Large enrollment courses present a challenge for instructors who want to engage students in authentic science practices that fit the recommendations of Vision & Change. Our lesson provides a meaningful science experience for undergraduates in the context of a large introductory biology course by guiding them to measure seasonal changes of plants on campus. Plant phenology is the study of periodic life cycle events in plants such as bud break, flowering, and leaf senescence. This lesson enables collaborative teams to collect, quantify, and analyze observable seasonal changes in nature. Students collect phenology data on an open-source digital database accessible from their mobile devices. Data are then imported into a simple analytical application hosted by the Quantitative Undergraduate Biology Education and Synthesis Hub (QUBES). Students use the collected data to develop a proposal with questions, hypotheses, and data visualizations. This lesson engages students in authentic inquiry about local and readily observable natural history patterns related to inter- and intraspecific variation in plants and promotes development of observational, quantitative, and communication skills. Our lesson design is highly flexible and suitable across different course levels, course sizes, and college campuses.
The Learning Goals of this lesson span across the entire semester.
Students will learn about plant phenology and how to quantify changes in local tree species.
Students will explore evolution, natural selection and adaptation through the lens of plant phenology.
Students will learn about plant-environment interactions and the role of plants in the ecosystem.
Students will work collectively as a group through multiple steps of the scientific process leading to the development of a research proposal.
Lesson Learning Objectives
The Learning Objectives of this lesson span across the entire semester.
Observe and collect information on phenological changes in local trees.
Become familiar with a database and how to work with large datasets.
Analyze and visualize data from the database to test their hypotheses and questions.
Develop a research proposal including empirically-driven questions and hypotheses.
Synthesize the results of their analysis in the context of plant biodiversity and local environmental conditions.
Bloom's Cognitive Level
Vision and Change Core Competencies
Vision and Change Core Concepts
Principles of How People Learn
Decades of research advocate that Science, Technology, Engineering, and Math (STEM) teaching could improve performance and retention in these fields by better engaging students in authentic science practices (1,2,3,4). More recently, increased attention has focused on engaging students in authentic research as an approach to biology education, such as course-based research experiences (5,6). However, while course-based research experiences have achieved many curricular goals, there are barriers to scaling up to large-enrollment courses that are currently offered at many universities for introductory biology. The challenge is to overcome perceived barriers to engage more than 100 students in authentic research practices in an auditorium-style classroom. The purpose of this lesson is to leverage mobile technology, campus natural resources, and databases to engage student learning of plant phenology while also developing their research skills.
Phenology is the study of seasonal changes in species. In plants, phenological changes include bud formation, flowering events, and leaf senescence. These changes occur everywhere and present an opportunity to introduce students to local ecological phenomena regardless of geographic location. By taking a quantitative and natural history approach to plant phenology, we can engage students in multiple scientific practices in our courses. This lesson was originally developed by Dr. Tammy Long and Dr. Sara Wyse (7) and has been modified to incorporate more scientific practices and take advantage of newer technology. The Beal Botanical Garden of Michigan State University collaborated with the authors to select and tag campus trees used for the lesson based on accessibility, species diversity, and known phenological differences among species. Garden staff also provided detailed maps and historical information about trees selected for study. With this information, campus resources and infrastructure, the authors were able to build a comprehensive course-long project that develops students' scientific and quantitative skills.
This lesson differs from previous phenology quantification efforts in its accessibility to students and emphasis on scientific practices. Citizen science initiatives such as Project Budburst (8) and the National Phenology Network (9) have simple observation protocols and data entry portals. These are effective for long-term monitoring and data management. However, the means of visualizing phenology data is constrained to past, curated data sets. Additionally, the lessons developed for these initiatives have mostly focused on K-12 students or adult education. This comprehensive phenology lesson enables college-level students to quickly collect their own data, develop hypotheses, visualize and analyze their data, and draw conclusions about tree phenology patterns on campus. This immersion in scientific practices results is a level of inquiry not found in other phenology-based lessons.
