PS 2: Environmental data + graphs

Credit: This activity was adapted from one designed by Dr. Stephanie Spielman for the course, Data Science for Biologists at Rowan University.

Learning objectives

• Gain an understanding of Pasa and the organization’s Nutrient Density Study
• Start getting familiar with the data for our project
• Practice identifying data types
• Practice indicating the appropriate graph to use for particular kind(s) of data and question
• Create a graph using ggplot2

Resources

• Refer to the notes you took in lab
• See lecture slides on Types of data and graphs (Moodle)

Background

This dataset, contains a subset of the data that Pasa Sustainable Agriculture shared with us from their Nutrient Density Study. Each row comes from a paired crop and soil sample. The columns provide information about either the farm that provided the sample, characteristics of the crop that was sampled, or characteristics of the soil that was sampled.

You will see some entries in the dataset that are NA, which implies a missing value. Sometimes, data is missing or unknown, and it doesn’t necessarily imply anything is wrong with the dataset. Instead, it implies that data collection can be challenging, and you can’t always get all the information you want.

Specific instructions

Ultimately you will submit your answers through the associated problem set on Moodle. Feel free to record your initial thoughts on a separate document if it helps you to organize your answers.

Set 1

For these questions, use your notes from lab to record your best answer based on the information that was shared with us by Sarah Isbell. Your responses will be helpful to refer back to later in the semester as we get deeper into the data analysis.

1. What do you understand to be the motivation(s) for the Nutrient Density Study? Why was the study initiated? (consider both Pasa as an organization and participating farmers; 2-4 sentences).

2. What do you understand to be the experimental design of the Nutrient Density Study? (2-4 sentences)
   • Is it an observational study or a manipulative experiment? (think back to ENST 162…)
   • What determined where and when samples were collected?

3. What do you understand to be the methodology of the Nutrient Density Study? (2-4 sentences)
   • What types of measurements were made and recorded on farm practices, crop characteristics, and soil characteristics?
   • Who made these measurements and how?

4. What do you understand to be the goals of our work this semester, from the perspective of our partner? (2-4 sentences)
   • What is Pasa hoping to get out of the collaboration?
   • What kinds of insights might be useful and interesting to Pasa’s farmer members?

Data related questions:

Explore the dataset

Use the “Show [] entries” dropdown menu below to see all 20 rows.

Scroll to the right to see all columns.

You can use the “Search” field to find values in the data.

Click on a column header to sort the data by that column.

Explanation of variables

• farmer_id: Unique identifier for farm/farmer contributing the samples
• state: State where samples were collected
• county: County where samples were collected
• survey_version: Relates to when the sample was collected and the survey in place at that time
• farm_practices.no_till: Does the sample come from a no-till field?
• crop_color: Color of the crop sample
• crop: Species of the crop sample
• variety: Specific variety (genetic line) of the crop sample
• antioxidants: Concentration of antioxidants in the crop sample
• polyphenols: Concentration of polyphenols in the crop sample
• produce_xrf_Ca: Concentration of calcium in the crop sample
• produce_xrf_K: Concentration of potassium in the crop sample
• om_percent: Percent soil organic matter in the soil sample
• ph: pH value for the soil sample

Set 2

Instructions: Answer these questions about the dataset variables and how one might visualize the data. When referring to variable names in the data, you must be absolutely precise with spelling and capitalization. Your answer will only be correct if it could be understood in R.

For example, there is no variable called Crop Color but there is a variable called crop_color.

1. Name one variable in this dataset that is categorical.

2. Name one variable in this dataset that is numeric.

3. Name one variable in this dataset that is logical.

4. Name one variable in this dataset that provides information about spatial location.

5. Name one variable in this dataset that provides information about time.

6. Name one variable for which do we NOT know all of the values. (in other words, there is missing data)

7. For which variable is there more than one value in each cell?

8. Examine the values in farmer_id, which indicates the identity of the farm/farmer that the sample came from. Why do you think the values are NOT ordinary names of farms or people?

