Life Lab School Gardens Program in PVUSD 2022-2023

In 2022-2023, Life Lab operated garden classrooms, each with a full-time garden teacher, in ten schools in the Pajaro Valley Unified School District. The classrooms serve a student population in which more than three in five students are multilingual learners and more than four in five are considered economically disadvantaged. In addition, 10% of students are migrant students and 14% do not have permanent housing. The implementation of the Life Lab School Gardens Program in every elementary school in the district is an ambitious goal that is well underway. This year, each garden classroom provided curriculum and learning opportunities aligned with the Next Generation Science Standards (NGSS). 

This report is the result of a study undertaken by researchers at the University of California Santa Cruz with the goal of learning more about the Life Lab School Gardens Program. We observed an exceptional program that provided opportunities for student engagement in science learning while fostering a sense of belonging. Our analysis suggests that participation in the garden classroom program has a positive and statistically significant impact on students' attitudes and beliefs about science and nature and on their science learning. Students who participated in the garden classrooms were more likely to believe that they could be scientists; there was a statistically significant increase in their agreement to the statement, “I can be a scientist.” There was also an increase in their actual science learning across three core areas: what scientists do (the NGSS Science and Engineering Practices), what scientists know (the NGSS Disciplinary Core Ideas), and how scientists think (the NGSS Crosscutting Concepts)

Methods

This evaluation was conducted by researchers at UC Santa Cruz. As a part of our evaluation we collected several types of data including pre- and post-surveys of students' attitudes and beliefs about nature and science (adapted from Cheng, & Monroe, 2012), a pre- and post-assessment of students' learning (aligned to NGSS standards), and six observations of garden classrooms by graduate students trained in ethnographic observation.  We attempted interviews with teachers and administrators, but the extreme weather events of the Winter of 2022-2023 created an emergency situation in the school district and made interviews impossible to schedule.  We were able to connect with one principal via email who responded to interview questions.

Attitudes & Beliefs

It is well-documented (e.g., Kuo, Barnes, & Jordan, 2019) that experiences in nature support students' content learning and provide intangible benefits such as self-discipline and collaborative skills. When students have positive attitudes and beliefs about nature and science, they are better positioned to access these benefits.

A pre- and post-survey of attitudes and beliefs was administered to 122 students at three schools (Ann Soldo, HA Hyde, and Hall District Elementary School) at the start of the school year and again at the conclusion of the fall semester. The survey contained five items to which students indicated their agreement (5-point Likert scale, 5-25 point range in total scores). Agreement with a statement indicates a positive disposition toward nature and science. The students were given the option to complete the survey in Spanish or English. 

Using the Wilcoxon signed-rank test, we saw a significant difference between the pre- and post-surveys (Z=-2.6, p < .05).  That is to say, when their scores on all five statements were added together and compared between pre- and post-survey, there was a statistically significant increase in the overall score. Based on this test, we can say that a student participation in the Life Lab School Gardens Program was associated with increased agreement with the statements in the surveys, and students generally held more positive attitudes and beliefs about science and nature after just a few months. That said, the mean scores on the pre-surveys (M=21.24) and post-surveys (M=21.86) were very similar. The students entered the school term with generally positive attitudes and beliefs, choosing "agree" or "strongly agree" for most survey items. Because of this, when we examined individual survey items, we were not able to make statements regarding the size of the change in attitudes and beliefs for most items. Indeed, as you will see below, only the final statement "I can be a scientist," yielded statistically significant change.

In this section, we review the survey results for each item in the pre- and post-survey. As a reminder, there were five matching statements in the surveys with five possible responses (1 = strongly disagree, 5 = strongly agree). For three statements, the mean score in the post-survey was higher than that in the pre-survey. These were "I like being in nature," "I think that humans are a part of nature," and "I can be a scientist." The other items showed a very small, non-significant decrease. One item showed statistically significant differences: “I can be a scientist” (t121 = -2.64, p < 0.05). 

"I can be a scientist."

An important goal of the Life Lab School Gardens Program is to provide students with experiences that are both educational and help to encourage a positive self-concept in science. Indeed, school-based science experiences and self-concept are strong predictors of students’ aspirations to pursue science in their future. This is especially vital for children from communities that are underrepresented in STEM fields - people of color, non-native English speakers, people from low socioeconomic backgrounds, women, etc (Vincent-Ruiz & Schunn, 2018). The belief that one can become a scientist is the essential first step. 

