Analysis Of Gui Testing And E-Recruitment Site Performance Using Katalon Studio And Jmeter With Two Way Anova Method

Abstract

A Graphical User Interface (GUI) is a method of interaction between users and software that displays a graphical interface easily understood by users when operating a system. It can measure the quality of a site. However, in addition to the GUI that can affect the quality of a website, performance can also affect the quality of a website. Performance is a process carried out to evaluate the performance of a website when the traffic load is high. This study aims to analyze the differences in GUI quality and performance on five e-recruitment platforms, namely Jobstreet, LinkedIn, Karir.com, Glints, and Kalibrr, using the Two-Way ANOVA method. GUI testing with Katalon Studio showed that JobStreet and Karir.com had high response times on the Login and Profile features due to difficulty in recognizing complex elements. Glints failed on the second and third tests in the Sign Up feature, while LinkedIn showed a high response time due to difficulty in recognizing attributes in the Search for Jobs feature, and Kalibrr appeared stable. Performance testing with JMeter, Jobstreet, Karir.com, and Kalibrr showed stable performance with low response time, stable throughput, and a 0% error rate. Glints experienced a 100% error rate because access was denied with a 403 code of “Forbidden”, while LinkedIn showed a spike in error rate as the number of threads increased. Two-way ANOVA analysis showed that in GUI testing, there were significant differences in response time and success rate based on application and feature. In performance testing, response time and error rate also showed significant differences, but throughput did not show significant differences by application and feature.