Exploratory Learning

Studying the Effects of Environment on Child Engagement in Museum Exhibits

​

​

1. Abstract + Project Summary 

​

        The goal of this project is to study the factors that influence children’s engagement in informal education (e.g. museum exhibits). We want to observe how children interact with museum exhibits, and if the time they spend engaging in the museum exhibits is affected by the kind of learning environment that exhibit creates. 

        My mentors, Zoe Liberman and Diane Lee, from the Social Cognitive Development Lab at UCSB, and I gathered data by asking participants (roughly between the ages of two and ten) to wear a GoPro on their head as they interacted with a museum exhibit. We recorded data from participants in two children’s science museums: the MOXI: The Wolfe Museum of Exploration and Innovation, in Santa Barbara, and Papalote Museo del Niño, in Mexico City. We set up an identical exhibit in either museum where kids were encouraged to build a paper rocket and launch it into the air. Each participant’s building and launching process was recorded by the GoPro. While the Papalote exhibit had a staff member present to guide the children through building their rocket, the MOXI exhibit did not. This difference allows us to compare the results from each museum to determine if a particular environment (staff member vs no staff member) increases the amount of time the children spend engaging in the exhibit. 

    The data videos we gathered were coded for different actions and behaviors by myself and other undergraduate students working on the project. The code then helps us determine how long children stay engaged in the exhibit. Even though this project isn't completely finished, our goal is to use what we learn from our results to work with museums and other informal education settings in maximizing children’s engagement in tasks and learning objectives.  

 

​

2. Introduction 

​

        Child engagement in informal education can be affected by numerous factors. One of the biggest factors is direct vs. indirect instruction. Direct instruction is an environment where a teacher, instructor, or staff member is present to guide children through the learning objective, whereas indirect instruction expects children to follow a simple set of instructions and complete the learning objection on their own. Staff members, parents, and peers can be present for aid in either environment. This difference in instruction is the biggest factor we wanted to study, which is why each of our museums represented one of these two types of instruction. Other factors that we wanted to consider were guidance from parents and staff members; certain cultures can foster different kinds of relationships between parents or other figures of authority and children, leading to a difference in how children ask for help (Briseño-Garzón).

          Previous early-education studies have found that children tend to be more creative in a less-structured learning environment, but they tend to be more successful at accomplishing an objective in an instructed learning environment (Legare, Bonawitz). Differences in the way the objective (for example, building a rocket) is communicated can influence the amount of effort the participant puts in or how long they stay engaged with the task. Children can infer information from their peers, parents, or teachers, as well as decide to follow or not follow guidelines set out for them. Thus, their engagement in certain tasks can be difficult to measure and often unpredictable. 

          After doing preliminary research on the topic, I began to suspect that children would engage more with exhibit materials if there was an instructor or staff member present, who could encourage the child to stay focused and accomplish an objective. However, I was eager to see if the results of our research project would prove or disprove my theory.

 

 

3. Methods 

​

          To gather data, we set up an identical exhibit in each museum, which encouraged children to build a paper rocket and then launch it (the stomping mechanisms in both museums were similar to a stomp rocket). Each child was provided with paper, scissors, and tape to build their rocket. 

         We sent researchers into either museum with GoPros to gather data. When a child seemed interested in the exhibit, we asked if they would like to take part in the study and wear a GoPro on their head. It was important that we got both the child’s consent as well as the parent’s consent to be filmed. Once we got the consent, we gave the participant the GoPro to wear on their head, and it would film their rocket-building and launching process. We gathered about eighty participant videos from the MOXI museum and thirty-five from the Papalote museum. We also asked parents to fill out a survey, asking them about their background in education, their opinions on the effectiveness of the staff members, and why they brought their children to either museum in the first place. Though we have not taken a closer look at those results quite yet, we believe that knowing this background information will help us better understand the public opinion of informal education settings, and how effective they are. 

         After we gathered the videos, our next step was to code them. To do so, we created a thorough coding manual that clearly defined every parameter we wanted to code for

. 

