Germs Everywhere Lab

Germs Everywhere Part 2 (Data Analysis) The time has come to gather data from your incubated plates, which will likely contain diverse colonies of bacteria. Each colony will contain millions of bacteria that are all descended from a single individual (often called a colony forming unit, or CFU) that was streaked across the plate last week. Pre-Lab(due prior to lab) 1. Anthony wants to make a batch of beer, and he is interested in determining whether the yeast will produce more carbonation (CO 2 gas) if he grows them in a higher concentration of sugar. He places yeast in a tube with either no sugar, 5% glucose, or 10% glucose. He measures the amount of CO 2 produced after 1 hour. Listed below are 4 different ways of presenting the same data. Write down 1 advantage and 1 disadvantage for each presentation style. Advantage: Advantage: Disadvantage: Disadvantage: Advantage: Advantage:

Germs Everywhere Part II Data Analysis Page 2 of 11 Disadvantage: Disadvantage: 2. Which graph makes it easiest to answer the follow research question: Which condition (5% or 10% glucose) produces more CO 2 than the control? Why did you choose this answer? 3. In the real world example of chronic migraine treatment comparing drug #1 to drug #2 in the “Analyzing your data” section page 4 of this handout, how does standard deviation show the difference between the results? Which drug is better in your opinion? Why? 4. Explain in your own words what you will be doing in lab today.

Germs Everywhere Part II Data Analysis Page 4 of 11 E. What is(are) the independent variable(s)? On which axis is the IV? - The independent variables is the time. It is on the x-axis. F. What is(are) the dependent variable(s)? On which axis is/are the DV? - The dependent variables is the length. It is on the y-axis. G. The legend actually describes key results. Is each shaded of the bars described at least once? Why isn’t every single bar described? - The legend shows results at least once. H. What do the “#” and “*” symbols represent? Why do we care about statistical significance - The numbers represent keys on the graph. We care about statistical signigicance because without it, things on the graph would not be correct. I. What are the limitations of this experiment? - The limitations of this experiment are when the graph runs out. II. Introduction to Calculating the Mean (Average) One of the most basic and effective ways of summarizing data is by calculating the mean value. This is an excellent way to provide a single value that best represents a larger number of cases. For your project, you will likely be comparing some measured variable (i.e. the number of round colonies, the density of colonies, etc.) across different groups (i.e. swab from unwashed hand vs. swab from hand washed with soap). The mean is calculated using the sum of all values divided by the total number of values: Accounting for Variation and Accounting for Sample Size When comparing groups of variables, the mean alone is not enough information to make conclusions. This is because mean values by themselves can be very misleading. Take the following two groups: # Colonies Group #1 # Colonies Group #2 6 10 7 10 4 1 10 0 6 12 Here is a real world example. Imagine you are in charge of testing two new drugs that were designed to cure chronic migraines in adults. When 20 adults with chronic migraine syndrome are treated with each drug, they rated the extent to which the drug reduced their migraine pain from zero to 100, with a score of zero meaning the drug did not reduce their pain and a score of 100 meaning that they no longer had any migraine pain. The following data was collected: % Pain Reduction from Drug #1 % Pain Reduction from Drug #2 100 70 100 50 100 70 100 60 100 60 Group 1: (6 + 7 + 4 + 10 + 6) / (5 values) = 6.6 colonies per plate Group 2: (10 + 10 + 1 + 0 + 12) / (5 values) = 6.6 colonies per plate The mean values for each group are the same (6.6), but the two groups have very different patterns. The values in Group #1 are generally much closer to the mean value than the values in Group #2. This is called the variance from the mean. When reporting the mean, the variance must always be reported. For both drugs, the mean score for pain reduction was 60. However, when you look at the raw data, you can see that Drug #1 completely removed the migraine pain in half of the individuals, but barely had an effect on the other half. The users of Drug #2 showed values that consistently were closer to the mean of 60. Some common ways to report the variance are standard deviation and standard error. These values represent how far the data values stray from the mean. For the example on the left, the calculated standard deviation for each Drug is: Drug #1 standard deviation = 41.04 Drug #2 standard deviation = 7.25 When we report the data we collected from the drug trials, we report the

