The larger the ratio, the better, the more accurate the model and the more significant the regression effect. R squared is between 0 and 1, and the closer it is to 1, the better the regression fitting effect is. Generally speaking, the goodness of fit of models exceeding 0.8 is higher.
Extended data
The higher the square of r, the more suitable the model is for your data. In psychological investigation or research, we usually find that the low R square value is lower than 0.5. This is because we try to predict human behavior, and it is not easy to predict human behavior. ?
In these cases, if the R-squared value is very low, but there are statistically significant independent variables (also known as predictive variables), it is still possible to understand the relationship between the changes in the values of predictive variables and the changes in response values.
When the horizontal line explains the data better than your model. Mainly occurs when intercept is not included. If there is no intercept, the regression may be worse than the sample mean when predicting the target variable. This is not only because there is no interception. Even if intercept is included, it may be negative. Mathematically, this is possible when the error square of the model is greater than the sum of the total squares of the horizontal lines.
Baidu encyclopedia -R plaza