The students targeted in this lesson were enrolled in the second of two introductory biology courses for majors. The lesson was developed for large-enrollment course sections of about 200 students that met twice weekly for 80-minute class sessions.
REQUIRED LEARNING TIME
This lesson is a course-long project that is revisited throughout the semester.
PRE-REQUISITE STUDENT KNOWLEDGE
Students will need to know how to use a web browser on their smart phone. Additionally, students may benefit from knowledge of how to navigate a campus map. If trees or plants are not currently marked on campus, students will have to develop the ability to recognize trees and mark their individual trees. We provided digital photographs of each individual tree to help students identify their specific tree.
PRE-REQUISITE TEACHER KNOWLEDGE
The instructor should have basic computer skills in order to work with MS Excel. Additional computer skills to navigate the QUBES Hub website and process student data are valuable as well. Access to the Internet is required. Teachers should be able to identify tree species on campus and be able to generate or annotate a campus map with tree locations and identifiable markers. While the lesson appears complex with several moving parts, the two online systems used in this system, Ona.io and QUBES Hub, are designed for ease of use with extensive help available to novice users, especially instructors.
SCIENTIFIC TEACHING THEMES
Throughout the lesson, students are interacting directly with course material with some guidance from instructors and teaching assistants (TAs).
Peer learning: This lesson is based on teams of students working together towards the learning objectives. Almost every step involves collaboration and discussion among students on how best to accomplish their goals
Inquiry-based learning: This lesson is a research exploration of changes in plant phenology in the students' local environment. They develop questions and hypotheses based on their observations and test those hypotheses by visualizing course-collected data.
The lesson consists of a multi-tiered assessment strategy with small and large-scale assessments spaced throughout the course. Formative assessments served to build student confidence and skills and to ensure participation throughout. Examples of formative assessments included:  weekly data collection - students received a minimal amount of points for completing weekly data uploads and instructors alerted students who fell behind,  phenology calibration - to improve data quality and consistency across teams, students used clickers to practice rating phenological change from projected images of campus trees,  homework - students received feedback on several homework assignments related to data collection, visualization, plant phenology, and developing their research proposal,  in-class activities - additional feedback was provided in class where students visualized data with guidance from instructors and TAs, and  proposal draft - toward the end of the course, student groups wrote a research proposal draft and were given feedback. A final research proposal served as a summative assessment of how students accomplished the learning objectives.
The lesson is a collaborative learning activity with all students collecting data for their group and consequently the whole course. Every student has the opportunity to collect, analyze, and visualize phenology data. Each group separately develops an original research question and hypothesis and has the opportunity to share their ideas with other groups. Providing participation opportunities for all students within a group is very important. As part of the course, each student in a group (4 per group) is designated a letter (A-D) and throughout the course, the student representative per group rotates each class session. Additionally, there is a team peer-review questionnaire administered to the students mid-semester to help them address any long-standing disagreements or controversies. Any students with significant discrepancies in self vs. peer scores are noted for further discussion with the instructor.
Prior to the start of the course, instructor(s) select and tag trees that will be used in the lesson. While the study could be modified to allow students to select and identify trees, we have found that this can result in data that are less reliable due to inaccurate identifications and/or less usable because the nature of selection may be more random and less conducive to hypothesis testing with replicate sampling. Trees can be selected based on multiple criteria. We advise selecting trees based on student accessibility, phylogenetic diversity, and known variation in phenological patterns (i.e., deciduous vs. evergreen, early vs. late season). For our course of 200 students (50 groups of 4 students), we selected 10 different species and 15 replicate trees per species. Once trees have been selected and tagged with a unique identifier, the instructor will need to build maps for students to find their trees, or alternatively, provide GPS coordinates for students to navigate to the tree using their mobile devices.
BUILDING THE DATA COLLECTION FORM
Students collect data on their mobile devices using a predesigned data collection form built using Ona (www.ona.io). The instructor will need to create a free account and start a new 'public' project. Public projects allow for unlimited data submissions and open access for anyone with the website link to the project page. This allows students to independently verify their data submissions and confirm their collected data was logged.
