CURE Report Week 9 (1)

docx

School

University of Louisville *

*We aren’t endorsed by this school

Course

243

Subject

Biology

Date

Apr 3, 2024

Type

docx

Pages

Uploaded by MajorMoon8437

Today, you will complete this worksheet as you analyze the data you gathered last week for the CURE report. We’ll be using the .csv file you created last week and the statistics website you used in Week 4 to: (1) calculate descriptive statistics for all your variables, (2) create histograms and plots to determine if your variables are normally distributed, (3) select an appropriate statistic test for each of your hypotheses, and (4) report those results. You’ll also produce graphs and captions of those results today, so that your GTA can provide feedback to you before you place these elements in your CURE report. Finally, you will draft a methods and results section of your CURE report for your GTA to provide feedback on this week. Unlike the introduction section in last week’s lab, providing a draft of both your methods and results section this week in lab is required rather than optional. INTRODUCTION Last week in lab, you formed two hypotheses using the CURE data we’ve collected so far in BIOL 241 and 243. Today, let’s review those hypotheses, and use what we’ve learned in previous labs and prelabs in BIOL 243 to classify our variables as (1) explanatory or response variables, (2) continuous or categorical, and (3) normally distributed or not normally distributed. To do this, we’ll need to refer back to the Shiny app statistics website you used in Week 4 of this lab to analyze an example dataset collected by one of the professors of the introductory course sequence, Dr. Abolins-Abols. A QUICK NOTE: Before you begin this lab, here’s quick note for those of you that extracted data from QGIS from either the imperviousness or canopy cover raster. If you did, take a look at your .csv of extracted data before lab. “No data” (or cells with missing values) in these 2 raster layers will have values larger than 100 in the .csv. Delete any values in that column that are larger than Photo by Bruno Pereira on Unsplash Week 9, BIOL 243 CURE Report, Part II Team members present: WEEK 9 LEARNING OUTCOMES By the end of this lab, you will:  Identify explanatory (independent) and response (dependent) variables  Identify continuous and categorical variables  Contrast descriptive statistics and inferential statistics  Choose which statistical test is most 1.(lead author) 2. 3. 4.

100 before you perform your statistical analyses. The Shiny app statistics website will ignore cells that are empty. Before we return to this website and perform our statistical analyses this week, be sure to also look over the feedback you received from your GTA on your hypotheses. MATERIALS  A computer  Soil Collection data .csv file (that contains your extracted raster data from last week!) PART 1: REVIEW YOUR HYPOTHESES & CLASSIFY VARIABLES Using your GTA’s feedback, rewrite your hypotheses below, if edits were suggested. If no edits were suggested, simply copy-paste your work from last week here. Hypothesis 1 (0.5 pts): Greator canopy cover maintains more ideal temperatures for the growth of microbes. Hypothesis 2 (0.5 pts): Greator canopy cover increases the opportunity for species specialization, which creates an ideal environment for microbespecies diversity. For hypothesis 1 above ( 2 pts )  what is your response variable? microbe abundance  is your response variable continuous or categorical? continuous  what is your explanatory variable? canopy cover  is your explanatory variable continuous or categorical? continuous For hypothesis 2 above ( 2 pts ):  what is your response variable? microbe species diversity  is your response variable continuous or categorical? continuous  what is your explanatory variable? canopy cover  is your explanatory variable continuous or categorical? continuous

PART 2: CALCULATE DESCRIPTIVE STATISTICS Click on this link to go to our Shiny app statistics website: https://abolins.shinyapps.io/biostats2/ We first need to upload your dataset. 1. Be sure you know where you saved the .csv file that you created at the end of lab last week. We need it, rather than the .csv file on Bb, because the .csv file on Bb doesn’t contain any of the QGIS data. 2. On the statistics website, click on “Choose CSV File” button and navigate to the location where you downloaded your dataset. 3. Your full dataset should now be uploaded to the webpage. You can toggle between seeing the beginning of the dataset and the full dataset by changing between “Head” and “All” options under “Display”. 4. You should keep the other dataset options at their default values (Header should be selected; Separator should be “Comma”; Quote should be “Double Quote”). Note that before you have uploaded the dataset and before variables are selected, the page may display the following error message: Error : An error has occurred. Check your logs or contact the app author for clarification. Ignore this error, it is displayed because the website cannot calculate anything because you haven’t chosen the data you want to plot/analyze. Recall that, once your dataset has been uploaded, it’s time to explore your data to determine what kind of statistical test you need to run for each hypothesis. Below the dataset you just imported into our Shiny app, you will see five tabs. To the left of each tab, you will see additional information about what you can do in each tab. Let’s use the Data summary tab first. Here, you will be able to do the following things:  Get summary statistics for each continuous variable (average, sample size, minimum and maximum values, as well as its variance (a statistical metric of how variable the data is).  Below the summary statistics, you will be able to inspect the histograms of each continuous variable. Remember from your pre-laboratory assignment that we use histograms to see if both the response and explanatory continuous variables are

