(2) The independent variable will cause the change of the dependent variable we want to study.

(3) Independent variables must be variable.

(4) During the experiment, appropriate checkpoints should be set, and the interval between checkpoints should be at least greater than the just perceptible difference, so as to ensure that when the independent variable changes, the dependent variable will also change accordingly.

In addition, when presenting independent variables, we should also pay attention to the size, presentation mode, spatial position, presentation sequence and presentation time of independent variables.

2. Determination of dependent variables:

The choice of dependent variable must have some inherent properties, which is the standard for us to choose a good dependent variable.

(1) validity, the change of dependent variable is really caused by the change of independent variable. The result of the dependent variable measured from the sample can represent the population in which the sample is located.

(2) Reliability, namely, the consistency of the change of the dependent variable when repeated measurements are made, with other conditions unchanged.

(3) Sensitivity, the dependent variable can sensitively reflect the response changes of the organism caused by the dependent variable.

(4) Quantification, quantifiable indicators can reflect the changes of response variables in more detail, collect objective data, and carry out it scientifically. & gt

Question 2: How to design an experiment between subjects? Generally, an experiment needs more than 30 subjects, and more than 30 people can be regarded as a large sample, so the reliability is reliable.

However, it is difficult to find 30 subjects in this experiment, and autistic children should not be so easy to find. However, there is also a solution for small samples, which is to design inside the subjects, that is, each subject has gone through all the processing levels. More than a dozen themes are used in the literature, which is probably the same (the requirement for the number of themes in the design is far lower than that in the inter-theme design). I feel that a dozen will do. According to the experiment, normal children can be used as the control group.

Question 3: How to design the experimental group of innovative experiments in experimental psychology: Let 10 subjects eat cake with excrement core, and the subjects randomly sample.

The reference group 1: 10 asked the subjects to eat cake in the shape of feces, and the subjects randomly sampled.

Reference group 2: 10 The subjects were given normal cakes, and the subjects were randomly sampled.

The purpose of the experiment is to see which group vomits first.

Adopt me, dear.

Question 4: How to design psychological experiments, zhidao.baidu/...=query?.

blog. 163/... 1 1308/

These are all examples.

Feeling is to design independent variables and dependent variables, control variables after work, find out the data distribution you want, write conclusions and improve methods.

If it is just an exercise, then the suggestion is: turn the abstract things in the topic into our common life scenes, and then think about what factors will change the occurrence of this scene. Of course, if it is a big paper or something, you must carefully consider the selection of materials, because first of all, you must ensure that there is no repetition. ...

Question 5: Experimental design of experimental psychology-Exploring the independent variable of facial emotional state of facial muscle sports teams related to laughter: facial muscle movements with different expressions. According to six basic emotions, it can be divided into six levels.

Dependent variable: the emotional state of the subject.

Experimental design:

It is easy to control the independent variable, just let the subjects make some kind of expression directly, but don't show him a picture of facial expression, otherwise the picture will be in * * *.

The first choice of the dependent variable is the self-report of the subjects. Because this is more real, the difficulty is not easy to quantify.

You can also choose a variety of physiological indicators, such as heartbeat, pulse, perspiration, skin temperature and so on. And directly measured by a multichannel physiological instrument. The advantage of this is more objective quantification, but the disadvantage is that there is still debate about what kind of emotions physiological indicators represent. I tend to use the latter, after all, objectification of experimental indicators is the best.

In the specific operation, it is best to use the internal design of the subjects. A subject made six kinds of expressions, one expression at a time, and recorded his physiological indexes. After an expression is finished, take a break to eliminate the subsequent influence of emotions. Then move on to the next one.

The order of each subject's six expressions should be random, not all in the same order.

That's all for the time being ~ ~ ~