The problem of estimation of the population mean in the presence of response bias and social desirability bias is common in surveys. Researchers have proposed estimators of the population mean using auxiliary variables in the presence of response bias. However, little work has been done on estimation of the population mean using auxiliary attributes and variables in the presence of response bias. The present study addressed the problem of estimation of the population mean using auxiliary information in the presence of response bias and randomized response. The objectives of this study were to propose estimators of the population mean using auxiliary information in the presence of non-response and measurement errors, to extend the proposed estimators to estimation of the population mean using the three-stage Randomized Response Technique model, to derive expressions for the biases and mean square errors of the proposed estimators, to compare the performances of the proposed estimators to other existing estimators, and to assess the impact of response bias and the three-stage Randomized Response Technique on estimation of the population mean. A generalized class of estimators was proposed using auxiliary information. Up to the first order of Taylors series expansion, biases and mean square errors of the proposed estimators were derived. The efficiencies of the proposed estimators were studied theoretically and numerically using real data. The values of mean square errors increased with an increase in non-response rate, inverse sampling rate, and sensitivity level of the survey question. The proposed estimators performed better than other existing estimators. The proposed estimators are useful in various fields, such as social sciences, agriculture, and education.