Germs Everywhere Part II Data Analysis Page 5 of 11 100 50 100 60 100 70 100 60 100 50 20 60 20 50 20 60 20 60 20 70 20 60 20 60 20 50 20 60 20 70 III. COLLECTING YOUR DATA Do not open the ziplock bag Remember these are live bacteria cultures. We have no idea what type of bacteria has grown – it could be pathogenic. Care must be taken to avoid contact with the cultures. You can maneuver the plates to see what is in them through the bag, WITHOUT opening it. During all data collection, you should always be thinking about how you can avoid bias. Are you gathering your data in any way that could impact your results? If so, this could degrade the value of the science you are doing by impacting your reliability of your results. Potential sources of bias will differ depending on your study design. Prior to beginning data collection, discuss the ways you could bias your results and how to avoid them. Call your TA over to help if necessary. You might find that you need to be more specific in the dependent variable that you chose during your experimental design. Cell phone cameras should be used to capture images of the work you do (data collection procedures, colonies on plates, etc.) for use in your final report. Such images can be very helpful in effectively communicating your methods and results to others. Data collection can be tedious. Divide roles so that everyone can be involved and help, but be sure to keep the same roles the entire time (changing cell counters can be a source of error in your counts). Record your data on the sheet provided. It should be neat and legible to others. At least one dependent variable must be quantitative (numbers) Plate #: Group (Exp, Control, +Ctl, -Ctl) Dependent Variable #1 (what is being measured): _______________ Dependent Variable #2 (what is being measured): _______________ Dependent Variable #3 (what is being measured): _______________ 1 Bacteria How much bacteria is on a phone case How much bacteria is on a aggie one card How much bacteria is on your shoe 2 3

Germs Everywhere Part II Data Analysis Page 7 of 11 V. COMMUNICATING YOUR EXPERIMENT – Flowchart of Methods To effectively communicate experimental results, you must first be able to effectively convey the experimental design. When you write a formal laboratory report or a research paper, you must include step by step instructions so that someone else could repeat your experiment. When communicating orally, however, a brief visual is preferred. Create a flowchart of your experimental design using the following tools to help you: Are there more bacteria on an internal or external door handle? Can hand sanitizer effectively sanitize the laboratory benchtop? Control group: Benchtop without sanitizer Benchtop with sanitizer Exp Group 1: Internal Doors Exp Group 2: External Doors Squirt 5ml sanitizer on benchtop, wipe with paper towel for 30s

Germs Everywhere Part II Data Analysis Page 8 of 11 VI. COMMUNICATING YOUR EXPERIMENT - RESULTS Draw a graph of your data and discuss your analysis with your TA. You should make the simplest graph that still includes all of your data. For example, you must include bars for positive and negative controls individually (do not average these). However, you can graph the mean of your experimental groups and add error bars to represent the standard deviation. Include an overall title, axes titles and labels. If you prefer to draw your graph using excel, no problem. . Count # colonies on each plate IV: handle side Swab 20 streaks across entire handle Swab LB Agar Plates (1 plate per door handle) Incubate 1 week Room Temp DV: # colonies Constant or Controlled Variables Count % area of plate covered by colonies IV: sanitizer Swab three 4”x4” sections of benchtop for each group Swab LB Agar Plate Incubate 1 week Room Temp DV: % Area Constant or Controlled Variables 3 Internal Door Handles 3 External Door Handles

Germs Everywhere Part II Data Analysis Page 10 of 11 VII Write your conclusions to your experiment. Refer back to your hypothesis. Did you results support or refute your hypothesis? Remember, as scientists, some of the most important discoveries have come when our results do NOT support our hypothesis. So we don’t want to have bias in our interpretation. You will not be judged on the results of your experiment, even if you completely screwed up. You will be evaluated on how you explain your results and the depth of your interpretation of your results. Be sure to explain all parts of your data…not just the “easy-to-explain” parts. VIII Write the limitations to your experiment. Did your anticipated positive and negative control turn out as expected? If not, this impacts how you can interpret your data, and you need to explain that impact. Were there large variabilities in your replicates? If so, why might that be? Were there technical challenges to your data? How do these limitations impact your interpretation of the data?

Germs Everywhere Part II Data Analysis Page 11 of 11 IX. Extend your conclusions . Can your finding be generalized beyond this single experiment? Why or why not? For example, if you found more bacteria on the door handle than the table, is this likely to be true of all door handles versus all tables? If not, what conditions might apply? All lab door handles versus all lab tables? Or just Biology lab door handles versus Biology lab tables? What about just BIOL 101 lab door handles / tables? What makes sense based on your data? What does this data mean to you and your everyday life? Consider washing your hands more frequently? What about where you consider putting food and drink? Think deeply on these extensions. X. Write suggested improvements to your experiment . No experiment is ever perfect. There are always things that can be improved. Think about not only technical improvements (eg. Human error), but more importantly, experimental design. If you had to do your experiment over again with unlimited resources, what would you do differently?