Students will learn how statistical analyses can be applied to real data. MATLAB is utilized as a way to perform repeat analyses on multiple sets of data. This activity encourages students to bring together many skills they've learned throughout the course, ideally providing a cumulative final project.

Students will undergo higher-order thinking skills such as computation, and data analysis in order to complete this project. They will also be required to practice writing skills.

This project was developed as a second-year community college course in engineering for class sizes of ~35. It was designed as a final project as it brings together many skills throughout the course of the term (loops, if-statements, statistics, and advanced plotting). Students are typically given a week to complete the project, and could be assigned as an individual, partner, or small group exercise.

To be successful, students should be confident in array manipulation and extraction and writing loops and if-statements. Students can still be successful even if they need some help reading in data or creating plots.

This project could be easily adapted by requiring less statistical analyses (e.g. stopping after calculating the means) or even by requiring more analyses!

Required materials:

Access to MATLAB with the Statistics and Machine Learning Toolbox
Copy of referenced paper (or at least the abstract)
Student Handout for "Put Out the Fire"
Copy of the data file (.txt format)
Optional: Live Script Worksheet

Students can read through the handout (or Live Script Worksheet) and follow the steps to complete the activity. They should submit a copy of their script file, as well as written answers to the final three questions.

This activity uses MATLAB with the intent to demonstrate how even larger sets of data could be quickly analyzed with MATLAB. This activity has also been performed using Microsoft Excel, or other spreadsheet software with basic statistical capabilities.

Student Handout for Put Out the Fire (Microsoft Word 2007 (.docx) 60kB Nov1 20)

PFAAsOnMicrobes_Data_Weathers.txt (Text File 1kB Nov1 20)

PFAAsOnMicrobes_Worksheet.mlx (MATLAB Live Script 34kB Nov1 20)

Solutions for this activity have been included, though keep in mind there are many different ways to approach and program MATLAB problems.

PFAAsOnMicrobes_Solution_Weathers.mlx -- educator-only file

Hide

This assignment can be delivered in two ways depending on the independence or overall level of comprehension of the class. These methods are: 1.) For Advanced/Independent Students: Provide the instructions in the Student Handout and the data file and have the students follow the steps to independently create their own MATLAB file, or 2.) For Students Requiring Additional Guidance: Provide the MATLAB Live Script Worksheet that describes the format and suggested steps for completing the activity in addition to the Student Handout and data file. The instructor can add lines of code to the Live Script Worksheet before distribution in order to adjust the level of difficulty.

Students will need to be confident with while/for loops in order to complete this activity. Often students will be able to complete the individual commands for a single step, but have trouble conceptually applying these calculations across an array by using loops. This activity provides another opportunity to practice this skill, but the instructor should be aware that students may require personalized guidance to correctly apply looping techniques.

Depending on the level of statistics that is provided or required for this course (or the ability to access the Statistics and Machine Learning Toolbox), there is flexibility for the instructor to change the required statistical analyses. For example, the students can find the mean and standard deviation for each triplicate measurement, and plot the mean values in a stacked bar chart as described. This will skip the step of assessing the statistical significance of the results using the t-test. Note that ttest2 is the only command used in this activity that requires the use of the Statistics and Machine Learning Toolbox. Therefore, if the students are unable to access this toolbox, this portion of the exercise may be skipped without detriment to the larger objectives of this activity.

The follow-up questions can also be adapted, removed, or amended to match each course's background and objectives. The third question (why might some microorganisms be impacted more than others), for example, has no true correct answer, and is still an area of scientific development. Encourage the students to think outside the box, perhaps they can think of a new way to explain this phenomenon that has not yet been considered!

Students have met the goals of the computation section of the activity if they are able to print the names of the microorganisms that are significantly different after exposure to PFAAs, and if they are able to create a stacked bar graph.

Students are asked three follow-up questions that require higher-order thinking. The first question requires the student to look through the data they plotted to check that it matches the results described in the paper. A yes or no, with a brief sentence of what they observed in their data would be sufficient to show understanding.

The second question can be used depending on the context of the course and how much background on hypothesis testing and significance has been discussed.

The third question requires the students to dive deeper into either the referenced paper or their background in biology. Two- to three-sentences that demonstrate that the students have taken time to think about the puzzle would be a sufficient response. This question, in particular, could be removed if the students are unfamiliar with microbiology.

The fourth question would ideally give the students a chance to be creative and play around with the advanced plotting options in MATLAB. An excellent response would include a new type of plot that shows the relationships, in addition to one- or two-sentences about which plot is best. An average response would only include a brief one- or two-sentence answer about which plot is better.

Project references the data described in "Perfluoroalkyl acids inhibit reductive dechlorination of trichloroethene by repressing Dehalococcoides" found here: https://pubs.acs.org/doi/10.1021/acs.est.5b04854