<ID> — participants subject number, birthday, test date, and gender

<LOOK> — what the participant is looking at (could be another person or a part of the exhibit)

<TOUCH> — who is touching any of the materials (child, parents, staff, other)

<STOMP> — how many times the rocket is launched, and by who (child, parents, staff, other)

<TALKTYPE>  — questions, answers, exclamations, commands

<WORDS> — a word-to-word write up of the conversations taking place during the video 

<EFFORT> — the effort put in by the participant (improvements, stomp attempts)

<COMMENTS> — to highlight any anomalies. 

​

We also set letters to identify the person completing an action, so that the coding process was fairly quick. 

c — Child (participant)

m — Mother 

f — Father (sometimes coded as “d”)

s — Staff 

p — Peer (other children in the exhibit)

o — Other 

​

         Before we began coding data, we assumed that coding the videos for all of the parameters would take about an hour to an hour and a half. However, when I began coding the first set of data videos we received from the Papalote museum, we found that coding each video for all 7 parameters (excluding ID and comments) took about six hours. Thus, we decided to break the task down into sections that would be split amongst myself and other undergraduate students working in the lab. My job became coding <TOUCH> and <STOMP> for all of the videos. 

         The program we used to code our videos is called Datavyu. It's a user-friendly video-processing program that allows the coder to scrub through a video and code different actions. Once we uploaded the coding manual parameters into the Datavyu spreadsheet, we could shuffle through each participant video and enter new ‘pieces’ of code whenever the participant exhibited a behavior we were targeting. Each piece of code included onset and offset time, as well as an identifier (who was doing that action). 

          After we finished coding a video, we exported it as an Excel file where we could find averages and compare those between museums. We used Excel to find total <TOUCH> times for all of the people present in each video, as well as the total number of stomps, and compiled those values onto a master Excel document, where we could record and compare the values from all twenty videos sampled. 

 

​

4. Results + Discussion 

​

          Since the project is not yet complete, we were only able to sample ten videos from either museum. Unfortunately, the coding process for all seven parameters is not yet complete, so we were not able to take those factors into consideration. However, we were able to compare the <TOUCH> and <STOMP> parameters from these videos, which at least gave us a good sense of how long children are physically interacting with the exhibit. 

From those twenty videos sampled, we actually found that children were spending roughly the same amount of time engaging with the materials in either museum. We also noticed that the parents tended to spend more time engaging with the materials in the Papalote museum than in the MOXI Museum. Though these results went against my original hypothesis, they are very suggestive that children are staying engaged in the exhibits and doing the majority of the building.  

        However, since these results are pulled from only a sample of the data, we have to wait to see if this trend will carry over once we’ve coded the rest of the videos. We also decided to look at the proportion of time the child spent touching the materials in comparison to the total touch time of everyone in the video, and actually found that the materials were being touched by the participant the majority of the time. We were also pleased to find that the child proportions for either museum were relatively close. Once again, we must wait until the rest of the videos are coded and compared before we can draw any definite conclusions. 

        We also compared the results of the <STOMP> code, and we found that kids were launching their rocket more in the Papalote museum.  However, we were unsatisfied with these results, so we decided to compare the number of stomps in a different way. Instead of taking an average for each museum, we separated all twenty participants into their engagement proportion. We then took the <STOMP> averages for all of the children who engaged less than 85% of the time and compared those to the <STOMP> averages of the children who engaged more than 85% of the time. 

        We found that the kids who engaged more with the materials beforehand tended to launch their rocket more. This could be a result of children being excited to see how their rocket will fly; they could be eager to test what their hard work has produced, which could be a sign of being incredibly engaged with the exhibit. Of course, we have to wait and see if this trend will carry across the entire dataset. 

        Even though I had initially suspected that children would spend more time engaging with the materials in the direct learning environment, I was surprised to find that participants’ engagements were roughly the same between either museum. Since our research is not yet complete, we still have to wait before we can make any definite conclusions. However, in the future, we plan on analyzing the interactions between children and staff members; both on a cultural level and an interpersonal level. On a cultural level, we want to see how long staff members are talking to kids between museums, and if the difference between direct vs indirect instruction plays a part in differentiating those. However, we want to look closely at the kinds of conversations children are having with the staff members at each individual museum, and if the staff to child interactions are equal or unequal between participants. This means that we want to see if some children spend more time talking with staff members than others do at each museum. Depending on what we find, we could determine if the presence of a staff member will increase the child’s engagement and overall interest in the exhibit.