This item was the only one to show a statistically significant change between the pre- and post-surveys. The pre group had lower values (M = 3.92, SD = 1.53) than the post group (M = 4.34, SD = 1.28). We can be 95% confident that participation in the Life Lab School Gardens Program is associated with an increase in students' science self-concept (t121 = - 2.64, p < 0.05). By calculating a confidence interval, we can say that the size of this increase will be between 0.1 and 0.74 points on the five point scale. 

"Students are learning to build trellises as an extension of their engineering units. Students are watching life cycles before their eyes during worm exploration time. Parents with cooking and carpentry skills have been leading other parents in the continued enhancement of our learning space. "

These observations of a school principal highlight the varied opportunities students have to engage in doing science in the garden classroom. By including families and involving students in solving real problems in the garden, students are able to be legitimate participants in a community working together to build a learning space. These types of community involvement and project-based learning can support students in developing a positive self-concept in science.

The following items were not statistically significant in our analysis. As you can observe below, the pre- and post-survey mean scores were very similar for these items. This does not mean that there is nothing to learn about students' attitudes from these results, only that we cannot say that the garden classroom affected the change between pre- and post-survey. In general, students expressed positive attitudes to the following four statements.

"I like being in nature."

"I think that people are a part of nature."

"I like to care for plants."

"I like trying new fruits and vegetables."

Science Learning

The Life Lab School Gardens Program uses Next Generation Science lessons to create authentic, place-based learning experiences for students. During the period of this study, second-grade students were engaged in a unit on how seeds travel.

We administered a pre- and post-assessment to over 450 students in garden classrooms at all ten schools. The assessments were identical; students were given a seed (the type of seed differed between schools and between the pre- and post-assessment) and asked to complete two sentences: "I think this seed travels by ..." and "I think this because..." The assessments were administered in a naturalistic rather than experimental context. This means that we did not attempt to control for every variable. Some students may have received help writing or with translation. Some students may have talked with their classmates about their response. All students completed the assessments amidst the joyful commotion of a class of 2nd graders gathered in the garden. 

The assessments were randomly coded and scored without knowing the student, school, or whether it was a pre- or post-assessment. We scored each assessment on three dimensions informed by the Next Generation Science Standards: Disciplinary Core Ideas, Science and Engineering Practices, and Crosscutting Concepts. These are explained more fully below. Each assessment was given a score between 1-4 for each dimension (1 = emerging, 2= approaching grade-level, 3= proficient, 4 = advanced). 

After scoring, we eliminated the assessments of any students for whom we did not have both assessments, which included all students at Ohlone Elementary because we did not gain access to their post-assessments until after the analysis period. That left 339 students from seven elementary schools. There were between 40 and 69 students from each school. We then eliminated sixteen additional students whose writing was illegible to the scorer, and we analyzed the scores of the remaining 323 pre- and post-assessments. 

Using the Wilcoxon signed-rank test, we observed a statistically significant difference between the assessments, Z=-12.8, p<0.001.  Furthermore, the sample size was sufficiently large to enable a paired samples t-test which compares the means of two measurements taken from the same individuals at two different times. The purpose of this test is to determine whether there is statistical evidence that the mean difference between paired observations is significantly different from zero. Based on paired t-test results, we can say that there is a significant mean difference between pretest and posttest scores t(322) = -17.63, p = <.001. This means that in our sample, on average, posttest scores were 1.85 points higher than pretest scores. To put this number in context, because there were 3 dimensions on both the pre- and post-assessment, each scored 1-4, the highest possible score was 12. So, on average, the post-survey scores were 1.85 points higher overall. We can go on to say we are 95% confident that this increase will be between 1.64 and 2.06 for any similar population of students (95% CI [.1.64, 2.06]. In sum, we can say that participation in the Life Lab School Gardens Program is associated with an increase in students’ total scores on an assessment aligned to the Next Generation Science Standards.

In this section, we review the pre- and post-assessment results for each dimension on which they were scored and supplement these with vignettes from the ethnographic observations. For each dimension, we used the Wilcoxon signed-rank test and observed significantly higher scores on all three dimensions: Disciplinary Core Ideas (Z=-11.46, p < .001), Science and Engineering Practices (Z=-10.33, p < .001), and Crosscutting Concepts (Z=-8.76, p < .001). Furthermore, because the sample size was sufficiently large, we were able to perform paired t-tests that describe the size of the increase for each dimension. The figures below display the mean scores on the pre-test (green bars) and post-test (blue bars) for each.

