Must the questionnaire be tested for reliability and validity? If you have multiple-choice questions, how to do reliability analysis, please ask experts for help.

There is no need to test the reliability and validity of the questionnaire: the reliability and validity analysis is mainly aimed at the scale questionnaire, and it is not suitable for the reliability and validity analysis if only some objective realities (such as age, gender, occupation, vehicle, salary, etc.) are involved. ) was investigated.

If multiple-choice questions are included, reliability analysis cannot be done. The reliability and validity of the dimensionless questionnaire are mainly verified by experts, and there is no need for reliability analysis.

Disadvantages of reliability measurement: because it involves the sum of squares of errors, it is necessary to repeat the measurement for the same respondent for many times, and then get the reliability estimation based on the estimation of T and X values ... However, repeated testing will bring memory effect and connection effect, which will arouse the resentment of the respondents, and it is difficult to achieve it in investigation and research.

Extended data:

The role of effectiveness:

1. Prediction error: The actual meaning of the validity coefficient is often expressed by the decisive coefficient, and the square of the intentional correlation coefficient indicates the proportion of the standard variance for testing the correct prediction or explanation to the total variance.

Second, the forecast standard is divided: the forecast standard is divided from the forecast.

For example, if two variables x and y are linearly related, as long as the regression equation between them is determined, one variable can be estimated from the other variable.

Third, the prediction efficiency index: the error that can be reduced compared with blind prediction.

Baidu Encyclopedia-Reliability

Baidu Encyclopedia-Effectiveness