Laboratory 1 The Scientific Method

Laboratory 1The Scientific MethodThe student dedicated to the idea that scientists in general (and perhaps biologists inparticular) can help make the world a better place should remember this admonishment:Question everything.Too often in their daily lives, people read information, assume it is correct, and take forgranted that those who report the information have gathered it in a logical, rigorous fashion.Unfortunately, this is oftentimes not the case. How can the well-informed person know whenreported information is flawed? One must consider how the data were collected, whether thesample size was large enough to be representative of reality, and whether the describedexperiments are repeatable and verifiable.Critical thinking is a double-edged sword. The scientist who (wisely) subjects his/hercolleagues' work to rigorous analysis must also be willing to submit to such scrutiny. In today’slab, you will perform an exercise that will help develop your critical thinking skills and ability todevise rigorous, repeatable experiments. At the end of the exercise, you will have theopportunity to critique the work of other teams in your lab, and to have your own experimentaldesign analyzed by your classmates.You are probably already familiar with the scientific method. Scientific investigators takea systematic approach to solving problems in biology by (1) acquiring knowledge about aparticular area of interest, (2) investigating unknown phenomena in this area, (3) testinghypotheses and predictions about unknown phenomena with rigorous experiments, and (4)integrating their findings with pre-existing knowledge.Although there is no single scientific method followed by all researchers, scientists dofollow a series of steps to determine which of two or more alternative hypotheses about aparticular observation is correct. Information is gained via experimentation, and carefulanalysis of results allows the scientist to draw logical conclusions.You MUST read Appendix I of this lab manual before coming to laboratory for thisexercise, or you will not be prepared to perform the lab.I. A Review of Basic ChemistryA. Polarity and SolubilityAn electric dipole is the separation of positive and negative electrical charges. A batteryis a common example of an electric dipole, but you are no doubt aware that many commonsubstances are composed of molecules forming electric dipoles. One of the most ubiquitous ofthese is water (H 2 O, Figure 1), said to be polar because of the charge differential at itsopposite poles.A molecule in which the electrons are distributed symmetrically, and do not form a dipole, issaid to be nonpolar. The polarity (or nonpolarity) of any substance is one of several importantaspects of its chemical nature, which determines how it will interact with other substances.A liquid that dissolves a substance is known as a solvent, and the substance that dissolvesin that liquid is known as the solute. Solute particles dissolve in a solvent to form a solution.The solubility (i.e., tendency to dissolve) of any substance depends on its chemicalproperties. A solvent must be able to separate the particles of the solute and occupy thespaces between the formerly bonded particles.1-1

Polar solvents (e.g., water) are able to separate the molecules of polar solutes (e.g.,sodium chloride, ammonia, etc.) from each other. As the positive pole of a polar solventmolecule approaches the negative end of a solute molecule, the attraction between them pullsthe solute away from its neighboring solute molecules and into solution. Polar solvents may beeither protic or aprotic. Protic solvents contain an easily removable hydrogen atom, mostoften bound to an oxygen atom (e.g., as a hydroxyl group, -OH). In short, protic solventmolecules can be hydrogen donors; aprotic solvent molecules cannot.Polar solvents can dissolve ionic solutes, that is, those held together by ionic bonds. Forexample, sodium chloride (NaCl, Our Friend Table Salt) dissolves in water because thenegative chloride ions are attracted to the positive end of the water molecules, and the positivesodium ion is attracted to the negative end of the water molecules. Pulled in two directions,the sodium chloride solid dissociates into Na + and Cl - , forming a solution of sodium andchloride ions in water.Nonpolar molecules have no affinity for polar molecules, and so exert no force to drawthem into solution. Nonpolar solvents, however, generally will dissolve nonpolar solutes.Figure 1. The Bohr model of the water molecule. In this diagram, theright side of the molecule is relatively more positively charged than the left,as the negatively charged electrons spend more time orbiting the oxygennucleus than the two hydrogen atoms at the positive pole. Water is thus apolar molecule, and can dissolve polar solids.B. Miscibility: Colloids and EmulsionsLiquids of different chemical nature can mix together, too, of course. The degree to whichthey form a uniform solution depends on their miscibility with each other. Miscibility is thedegree to which two liquids can combine to form a homogeneous, uniform solution. Forexample, ethanol and water are highly miscible, forming a uniform solution when mixed. Bycontrast, water and the long-chain fats making up vegetable oil are not miscible: after you stopshaking that bottle of salad dressing you just made, the oil will gradually separate from theaqueous vinegar solution, and you will finally see two separate layers of immiscible liquids (oiland water).1-2