From the Ona.io website, instructors can use the form builder, or alternatively, build a form using XLSform syntax (http://xlsform.org/). An example of the form used by the authors can be found in the supplemental materials (Supporting File S1 - Data Collection XLS Form Example). One of the many advantages of this system is that students can access the data collection form using a website URL. Once they have loaded the webpage on their mobile device, they can go offline to log data (including GPS points and photos) and upload it once they reacquire Wi-Fi access. In this way, students don't need to use their data plan or rely on Wi-Fi while collecting data outside. The authors have tested this system with two sections of 200 students each and all web browsers and phone types were found to be compatible.
The authors used CATME Team Maker (http://info.catme.org/) to build student groups. To balance the groups, the authors used gender, work schedule, motivation, and home address as parameters for forming groups. While CATME now charges for their services, the instructor may form groups using any system they choose.
WEEK 1: INTRODUCTION AND DATA COLLECTION PROTOCOL
Early in the course, students are introduced to the topic of plant phenology and the importance of measuring plants over time (Supporting File S2 - Week 1 Slides).
DATA COLLECTION IN-CLASS TEST
After introducing plant phenology, students used their mobile devices in class to access the webform and collect "data" on each other as if they were trees. This provided a chance for students to interact with the interface and for instructors to walk through the data collection procedure and troubleshoot issues that arose prior to sending students in the field. In this same class session, the instructor projected the online database to show students the tabulated data and to model how they would access and view data from the class (Supporting File S3 - In-Class Data Collection Protocol). A data collection protocol developed for students can be found in supplemental material (Supporting File S4 - Data Collection Protocol).
TREE SELECTION IN-CLASS
Using a Google Spreadsheet, student teams signed up for trees on a 'first-come, first-served' basis. Each team selected three (of the ten available) tree species and sampled two individuals each for a total of six trees per team. In our design, we prioritized replicate sampling by different student groups to promote data integrity. Therefore, each individual tree in the study was selected by two groups in each of the two sections of the course, ensuring each tree was measured weekly by four different student groups.
ONGOING DURING DATA COLLECTION: MAINTAINING PHENOLOGY DATA INTEGRITY
To account for user error, each individual tree was selected by four different groups (two groups in each section of the course). Ideally, this produced 4 data points per tree, per week. The instructor can then take the mean measurements and provide a cleaner dataset for students to visualize and analyze. It also provides a teaching opportunity to discuss the importance of replicate sampling and data integrity and management.
WEEKLY SCHEDULE OF DATA COLLECTION
Students were instructed to collect phenology data weekly for their 6 trees. The data form consisted of group number (text entry), student name (text entry), GPS coordinates (using phone's GPS), tree species (dropdown menu), tree unique ID (conditional on species selected, dropdown menu), percentage of fall color (0-100%), percentage of leaf fall (0-100%), and a photo of the tree (Uploaded by student) (See Supporting File S4 - Data Collection Protocol). Each Friday, the instructor downloaded the most recent database file and ran it through an R script (See Supporting File S5 - R script to process phenology data) to determine the number of trees logged per group (this analysis can be done in MS Excel as well). The instructor then emailed the report to all students as a way of notifying students of missing data before the weekly deadline of Sunday midnight. Every Monday, the instructor re-ran the R script and posted participation scores on the course management system. Although students have access to the database at all times, as the table of data collected is visible on Ona.io (Figure 1), the weekly report on Fridays served as an additional reminder for students to hold one another accountable for their contribution to data collection.
Estimates of both percentage color change and leaf fall tend to be highly variable among groups and among students within a group. This may be due to differences in visual acuity among students, and changes in student abilities to estimate the percentage of fall color and leaf senescence. To standardize across the entire class, in-class training sessions were implemented in the first week of class and halfway through data collection. Trees with different percentages of color and leaf fall were projected and students estimated percentages of fall color and leaf fall using iclickers. Calibration was attained by displaying values agreed upon by the instructors and comparing student responses to the instructor-consensus value. This in-class calibration was conducted in week 2 and week 6.
