Innovative Inter Quartile Range-Based Outlier Detection and Removal Technique for Teaching Staff Performance Feedback Analysis

The teaching-learning process plays an important role in education. To improve this process valuable and timely feedback is taken from the students. That feedback can be used for two main purposes: one is to improve the process according to student expectations and another is to evaluate the teaching faculty's performance for the sake of appraisal. However, some students can give unfair and biased feedback. Such feedback may produce an adverse effect on appraisal. To remove these anomalies generated due to favoritism and biasness in the feedback innovative interquartile range (IQR) based outlier detection and removal technique is implemented in this article. The proposed technique removes the outliers based on IQR and precisely selects the central tendency (mean or median) from the feedback data distribution based on skewness. Then the identified central tendency will be considered to compute the appraisal indicator. To conduct the experiments feedback data is collected from the students via questionnaire. The questionnaire is prepared by expert academicians. The questionnaire contains twenty-one questions which are divided into five categories. By confirming experimental results, the proposed IQR-based outlier detection and removal technique removes the outliers from the data set and improves the performance analysis which intern helpful for teaching faculty performance appraisal.