When to Use Which t-test?

Research Scenario and Test Choice

A hypothetical research scenario could involve investigating whether the use of ChatGPT improves the academic writing scores of EFL students at HUFLIT University.

In this study, there are two groups of participants:

• Group 1: students who use AI writing tools such as ChatGPT for their writing assignments
• Group 2: students who complete the same assignments without using AI tools

These are two separate and independent groups, meaning that the students in one group are different from those in the other group. No participant belongs to both groups.

Because the data come from different participants, I would choose an independent samples t-test. This test is appropriate when comparing the mean scores of two unrelated groups to determine whether there is a statistically significant difference between them.

For example, the writing scores of the AI group can be compared with those of the non-AI group. If the p-value is less than .05, we may conclude that the difference between the two groups is statistically significant.

I would not use a paired samples t-test in this case because that test is designed for data collected from the same participants measured twice, such as pre-test and post-test scores.