WEEK 2: PHENOLOGY DATA EXPLORATION HOMEWORK - BECOMING FAMILIAR WITH VISUALIZING AND ANALYZING DATA
To promote familiarity with data visualization and interpretation, and to prepare students for working with the much larger phenological data set later, students were given a homework assignment that asked them to visualize and answer questions about a smaller, less complex dataset (Supporting File S6 - Week 2 Phenology Data Exploration Homework). The homework dataset was generated by students completing an anonymous survey before the course began that collected data on height, sex, and diet. We included language to make clear the reasoning for the data collection, and that students could opt out entirely. The purpose of requesting parental and student trait data was to visualize the relative influence of parental genetics on student traits and what percent of variation is possibly explained by genetics (parental height) vs. the environment (frequency of protein or fruit consumption). The homework also modeled how data collected from multiple users could be aggregated and guided students to explore the data on their own, asking questions on how genes and the environment affect phenotypic traits. Students were directed to the QUBES Hub (https://qubeshub.org/) where they created a free account and used the Serenity web app to practice visualizing data (https://qubeshub.org/resources/serenity). This web app is a simple user interface for students to upload a CSV data file and quickly graph the data in multiple ways. A user guide was written by the instructors to help students through the Serenity interface (Supporting FileS7 - Serenity data visualization protocol). Prior to distribution of this homework and the student trait data, we suggest that instructors acknowledge the anonymity of the data and possibly discuss the difference between biological sex and gender to allay student concerns and anxieties about identity issues.
WEEK 4: PHENOLOGY DATA EXPLORATION EXERCISE I
Students were reminded of the phenology learning objectives and conducted a group exercise in class to deepen their understanding of plant phenology and lead them to ask questions and propose hypotheses about tree phenology (Supporting File S8 - Week 4 Slides). Students spent time researching the tree species online and answered several questions that were meant to lead them toward questions about the changes in fall phenology across different species.
As they developed their research questions and posed hypotheses, students learned about different types of data (categorical vs. continuous) as well as the value of a null hypothesis. Lastly, each group practiced visualizing the data using Serenity in an attempt to address their hypothesis (Supporting FileS7 - Serenity data visualization protocol).
WEEK 6: IN-CLASS CALIBRATION
Students were re-calibrated for phenology measurements. This was an opportunity to address on-going issues with data collection and data quality. Re-emphasizing the goals of the phenology project helped with motivating students to continue collecting data and work as a team.
WEEK 6: PHENOLOGY DATA EXPLORATION HOMEWORK - EXPLORING THE PHENOLOGY DATASET
To emphasize scientific practices and the phenology project, students were given a homework assignment to learn more about the tree species they were measuring, develop a question of their own design, and test their hypothesis by visualizing phenology data (Supporting File S9 - Week 6 Phenology Data Exploration Homework). Students used the Serenity web app to practice visualizing data (https://qubeshub.org/resources/serenity). This homework assessed students' ability to develop a research question, create a hypothesis, and visualize the phenology data.
WEEK 8: PHENOLOGY DATA EXPLORATION EXERCISE II
Students were reminded of the phenology learning objectives and conducted a group exercise in class to narrow their research question and build a scientific model of their hypothesis (Supporting File S10 - Week 8 Slides). The instructors collected weather data for campus including temperature, day length, and precipitation. These data can be accessed on Weather Underground (https://www.wunderground.com/history/). These climatic variables were introduced to the students to help them develop coherent scientific questions and hypotheses. Then the groups were asked to build box-and-arrow models (10) and present them to their neighboring group. This exercise promotes peer-learning, developing questions and hypotheses, practicing scientific inquiry, and group communication skills.
WEEK 9: PHENOLOGY RESEARCH PROPOSAL ASSIGNMENT - ROUGH DRAFT
Student groups convened outside of class to develop a research proposal based on the feedback they received on their question/hypothesis that they developed in class (Supporting File S11 - Research Proposal Assignment Description). Instructors evaluated the proposal drafts using a standard rubric accessible to students (Supporting File S12 - Research Proposal Draft Rubric). Many students inquired about using statistical analysis, so the instructors directed them to use the Radiant web app on QUBES Hub (Supporting File S13 - Radiant data visualization protocol), which is Serenity with analytical capabilities. For example, Radiant allows for analysis with linear regression and ANOVA.