A colloid is defined as a mixture of two immiscible substances, with one evenly distributedthroughout the other. An emulsion is a class of colloids in which both mixed substances areliquids. When well shaken, the salad dressing described above is an emulsion. So ishomogenized milk, an emulsion of milk fat particles suspended in an aqueous solution ofwater, milk sugars, and milk proteins. An emulsion consists of a continuous phase (e.g., theaqueous portion of milk) in which a discontinuous phase (e.g., the fat globules in milk) issuspended.II. Applying Your Knowledge to a Public Health IssueA. ProcedureBelow, you will find a case study, a scenario of scientific interest. Work in groups of four toanalyze this case study and do the following.1. Consider the case study and its premise.2. Formulate sets of competing hypotheses about the premise.3. Choose one set of competing hypotheses to test.4. Make a prediction about results you would expect if you were to experimentally testthese hypotheses.5. Design and perform a careful, rigorous experiment to test these hypotheses.6. Analyze the data from your experiment.7. Decide which of your original, competing hypotheses is incorrect.8. Discuss conclusions about the hypothesis that was not rejected. Are your explanationsthe only reasonable, logical ones? What others might there be? How would you testthem?B. Case Study: Preventing sexual transmission of HIV with latex condomsAs biology majors, possible future health professionals, and concerned members ofspecies Homo sapiens, you undoubtedly are aware of AIDS, Acquired ImmunodeficiencySyndrome. The causative agent of this disease is HIV, the Human ImmunodeficiencyVirus. HIV is a retrovirus, a virus that stores its genetic code in the form of RNA, and thenuses the cellular machinery of the host cell to create DNA versions of its genome. These arethen copied into RNA to manufacture more HIV particles. HIV is one of several viruses knownas lentiviruses (from the Latin lenti, meaning “slow”), named for their relatively long incubationperiod. They can exist (and, in the case of HIV, replicate) in host cells for long periods withoutcausing clinical signs of disease.One reason that HIV is such an effective pathogen is that it targets cells vital to the host’simmune response. HIV enters specific cells of the immune system known as CD4 T cells.While the virus can lie dormant within these cells and temporarily cause no harm, theyeventually kill the cells, either by lysing them, increasing their rate of apoptosis (programmedcell death), or by causing them to be killed by other immune cells (CD8 cytotoxic lymphocytes).Once the host’s T-cells are depleted below a certain level, the immune response is severelycompromised, and AIDS patients often die from infections caused by opportunistic pathogensthat would not be able to harm a person with a healthy immune system.HIV is transmitted between humans via contact with virus-infected bodily fluids such asblood, breast milk, vaginal fluid, and semen. Unprotected sexual intercourse is known to be amajor source of HIV transmission, and is one reason the virus has become a global pandemic.1-3