Disciplinary Core Ideas

Pre-Test and Post-Test Mean Scores

Science & Engineering Practices

Pre-Test and Post-Test Mean Scores

Crosscutting Concepts

Pre-Test and Post-Test Mean Scores

Sticky seeds in the orchard - Semillas pegajosas en el huerto

Eager students gather around the garden teacher as she says in a mysterious voice, "I am practicing to be a magician. I'm going to turn you all into animals." As she hands pairs of students squares of furry fabric, each one really does seem to turn into a tiny mammal - crouching, growling, meowing, and hopping. The group makes its way to the "orchard" - a lovely hillside behind some classroom buildings dotted with fruit trees.

Pairs of "animals" scamper up and down the hill, laughing, calling out "Look at this!", and pressing their fabric squares into the dried grasses and rocky soil. Sure enough - seeds cling to the fabric, and the students race back to the teacher to show what they found. 

Back in the garden classroom, the students use child-sized magnifying glasses to examine the seeds embedded in their "fur". Up and down picnic tables protected from the sun by a canopy and surrounded by growing things, the children bend over their treasures, carefully drawing what they see. "It has spikes!" one calls out. "¡Espinas!" says another. 

On a large piece of paper in the center of the classroom space is written, "I think ____ allowed the seed to cling to the blanket." The children caption their drawings by completing the sentence. As they finish they wander around with their magnifying glasses, eager to explore more tiny things in the garden. 

Disciplinary Core Ideas: What Scientists Know

This assessment measured students understanding of the following Disciplinary Core Idea: LS2.A: Interdependent Relationships in Ecosystems. Each students' assessment was scored according to their demonstrated understanding of how their seed travelled. The scores were defined as follows 1 = no or little demonstrated understanding, 2 = a partial understanding, 3 = a clear understanding (e.g., "this seed travels by soaring on the wind"), and 4 = an advanced understanding that goes beyond grade-level expectations. It is important to note that a score of 3 is considered proficient. The pre-assessment scores were lower (M = 1.77, SD = 0.81) than the post-assessment scores (M = 2.54, SD = 0.68). A t-test for dependent samples showed that this difference was statistically significant, t(322) = -15.59, p = <.001. On average, our students' scores increased 0.77 points. We are 95% confident that any similar population will experience an increase on their DCI score between 0.68 and 0.87 points. 

The life of a flower - La vida de una flor

It is first thing in the morning, but the second-graders sit up straight on benches made of wooden planks and tree stumps. They seem excited to be in the garden classroom. The Garden Teacher explains that they are going to search all over the garden and find the most beautiful, healthy-looking flower they can. She then teaches them an important lesson in sustainability - "Only pick a flower if there are more than ten of them on the plant. That is the Rule of Ten." 

The students search the garden. Some rush in, grab the first flower they find that seems healthy and plentiful. Others take their time, deliberating between blooms. When all the students have a flower the teacher presents the next challenge. "Now find a flower that is older than your flower." As the students approach with their second flowers, she asks "How do you know this one is older?" "It is brown and flopped over," one replies. The students all chat amongst themselves, showing their flowers to their classmates and comparing whose healthy flower is most beautiful and whose older flower is most wizened. They repeat the exercise, this time looking for a flower that looks younger than their original flower. They soon return, small hands clasping tiny buds.

After the flower hunt, the students return to the benches and take careful notes as the Garden Teacher tells the story of a flower becoming a fruit. She uses a tomatillo as her example, and the students diligently draw the stages from bud to blossom to fruit. 

The most exciting part of the lesson happens at the very end when the teacher brings out a basket of snap peas to eat. The children swing their legs and chat happily as they munch on this final stage in the life cycle of a flower. 

Science and Engineering Practices: What Scientists Do

This assessment measured students' understanding of the following Science and Engineering Practice: 2-PS1-3: Make observations to construct an evidence-based account for natural phenomena. Each students' assessment was scored according to the presence and quality of evidence-backed reasoning. The scores were defined as follows 1 = no or little demonstrated evidence, 2 = partially explained or vague evidence, 3 = correct evidence (e.g., "this seed travels by soaring on the wind because when I throw it, it soars"), and 4 = an advanced explanation that goes beyond grade-level expectations. It is important to note that a score of 3 is considered proficient. The pre-assessment scores were lower (M = 1.25, SD = 0.52) than the post-assessment scores (M = 1.91, SD = 0.85). A t-test for dependent samples showed that this difference was statistically significant,  t(322) = -13.21, p = <.001. On average, our students' scores increased 0.66 points. We are 95% confident that any similar population will experience an increase on their SEP score between 0.56 and 0.75 points. 