WEEK 11: END OF DATA COLLECTION
Student groups complete their data collection and work on their research proposals.
PHENOLOGY RESEARCH PROPOSAL ASSIGNMENT - FINAL DRAFT
This summative assessment incorporates all of the skills and scientific practices that the students developed over the semester (Supporting File S14 - Research Proposal Final Rubric).
This lesson was implemented in the Fall of 2017 and 2018 and while there were several issues with data collection at first, the overall workflow went smoothly. Students were able to collect over 4,000 plant phenology observations in approximately 10 weeks. The strength of this data set lies in the number of data points and replicated measurements for each tree within each week of data collection. Overall, student feedback was positive. They enjoyed collecting and analyzing their own data, and by pooling data across all students, they were able to work with a fairly large data set to answer questions of their own design. By the end of the course, students were more comfortable working with, visualizing, and interpreting data in the context of phenology.
CONFIRMATION NOTICE OF DATA SUBMITTED
Students became confused when the technology appeared to fail and data did not upload. It needs to be clear that the "Queue" on the Data Collection Webform needs to be emptied for the data to have been submitted. Additionally, students need to check the online data table and search for their data to confirm the submission completed. This promotes independence and self-reliance in data collection, but can prove frustrating as students are learning the procedures.
The Ona.io webform is locked once data has been collected. If there is a mistake in the data collection form, the instructor will have to create a new one and have students access it through a new link. We discovered this when an attempt to fix an error in the data collection form temporarily resulted in several versions of the form to exist at once, creating confusion among students.
COURSE SIZE AND/OR NUMBER OF TREES
This lesson can be applied to diverse contexts and may include large or small numbers of trees or herbaceous plants. We found that 6 individual trees per group of 4 students was a reasonable number and students were able to consistently collect weekly data on their trees. With a smaller course, it may not be necessary to collect data using Ona.io. The Ona system allowed us to correct errors in data collection and manage over 200 students inputting data throughout the semester.
POTENTIAL ADDITIONS TO THE LESSON
CONFIRMATION NOTICE OF DATA SUBMITTED
It is possible to write code or manipulate the Ona.io data collection form to send a notification of submission. This would have addressed several of the issues we encountered with students.
DATA INTEGRITY LESSON
Using the calibration and data integrity cleaning to teach about the importance of data management and accuracy was valuable and reduced errors in student data collection. The majority of variance in student responses occurred when students attempted to discern the "percent color" or "percent leaf fall" around the 50% mark.
SELECTING TREES FOR DIFFERENT OBJECTIVES
Different courses might be interested in different types of questions that could be addressed by quantifying phenology. Selection of tree or other plant species may change depending on these questions or on the species locally available during a given season. We opted for phylogenetic diversity and differences in deciduousness when selecting campus trees.
Students learned about phylogenies and cladograms in the course. The tree species in the phenology lesson were used to create a phylogeny and explain patterns of relationships among tree species on campus. Fall color and leaf senescence data were then used by students to test the genetically-based cladogram and determine whether closely related species are more similar in phenology then distantly related species.
To promote the sharing of ideas and to practice oral presentation skills, students can opt-in for a mini-conference whereby they present their question, hypotheses and results to the rest of the course and possibly across sections of the course. The authors considered this idea and look forward to implementing it in future iterations of the lesson. This is also an opportunity to have graduate students and faculty visit and observe undergraduate research on campus phenology.
We thank Dr. Frank Telewski, Jeff Wilson, and the MSU Beal Botanical Garden staff. This study was significantly improved as a result of their assistance in identifying and tagging trees that were used in this study.