9. Consider the guiding question for Pasa’s project: “Do healthier soils grow healthier food?” Given this question…

   • What is one variable in this dataset that could be considered a dependent variable?
   • What is one variable in this dataset that could be considered an independent variable?

Set 3

Instructions: For these questions, indicate if the presented information is a variable or a value in the dataset. Remember that a variable is an attribute (e.g. height), while a value is a particular instance of that attribute (e.g. 5’8”).

As an example, crop_color is a variable, and one of its values is “orange”. By contrast, 289.6 is a value that is associated with the variable produce_xrf_K.

Hint: Use the search bar above the table to help you find values!

1. variety

2. butternut_squash

3. 7340.1

4. om_percent

5. false

6. lancaster

7. 7.3

8. farmer_id

Set 4

Instructions: For these questions, suggest which type of graph(s) would be appropriate to accomplish the stated goal. Your answers should be based ONLY on whether the type of plot is technically appropriate for visualizing those data types (don’t worry about whether it might “look pretty” or not). Select ALL that would be appropriate.

Your graph options are:

• scatterplot
• bar plot
• histogram
• density plot
• boxplot
• heatmap

1. You want to visualize the distribution of om_percent, the soil organic matter content.

2. You want to visualize how many samples come from each county in the dataset.

3. You want to visualize the relationship between om_percent and the crop concentration of antioxidants.

4. You want to compare the distribution of om_percent between farms that till vs. those that don’t (as indicated by farm_practices.no_till).

5. You want to find out whether crop calcium concentration (produce_xrf_Ca) increases with ph.

6. You want to summarize the distribution of crop_color across crop.

Set 5

Instructions: Choose two of the examples in Set 4 and write the code that would be necessary to create the corresponding graph using ggplot2.

• One example should involve a single variable
• One example should involve two variables
• Assume the dataframe is called pasa.

See the ggplot2 cheatsheet here

Session information

R version 4.3.2 (2023-10-31)
Platform: x86_64-apple-darwin20 (64-bit)
Running under: macOS Monterey 12.4

Matrix products: default
BLAS:   /Library/Frameworks/R.framework/Versions/4.3-x86_64/Resources/lib/libRblas.0.dylib 
LAPACK: /Library/Frameworks/R.framework/Versions/4.3-x86_64/Resources/lib/libRlapack.dylib;  LAPACK version 3.11.0

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

time zone: America/New_York
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] dplyr_1.1.4     readr_2.1.5     DT_0.31         workflowr_1.7.1

loaded via a namespace (and not attached):
 [1] sass_0.4.8        utf8_1.2.4        generics_0.1.3    stringi_1.8.3    
 [5] hms_1.1.3         digest_0.6.34     magrittr_2.0.3    evaluate_0.23    
 [9] fastmap_1.1.1     rprojroot_2.0.4   jsonlite_1.8.8    processx_3.8.3   
[13] whisker_0.4.1     ps_1.7.5          promises_1.2.1    httr_1.4.7       
[17] fansi_1.0.6       crosstalk_1.2.1   jquerylib_0.1.4   cli_3.6.2        
[21] rlang_1.1.3       crayon_1.5.2      ellipsis_0.3.2    bit64_4.0.5      
[25] withr_3.0.0       cachem_1.0.8      yaml_2.3.8        tools_4.3.2      
[29] parallel_4.3.2    tzdb_0.4.0        httpuv_1.6.13     vctrs_0.6.5      
[33] R6_2.5.1          lifecycle_1.0.4   git2r_0.33.0      stringr_1.5.1    
[37] fs_1.6.3          htmlwidgets_1.6.4 bit_4.0.5         vroom_1.6.5      
[41] pkgconfig_2.0.3   callr_3.7.3       pillar_1.9.0      bslib_0.6.1      
[45] later_1.3.2       glue_1.7.0        Rcpp_1.0.12       xfun_0.41        
[49] tibble_3.2.1      tidyselect_1.2.0  rstudioapi_0.15.0 knitr_1.45       
[53] htmltools_0.5.7   rmarkdown_2.25    compiler_4.3.2    getPass_0.2-4