Unable to manipulate independent variables

In the process of collecting data, we have to face reality

Main points

evaluate

Indicators and meaning

evaluate

Independent variable: Stimulus variable, which is a variable selected and controlled by the experimenter. It determines behavioral or psychological changes.

Dependent variable: The subject's response variable, which is the result of the independent variable and is the behavioral variable observed or measured by the experimenter.

Advantages: The experimental design and implementation are simple; the number of subjects receiving each treatment level can be unequal; there is no need to match subjects; each subject only receives one treatment level;

Disadvantages: not all within-group variation consists of random errors, including individual differences among subjects;

Advantages: The effect of an additional variable is separated from the total variation, thereby reducing experimental error and obtaining a more precise estimate of the treatment effect;

Disadvantages: If the experiment contains many treatment levels, it may cause difficulties in forming homogeneous intervals and seeking homogeneous subjects;

Advantages: fewer subjects; more effective control of individual differences;

Disadvantages: fatigue effect; sequence effect; practice effect;

Applicable situation: There are two independent variables in the study, each independent variable has two or more levels; the experimental combination contains p*q combined treatments, and the researcher is interested in the combined effects of all treatment levels;

Basic method: randomly assign subjects to receive a combination of experimental treatments, with each subject receiving a combination of experimental treatments;

Differences from a single factor; the interaction between two or more independent variables can be evaluated to obtain richer information than a single factor; each subject receives a combination of treatments rather than a treatment level ;

Applicable situation: There are two independent variables in the study, each independent variable has two or more levels; the experimental combination contains p*q combined treatments, and the researcher is interested in the combined effect of all treatment levels; there are An additional variable that the researcher is not interested in, and there is no interaction between this additional variable and the independent variable, and the researcher hopes to isolate the variation of this additional variable;

Basic method: subjects are matched on an additional variable in advance (if this variable is a subject variable), and then each selected group of homogeneous subjects is randomly assigned, and each subject receives a combination of experimental treatments

Applicable situation: There are two independent variables in the study, and each independent variable has two or more levels. Both independent variables in the study are within-subjects.

Basic method: In an experiment, each subject is allowed to accept all treatment levels of a variable. This is a repeated measurement experimental design. When one factor is repeated measurement and the other factor is non-repeated measurement, it is called a mixed design. When both factors are repeated measures, it is called a within-subjects design.

Definition: When an experimental design contains both non-repeated measurement factors (between-subjects factors) and repeated measurement factors (within-subjects factors), it is called a mixed factor design.

Applicable situation: There are two independent variables in the study, and each independent variable has two or more levels. In the study, one independent variable is within-subjects, that is, each subject receives all levels of treatment, and the other independent variable is between-subjects, that is, each subject only receives one level of treatment, or itself. is a subject variable. Such as gender, age, etc. Researchers are more interested in the treatment effects of the within-subject factors in the study and the interaction of the two factors, hoping to estimate them more accurately.

Basic method: first determine the within-subjects variables and between-subjects variables in the study, randomly assign subjects to each level of the between-subjects variables, and then make each subject receive a test combined with a certain level of the between-subjects variables. All levels of the within-subjects variable.

between-groups design

within-group design