This study examined visitor behaviors with a hands-on exhibit for children under two conditions: when visitors chose to view a scaffolding video compared to when they did not. The premise of the research was based on two assumptions. First, video can mirror more closely than any other tool the actions of a facilitator in a museum exhibition. Second, video can provide more information through a lower threshold (quicker and easier access to understanding) than a traditional text label. At the time of this study, research on the use of video in museums was limited.

The study took place at the Children's Museum of Houston. Two scaffolding videos were created and installed at a hands-on exhibit. Four data collection methods were used: direct observation, indirect observation through audio-video recordings, paper surveys, and exit interviews. Although study participants ranged from young children through adults, the engagement of 7–13 year olds was of primary interest.

Data were coded and then analyzed according to differences in visitor interactions in the two study conditions: video viewed and video not viewed. The analysis followed three study sub-questions which were related to video viewership frequencies, length of engagement time and quality of interactions when adults were present. Results indicated that visitors viewed videos more frequently than not; visitor engagement time was longer when a video was viewed; and adult conversations with children were more educational when a video was viewed.

The study results clearly indicated that the differences in visitor interactions when a video was viewed were positive. However, this statement was made in light of two major factors exposed in the study that need further investigation. Engagement length was significantly higher when an adult was present and also when a scaffolding video was viewed. These factors need to be isolated in further study to more clearly identify the additive effect of video. In addition, this study did not measure the length of time that a video was watched after being activated. Therefore, it was not known how much of the difference in visitor engagement time was related to viewing the video as compared to hands-on interaction.