This study examined the differences and patterns for three categories between an argument-based inquiry group and a traditional group over the period of the SWH (Science Writing Heuristic) project: (1) teacher talk time, (2) structure of questions (question types), and (3) student responses. The participating teachers were chosen randomly by a convenient sampling method because the data were collected previously from the SWH project. Each group had thirty teachers. A total of sixty teachers participated in the study. Student responses were part of the study to evaluate the effect of open-ended question types but students were not direct participants in the study. Each teacher was asked to send a recorded video clip of their class at the end of each semester (spring and fall) over two years. Each teacher sent four video clips for the project. A total of two hundred forty video clips were analyzed to gather the information regarding the three categories. The first category was teacher talk time. It was measured in seconds only when teachers interacted with students with the topic. The second category was the structure of questions (question types). It consisted of two question types (open-ended and closeended). Under the open-ended question category, there were three sub-question types: (1) asking for explanation (AE), (2) asking for self-evaluation of reasoning (AF), and (3) asking for selfevaluation of others' reasoning (AFO). Under the close-ended question category, there were two sub-question types: (1) asking for factual information (AI) and (2) asking for confirmation (AC). Each sub-question type was counted numerically. The last category was student responses. Student responses consisted of higher-order thinking and lower-order thinking. Under the higherorder thinking category, there were three sub-types: (1) explanation responses (E), (2) selfevaluation of reasoning responses (SE), and (3) self-evaluation of others' reasoning responses (SEO). Under the lower-order thinking category, there was one sub-type: simple responses (S). v Each sub type was counted numerically. Based on the descriptive results (the length of teacher talk time in seconds, the number of question types, and the number of student responses), repeated measures ANOVA was conducted to find any differences and patterns for teacher talk time, structure of questions and student responses between the treatment and control groups over the period of the project and across time (four different time points). The results showed that there were clear differences for teacher talk time, the structure of questions, and student responses between the treatment and control groups over the period of the project and at each time point. The treatment group teachers talked less and used more open-ended questions than the control group teachers. The treatment group students displayed more higher-order thinking responses than the control group students.