Comparing instructor-led, video-model, and no-instruction control tutorials for creating single-subject graphs in...

Area of Science:

• Behavior Analysis
• Educational Psychology
• Data Visualization

Background:

• Visual inspection of single-subject data is crucial for behavior analysts.
• A lack of consensus exists regarding optimal methods for teaching graph construction in single-subject designs.
• Previous research by Tyner and Fienup (2015) provides a foundation for this study.

Purpose of the Study:

• To compare the effectiveness of instructor-led, video-model, and no-instruction control tutorials on graphing performance.
• To extend previous findings on teaching graph construction for single-subject designs.
• To evaluate the impact of different instructional methods on master's students' graphing skills.

Main Methods:

• A systematic replication and extension using a repeated-measures between-subjects design.
• 81 master's students with prior Microsoft Excel experience participated.
• Comparison of three tutorial conditions: instructor-led, video-model, and a control group.

Main Results:

• A statistically significant main effect was found for pretest, tutorial, and posttest submissions across all groups.
• The main effect of the tutorial group itself was not statistically significant.
• A significant interaction between tutorial group and submissions indicated instructor-led and video-model tutorials were superior to the control condition.
• Training positively impacted performance on both trained and untrained graph types (multielement graphs).

Conclusions:

• Both instructor-led and video-model tutorials are more effective than written conventions for teaching graph construction to graduate students.
• Instructional methods significantly influence graphing performance in single-subject designs.
• Training generalizes to untrained graph types, suggesting robust skill acquisition.