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

normally distributed, to determine whether we should use Pearson or Spearman Rank correlation coefficients to analyze the relationship between two continuous variables.  At the bottom of the Data Summary tab ,you will be able to inspect the distribution of a continuous variable across two different groups from an explanatory categorical variable that only has two groups . Remember from your pre-laboratory assignment that we need to know if a continuous variable is normally distributed in both groups of a categorical variable to determine if we can use a T-test or a Mann-Whitney U test to analyze differences between two groups.  If you have more than two groups in your categorical variable, than you can’t use this tab of our Shiny app – and cannot do much in this section (Part 2) of the worksheet either. When prompted below, report that your variable has more than two categories, and move on to Part 3. Below, select the appropriate tables to report data on each of your hypotheses. Note that, since you only have two hypotheses, only two (at most) of the provided tables should contain information. We’ve provided enough tables for you to use regardless of your chosen hypothesis. If you need help understanding which table to complete – and which tab to use on our Shiny app on our website – raise your hand and ask for questions from your GTA. HYPOTHESIS 1 (2 pts): Do you have two continuous variables for hypothesis 1? If so, fill out this table. If not, skip it. Continuous variable 1 Continuous variable 2 Average 18.98413 169.5257 Sample size 63 63 Minimum value 0 39.75 Maximum value 91 400 Variance 664.661 6380.619 Is it normally distributed based on its histogram? no no Do you have one continuous variable and one categorical variable with 2 categories, only, for hypothesis 1? If so, fill out this table. If not, skip it. Name of continuous variable Name of categorical variable Sample size of the continuous variable

Minimum value of the continuous variable Max value of the continuous variable Variance of the continuous variable Is this variable normally distributed across both groups of the categorical variable? Based on these responses, what statistical test should you use to test Hypothesis 1? Spearman Correlation Test Did you not fill out a table above? Why not? (If you did fill out a table above for Hypothesis 1, you can skip this question) no, neither of our variables were categorical HYPOTHESIS 2 (2 pts): Do you have two continuous variables for hypothesis 2? If so, fill out this table. If not, skip it. Continuous variable 1 Continuous variable 2 Average 18.98413 57.42674 Sample size 63 63 Minimum value 0 0.5985 Maximum value 91 100 Variance 664.661 1700.794 Is it normally distributed based on its histogram? no no Do you have one continuous variable and one categorical variable with 2 categories, only, for hypothesis 2? If so, fill out this table. If not, skip it. Name of continuous variable Name of categorical variable Sample size of the continuous variable Minimum value of the continuous variable Max value of the continuous variable Variance of the continuous variable

Is this variable normally distributed across both groups of the categorical variable? Based on these responses, what statistical test should you use to test Hypothesis 2? spearman correlation Did you not fill out a table above? Why not? (If you did fill out a table above for Hypothesis 2, you can skip this question) neither variable was categorical PART 3: CALCULATE INFERENTIAL STATISTICS Recall from Week 4 that, under the Group tests tab, you will be able to calculate the statistical significance of differences in a continuous variable for 2 groups, only, using either a T-test of a Mann-Whitney U test, based on which test is appropriate for your data. Under the Correlation tests tab, you will be able to calculate the statistical significance of a correlation between 2 continuous variables using either Pearson or Spearman correlation coefficients, based on which test is appropriate for your data. Does one of your hypotheses have a categorical variable with more than 2 groups ? If so, use the Advanced group tests tab and follow the directions provided on this tab to (1) determine which test to use (either an ANOVA or a Kruskal-Wallis test) by graphing your residuals and (2) run the appropriate test and fill out the table below. Use your answers from Parts 1 and 2 above to decide what statistical test you need to run for each hypothesis. Hypothesis 1 (1 pt): Name of explanatory variable (repeat from Part 1): canopy cover Name of response variable (repeat from Part 1): microbe abundance Name of the statistical test: spearman correlation Observed correlation coefficient: 0.0221259 p-value: 0.2446297 Hypothesis 2 (1 pt): Name of explanatory variable (repeat from Part 1): canopy cover Name of response variable (repeat from Part 1): microbe diversity Name of the statistical test: spearman correlation Observed test statistic: 0.01143928 p-value: 0.404099 Does one of your hypotheses have a categorical variable with more than 2 groups ? If so, summarize your post-hoc test results here. If not, skip this step.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

