Professor rating has become an important aspect in university. While there are many types of professor rating, RMP ratings has been encouraged. The rankings in campuses provide important insights on similarities in departments or across schools. Understanding the department quality in an institution is directly affects the quality of education provided to students. Technically, students spend most of their time enrolled in a class offered by a single department for their major. Thus, being able to compare departments or school quality and department would provide potential results for making informed decision about the current level of education.
This project was intended to provide important skills in analyzing data using python programming language. The project analyses sqlite data consisting of professors’ names, descriptive tags, reviews, attractiveness effect ratings, departments etc. the variables will be analyzed and visualized to provide meaning (Sena & Crable, 2017).
RateMyProfessors.com (RMP) is a website that provides information about institutions and educator rating. I collected from the website to provide prospective information about which professors should be considered and which ones to be avoided.
I was able to rate their institutions (university) using 10 metrics. The ratings are based on attributes such as food, safety, happiness, reputation, professor quality and difficulty. Students were also able to descriptive their professors based on their personalities: hilarious, inspirational, and tough grader. Students also summited text reviews that are entirely pen ended. The data comprised of both genders to have an all-inclusive opinion.
Since students rated the professors themselves, there were similar departments given different names such as Biology and Biology Sciences. Thus, after department names were grouped, final data consisted of 35 departments.
Data Analysis and Visualization
The data also shows that the difficult professors had lower average overall rating compared to the less difficult ones (Owen, 2017).
Descriptive tags attached on the reviews showed that some professors were not preferred by students because of assigning them lots of homework. The tag was however more in some departments compared to others. Of the available tags, 25% of the reviews for Accounting professors were more by about five times.
I conducted the wordcloud analysis. The visual representation was recommended since it would show the frequencies of terms in open-ended reviews. Most frequent term was good and easy. A good percent of students stated that they learn, homework and understand. Figure 2 shows the frequencies.
Figure 2: Frequency wordcloud
The program can also be developed to predict the best university or professor. Python allows predictions through scikit-learn model. The model learns from available data through mapping input features in the data. Prediction of new data instances using predict() and the LogisticRegression.
Professor rating is an important factor in the education sector. It is a tool that can be used to help decide which college would be best for a given course. Through examination of student opinion, both the management and students can be able to make informed decisions. The rating methods can be based on different valuation methods.