HIV, approximately 116 nanometers in diameter, cannot pass through a synthetic condom.Hence, use of a synthetic condom during sexual intercourse is believed to be an effective wayto prevent transmission of the virus from an infected individual to his or her sex partner. TheCenters for Disease Control (CDC) provides an overview of the use of condoms to preventsexually transmitted disease at http://www.cdc.gov/condomeffectiveness/latex.htmTo be an effective barrier to virus transmission, a condom must be intact. Any compromiseto the structural integrity of the condom puts the uninfected partner at risk of contracting HIV.Latex is a general term used to describe an emulsion of short polymers (i.e., chains ofrepeating chemical subunits) in an aqueous solution. Modern condoms are most oftenmanufactured out of latex derived from natural rubber. Commercial rubber is harvested fromthe Para Rubber Tree, Hevea brasilensis (Family Euphorbiaceae), native to the AmazonRainforest. The plant produces a sticky, milky, saplike substance, beneficial to the tree as adeterrent to herbivory. In its raw form, the latex harvested from H. brasilensis contains variousproteins, secondary plant metabolites, and solvents, but most of these are removed duringprocessing. (Some impurities remain, however, and are the allergens that cause potentiallydangerous anaphylactic reactions in about 1% of U.S. residents. If you have a known allergyto latex, DO NOT handle the condoms provided for this laboratory exercise!)Purified latex is an emulsion of isoprene polymers (Figure 2) in an aqueous solution.Condoms are manufactured by dipping forms into the liquid latex, and then subjecting thecoated forms to slow drying in a convection oven. This removes much of the aqueous phase,leaving a network of isoprene and other short polymers to form a thin, elastic sheath.As has already been mentioned, the use of latex condoms during intercourse is consideredone of the most effective ways to prevent transmission of HIV. However, some sexualpractices may reduce the efficacy of this barrier. For example, various types of lubricants arecommonly used with latex condoms during sexual intercourse. These include various gels,lotions, and oils which may or may not interact with the molecules of the condom. At the backof the laboratory, you will find an array of such substances that might be used for this purpose.Check the labels of these substances. Then, using the resources available to you, determineas much of the chemical nature of these substances as possible.C. Questions inspired by the Case StudyConsidering your knowledge of basic chemistry and the Case Study you have just read,use the space below to record your ideas about solvents, lubricants, and the structural integrityof latex condoms. What hypotheses can you formulate regarding the interaction of particularsolvents in lubricants and the latex material of which most condoms are made?1-4

III. Designing your experimentThis is arguably the most critical aspect of the scientific method. The most meticulousmeasurements in the world will not be useful if you have not formulated meaningfulhypotheses and devised logical, rigorous methods for collecting data.As a team, discuss the ideas you wrote down about the case study. Which of these lead topredictions about the effects of a particular solvent (or solvents) on the integrity of thecondom? Devise hypotheses that could be tested with experimental trials in this laboratory inthe next two hours or so. Select the most interesting, relevant question and move to the nextstep.A. Introduction and Formulating HypothesesIn the space below, give a brief introduction to the problem you are addressing. Give asmuch background as possible, considering the brief time you’ve had to acquaint yourself withthis issue. (Hint: If you have laptops with you, use them!) Why is this problem interesting?Why is it relevant and important to examine?By now, you have read Appendix 1 and are aware of when it is appropriate to use statisticalhypotheses versus the strong inference method. Unless you already know the effect of aparticular lubricant on condom integrity, and are prepared to now propose multiple, competinghypotheses to explain why that lubricant affects the condom, then you are still in theexploratory phase of this investigation. Hence, it is most appropriate to use the statisticalmethod if you are trying to answer a broad question. What are your hypotheses?Null Hypothesis :Alternative Hypothesis:1-5