Structure and function - Estructura y función

It is a cool, foggy morning and the school grounds seem mysterious as I make my way to the garden classroom. I find it tucked away in a quiet corner under trees. It is lovely and very much makes me wish I were back in second grade. The second graders seem excited to be there too, though their excitement expresses itself in goofy movements and the occasional student bursting into a popular TikTok song, "It's corn!" The Garden Teacher handles their energy with care and empathy and before we all know it, they are seated calmly on benches before a blackboard that contains just four words: Structure. Estructura. Function. Función

Structure and function seems like a difficult concept for 7 year-olds, but the teacher starts by connecting to students' at-home experiences. "Who has built with legos?" she asks them. They clamor to share stories of their creations. The students speak in both Spanish and English, and the teacher moves seamlessly between the two languages, often rebroadcasting a student's words in whichever language they didn't speak. When one student shares about building an airplane, the teacher jumps on the opportunity to make the all-important connections. 

"Why does the airplane fly?" 

"Because it has wings!"

"Yes. The wings are the structure. Flying is the function."

Later, they will examine seeds with "helicopter wings" and build models that spin and fly through the air. A concept that seemed quite challenging at first is made clear with hands-on activities and opportunities to connect to lived experiences. 

Crosscutting Concepts: How Scientists Think

This assessment measured students understanding of the following Crosscutting Concept: 2-LS2-2: The shape and stability of structures of natural objects are related to their function(s).  Each students' assessment was scored according to their connections between structure and function. The scores were defined as follows 1 = no or little connection to seed structures, 2 = partially explained or vague connections, 3 = clear connection to seed structures (e.g., "this seed travels by soaring on the wind because it is shaped like a wing"), and 4 = an advanced explanation that goes beyond grade-level expectations. It is important to note that a score of 3 is considered proficient. The pre-assessment scores were lower (M = 1.11, SD = 0.36) than the post-assessment scores (M = 1.54, SD = 0.73). A t-test for dependent samples showed that this difference was statistically significant,  t(322) = -10.39, p = <.001. On average, our students' scores increased 0.43 points. We are 95% confident that any similar population will experience an increase on their CC score between 0.34 and 0.50 points. 

Conclusion

We saw evidence that the Life Lab School Gardens Program in the Pajaro Valley Unified School District positively impacted second-grade students' attitudes and beliefs about science and nature and their science learning. Each garden classroom we visited was a truly beautiful space that was clearly loved and cared for by the school community. In every observation, we saw excited groups of students learn in an environment that allowed them to shout, move their bodies, and explore the world. The garden teachers were calm in the midst of the sometimes chaotic energy of their students, and used a variety of positive discipline strategies to encourage, redirect, and teach.

Analysis of the pre- and post-surveys showed a statistically significant overall increase in students’ positive attitudes and beliefs about science and nature. The most significant gain was in response to the statement "I can be a scientist." Though there were no statistically significant gains for the other survey items, the mean for each item was generally high (meaning agree or strongly agree) in both surveys. There is strong evidence that participation in the Life Lab School Gardens Program supported students’ self-concept around science and their beliefs that they can be scientists.

Analysis of the NGSS-aligned pre- and post-assessments also showed a statistically significant overall increase in students' scores. On all three dimensions that were scored, Disciplinary Core Ideas, Science and Engineering Practices, and Crosscutting Concepts, average scores were higher on the post-assessments. Participating in the Life Lab School Gardens Program  has real, measurable affects on students learning about what scientists know, what scientists do, and how scientists think. Though the mean scores were low overall, it should be noted that these assessments were given in the fall term when students were still developing the writing skills to demonstrate their understanding. Based on our ethnographic observations, we suspect that we would have seen higher scores if the students were interviewed about their responses. 

Finally, observations and conversations with school employees provide evidence that the Life Lab School Gardens Program is providing a space for community-building with students and their families. This is an important element of the garden classroom model which creates opportunities for students to make connections between home and school learning.