Author(s)

Fan Bu, Rongchang Ji, Zixuan Guo

Corresponding Author

Fan Bu

Abstract

There are more and more students finding a summer job to spend their summer vacation. Facing various summer part-time jobs in various fields, such as the field of the beverage industry, education, express delivery, sales, and even emerging online jobs, students do not know how to make a smart choice for these jobs. This research attempts to establish an effective model to solve this problem. Firstly, through searching the literature, this paper selected 6 main factors that influenced the job decision. Since summer jobs focused on a short-term process, not every factor had a significant impact on the short-term job. This is why, in this paper, the principal component analysis was used to screen factors, and four significant factors (work environment, work pay, entry requirements and workload) were selected based on the results, which could explain about 95% of the original data. Secondly, this paper adopted two methods (AHP and entropy method) to determine the weight of the index. Among them, AHP calculated the weight through the expert score and experience, which was more empirical, while the entropy method computed the weight based on the entropy value of the data, which was more objective. After the weight was calculated, the job score (individual's expectations for the job) could be calculated using TOPSIS method. Thirdly, after substituting actual data, two methods of AHP and entropy were compared and analyzed. The results show that in AHP, job score can effectively distinguish different types of work (Tutoring, 0.65-0.75; Programmer, 0.8-0.9; salesman, 0.55-0.7; waiter, 0.5-0.55); in entropy method, the entropy method does not differentiate the waiter and tutoring sufficiently (Tutoring, 0.65-0.8; programmer, 0.8-1; salesman, 0.5-0.6; waiter, 0.6-0.8). Through comparison, it can be seen that the AHP method is superior to the entropy method. Finally, a simple diagram of the job selection model was made for visualization to make the public easier to understand. When a user filled in his or her expectations for summer jobs, the model would output the corresponding score for the user.

Keywords

Principal component analysis; Ahp; Entropy method; Topsis