S1. Tree Phenology - Data Collection XLS Form Example: An example data collection form built in MS Excel to be imported into Ona.io
S2. Tree Phenology - Week 1 Slides: Introductory instructional slides on plant phenology for class instruction
S3. Tree Phenology - In-Class Data Collection Protocol: Data Collection Protocol for students to use their mobile devices in class to test out data collection and familiarize themselves with the interface. It includes screen shots of the data collection form used for this lesson. This file is meant to be a PDF file distributed to students. It is in PowerPoint form for editing purposes.
S4. Tree Phenology - Data Collection Protocol: Data Collection Protocol for students to use on their mobile devices in the field. It includes screen shots of the data collection form used for this lesson.
S5. Tree Phenology - R script to Process Phenology Data: R code for reading in CSV exported from Ona.io, cleaning and processing data, and reporting out student group counts of observations, aggregated data to be used for analysis, and visualizing variance in student observations.
S6. Tree Phenology - Week 2 Phenology Data Exploration Homework : Homework on Phenology Data Exploration for individual students. This assignment uses a data set provided to the students beforehand and allows them to practice developing scientific questions, hypotheses and visualizing data in Serenity.
S7. Tree Phenology - Serenity Data Visualization Protocol: Data Visualization Protocol for navigating Serenity on QUBES Hub. This web-app provides a simple means of visualizing data in a user-friendly interface.
S8. Tree Phenology - Week 4 Slides: Instructional slides on Phenology Data Exploration to reacquaint students with the data set and have them practice as a group developing questions, hypotheses and visualizing data. Additionally, they have an opportunity to learn more about the tree species.
S9. Tree Phenology - Week 6 Phenology Data Exploration Homework: Homework about Phenology Data Exploration for students to work as individuals on the phenology data set and explore how the data can be analyzed. Additionally, they have an opportunity to learn more about the tree species.
S10. Tree Phenology - Week 8 Slides: Instructional slides to Research Proposal Assignment to have them practice as a group developing questions, hypotheses and visualizing data for the phenology data set to develop a proposal.
S11. Tree Phenology - Research Proposal Assignment Description
S12. Tree Phenology - Research Proposal Draft Rubric: Rubric for scoring the Research Proposal Outlines
S13. Tree Phenology - Radiant Data Visualization Protocol: Data Visualization Protocol for navigating Radiant on QUBES Hub. This web-app provides a simple means of visualizing and analyzing data in a user-friendly interface.
S14. Tree Phenology - Research Proposal Final Rubric
NGSS Lead States. 2013. Next Generation Science Standards: For States, By States (Washington, D.C.: National Academies Press).
Brewer CA, Smith D. 2011.Vision and change in undergraduate biology education: a call to action. American Association for the Advancement of Science, Washington, DC.
National Research Council. 1996. National Science Education Standards. Washington, DC: The National Academies Press.
National Research Council. 2011. A framework for K-12 science education: Practices, crosscutting concepts, and core ideas. Washington, DC: National Academies Press.
Corwin LA, Graham MJ, Dolan EL. 2015. Modeling course-based undergraduate research experiences: an agenda for future research and evaluation. CBE-Life Sciences Education, 14:1.
Weaver GC, Russell CB, Wink DJ. 2008. Inquiry-based and research-based laboratory pedagogies in undergraduate science. Nature chemical biology, 4:10, 577.
Long T, Wyse S. 2012. A season for inquiry: investigating phenology in local campus trees. Science, 335:6071, 932-933.
Meymaris K, Henderson S, Alaback P, Havens K. 2008. Project budburst: Citizen science for all seasons. In AGU Fall Meeting Abstracts.
Betancourt JL, Schwartz MD, Breshears DD, Cayan DR, Dettinger MD, Inouye DW, Post E, Reed BC. 2005. Implementing a US national phenology network. Eos, Transactions American Geophysical Union 86, 51: 539-539.
Goel AK, G?mez de Silva GA, Gru? N, Murdock JW, Recker MM, Govindaraj T. 1996. Towards design learning environments -- I: Exploring how devices work. In: Frasson C., Gauthier G., Lesgold A. (eds) Intelligent Tutoring Systems. ITS 1996. Lecture Notes in Computer Science, vol 1086. Springer, Berlin, Heidelberg