 

​

6. Importance 

​

          Though the project is not yet complete, we hope to share the conclusions we draw from our research with our partner museums and other informal education settings across the country. Though we are unsure of what those conclusions will be exactly, we hope that they will be consistent with what we have seen in the first 20 videos we sampled. We are already beginning to see that our two partner museums are doing a excellent job of keeping their participants engaged in their exhibits, and we hope that, by having access to the data we gather, they will be able to continue to improve the quality of their exhibits. We also want to look at informal education settings outside of museum exhibits, and see if the conclusions we draw could also apply to those as well. Overall, we want to further our understanding of how young children learn, and what environments they learn best in. The lab plans on diving further into researching what factors can influence children’s engagement in these kinds of settings, possibly targeting factors that weren’t targeted in this particular research project. 

           Overall, it is very important to study different forms of informal education; young children spend so much time learning from experience outside of the classroom, and we want to make sure that they are exposed to new and interesting ideas that may not be introduced in school. We hope that we can foster creativity and exploration in young children, and encourage them to pursue an interest in a topic they might’ve learned outside the classroom. 

 

​

7. Reflection 

​

    Working with the lab has been a wild experience, and I’m so grateful that I’ve had the chance to work on this project with Zoe and Diane. When I first joined the project, the lab was still trying to get approval from the MOXI for the project to even take place there. I was able to experience what it takes to get a research project off its feet, and now I understand how hard it can be to create a well-rounded research project that targets a certain question that we wanted to answer. Though the beginning of the process was slow, I was able to spend a lot of time working with Diane and Zoe on the coding manual, making sure that it was clear, concise, and inclusive to all of the parameters we wanted to code before we began gathering data. We ended up changing the manual often and the coding process was tedious and difficult at times, but by early March I was able to start coding four videos a week with the finalized coding manual. 

    Throughout this entire process, I have really developed an interest in child development and how certain levels of engagement can lead to interest and curiosity in particular subjects later in life. At the beginning of the year, I was a little bummed that I couldn’t find a mentor who focused on language acquisition, a topic I was incredibly interested in, but I was pleasantly surprised with how much I enjoyed learning about early education and learning environments. In a perfect world, I think we would’ve finalized the coding manual earlier, so I wasn’t spending 6 hours coding a video, but otherwise, the project went smoothly and I am proud of what I have accomplished. 

    Looking forward into the future, I do want to try and transition what I’ve learned from this project into a more language-based project. Though I have no idea what that will manifest into next year, I know that I would really like to study brain development in bilingual children and research what kind of effects learning another language can have. I hope that a piece of that project/research can include studying the kinds of learning environments that maximize language learning/proficiency. 

 

​

8. Acknowledgements

​

         I want to express my gratitude to Zoe Liberman and Diane Lee for giving me this amazing opportunity. Working with the two of them in the UCSB Social Cognitive Development Lab has been a wonderful experience, and I have learned so much from being a part of the Museum Learning research project. I also want to thank Ms. Richard for giving us the chance to dive into real-world research projects and for everything she has done to make the STEM program as amazing as it is. 

​

​

9. Works Cited

​

Bonawitz, Elizabeth & Shafto, Patrick & Gweon, Hyowon & D Goodman, Noah & Spelke, Elizabeth & 

        Schulz, Laura. (2011). The Double-edged Sword of Pedagogy: Instruction limits spontaneous 

        exploration and discovery. Cognition. 120. 322-30. 10.1016/j.cognition.2010.10.001. 

 

Briseño-Garzón Adriana, and David Anderson. “‘‘My Child Is Your Child’’: Family Behavior in a 

        Mexican Science Museum.” University of British Columbia, Curator: The Museum Journal, 

        2012, pp. 179–201.

 

Legare, C.H., Sobel, D.M. & Callanan, M. Psychon Bull Rev (2017) 24: 1548. 

         https://doi.org/10.3758/s13423-017-1351-3