PART 4: CREATE A GRAPH FOR EACH OF YOUR HYPOTHESES Recall from your prelab that there are two ways you could create your graphs from your CURE report: (1) you can use the Plots tab on our Shiny app statistics website (where you’ve been calculating your descriptive and inferential statistics in Parts 2 and 3), and then use a program like Microsoft Powerpoint to edit axes labels or (2) to you can make your graphs within Microsoft Excel, as we did last semester. Let’s recall the rubric elements that we’ll use to evaluate each of your graphs: Using this as a guide, create a graph for each of your hypotheses and upload them below. Hypothesis 1 graph (1 pt): Hypothesis 1 graph caption (0.5 pt; remember to number your figure – and that your QGIS graph is Figure 1)

Hypothesis 2 graph (1 pt): Hypothesis 2 graph caption (0.5 pt; remember to number your figure – and that your QGIS graph is Figure 1, while your Hypothesis 1 graph also received a number above) PART 5: DRAFTING THE TEXT OF YOUR METHODS AND RESULTS Now, let’s draft a methods and results section! Take a moment to first review the rubric elements for your methods section. Figure 2: Relationshio Between Percent Canopy Cover and Catalase Concentration, Which is Utalized to Measure Microbial Abundance. Figure 3: Relationship Between Percent Canopy Cover and Percent Functional Diversity to Measure Species Diversity.

Do you need to make any changes to your QGIS map, based on feedback from your GTA? If so, upload your new QGIS map here (0.5 pts) . If not, just upload your map from last week’s worksheet here. Do you need to make changes to your QGIS map caption from last week? If so, make those changes here. If not, copy-paste from the last worksheet ( 0.5 pts ). Figure 1: Locations of Soil Collection Points Throughout the City of Louisville, KY and Surrounding Counties

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

Now, compose a draft of the text for the methods section of your CURE report below. Methods (text only – not the map, 2 pts) Finally, use the elements of your rubric related to the text of the results section below to write a draft of the results section of your CURE report. Results (text only – not the graphs, 2 pts) Next week, you will put everything together and finish your CURE report. Make sure you have a solid draft of the Methods and Results sections before you leave lab today. Your GTA will check your work. GTA Sign-Off (1 pt) Throughout experimentation, microbial abundance was measured through catalase concentration (u/mL), as it is evident that increased catalase activity indicates increased microbial aubndance in soil samples. Similarly, percent functional diversity was utalized to measure the second response variable, microbial diversity. Both micribial abundance and diversity trends were tested against percent canopy cover data. Quantum Geographic Information System (QGIS) was used in order to visualize and compare data points from soil collections, as the application provides visual capabilities to map and export data files. With QGIS, previous collection data--which includes percent functional diversity and catalase concentration (u/mL)--could be compared with geographical data like percent canopy cover for statistical analysis. The application ShinyApps was used for said statistical analysis, through which two Spearman Correlation tests were preformed. This statistical test was chosen because in both hypothesis, both the response and explanatory variables were continuous and at least one variable was not normally distributed. Graphical descriptions of the data could be developed through ShinyApps. No clear trend is present, save for a large cluster of data points at low catalase concentration and low percent canopy cover (Figure 2). In the spearman correlation test run on this data, there was an observed correlation coefficient of 0.0221259, and a p value of 0.2446297, which indicates very little correlation between the two variables. In the second hypothesis, there is no clear trend present, save for a large cluster of data point at low percent functional diversity and low percent canopy cover (Figure 3). A spearman correlation test was also run for this data, where a correlation coefficient of 0.01143928 and p value of 0.01143928 indicates low correlation between the two variables. Fisher

CURE Report Week 9 (1)

Related Documents