Is your alternative hypothesis one-tailed or two-tailed? Explain why.B. MethodsOnce you have constructed your hypotheses, you must determine what you will measure inorder to test them. For example, if you have predicted that a particular polar solvent willdamage condom integrity, you might wish to measure some aspect of the condom's physicalcharacteristics (e.g., its volume capacity before breaking, its permeability to varioussubstances, its stretching capacity before breaking, etc.).Your parameter should be something that can be measured in two different groups, thecontrol group, in which all conditions are held constant, and the treatment group, in which allconditions are exactly the same as for the control group with the exception of a singleexperimental variable.In the spaces provided below, describe the experimental protocols you will use to test yourhypotheses. How many replications will you perform? Do you have treatment and controlgroups? What is the difference between the treatment and control groups? What are theunique aspects of your experiment? (You need not give a detailed explanation of equipmenthere, unless it is unique to your experiment and really requires an explanation for the educatedreader to understand what you will be doing.)Survey the materials and equipment available in the lab. There should be various types ofglassware, syringes, meter sticks, rulers and other measuring tools. There are also tools thatyou may use to design a mechanical test of condom integrity. Finally, there is a selection ofvarious liquid or colloidal substances, polar, non-polar, and combined, that could be used inconjunction with a condom during sexual intercourse. Using the available materials (or othersyou can obtain quickly--be creative), design an experiment to test your hypotheses.IMPORTANT: If you have decided to collect parametric (continuous numerical) data, youshould use only two different lubricants, or one lubricant and a control group (receiving aplacebo treatment). If you use more than one type of lubricant, you cannot use the two-samplet-test explained in Appendix I. A different statistical test (e.g., ANOVA) must be used to testwhether there is a significant difference among multiple continuous numerical means. Unlessyou already know how to do this, we strongly suggest that you design your experiment tocompare the independent means of two different experimental groups. If you choose to collectattribute data, you may still use the Chi-square test. But apply it with caution, and know how tointerpret the results. (HINT: For an experiment such as this one, comparison of treatment andcontrol means will usually yield meaningful results.)IMPORTANT AGAIN: Once you subject a condom to a particular experimental procedure,its physical properties will likely have changed. Although you might be able to design anexperiment in which you subject the same condom to two different treatments, for the purposeof keeping things relatively simple for this lab, it is best to use a condom only once, unless youare truly able to ensure that using the same one twice will not introduce some type of error intoyour results.1-6

1. List the materials your group will use2. Briefly describe your experimental set-up and procedure.3. How many times will you replicate your experiment in each of thegroups (treatment and control)?C. Parameters and Data AnalysisYou will also have to decide whether to collect attribute, discrete numerical, or continuousnumerical data, as described in Appendix 2. For example, would it be more informative tomeasure the degree to which condom integrity is affected by a particular substance? Or willyou simply test whether integrity is affected or not? You must decide which type of data willmake your experiment the most powerful and convincing, once analyzed with an appropriatestatistical test. Whatever you choose to do, you must decide in advance the type of data youwill collect and which statistical test you will use to analyze your data.1. What parameter will you measure?2. What statistical test will you use to analyze your data?D. Predictions and InterpretationsIn the spaces below, list all possible outcomes of your experimental trials. (You might nothave five possible outcomes, but we wanted to give you space in case you’re feeling verycreative.) Each outcome should be accompanied by an explanation of the most plausiblereason for this outcome. Remember that the result is not the same as the interpretation. Theoutcome is the summary of the results themselves, whereas the interpretation is the logical,reasoned explanation for the observed results.The explanation of your outcome can serve as the basis for the next level of posinghypotheses, so be prepared to go to that next level.1-7

Outcome 1:Explanation:Outcome 2:Explanation:Outcome 3:Explanation:Outcome 4:Explanation:Outcome 5:Explanation:1-8

IV. Conducting your experimentOnce you have designed your experiment and are confident that it has a good chance ofyielding meaningful results, collect the materials necessary from the materials supplied in laband set up your trials.A. Collecting DataRemember to keep your treatment and control group as similar as possible, with theexception of a single experimental variable. This means that the same group members shouldbe assigned the same tasks each time, to eliminate as much human error as possible. (Thismeans that the same person should do the same tasks for both treatment and control, so thatyou don’t introduce a difference between your groups simply because of a difference in theexact techniques used by two different people.)Below is a grid for recording your raw data. We will let you label the columns and rows asyou see fit, and of course you may not need the entire grid.Grid for Data Collection:B. Analyzing DataA statistical test is a tool used by scientists to determine whether any variation from anexpected result is due simply to random error, or due to some non-random phenomenon thatbears further investigation. In next week's lab, you will engage in several exercises designed1-9

to teach you how to use statistics to analyze your data. We will do a shortened "preview" ofsuch statistics now.If your team decided to take non-parametric data, refer to Appendix I for instructions onhow to apply the Chi-square test. If your team collected continuous numerical data from atreatment and control group, use the following information to analyze your data.The Independent sample t-testIf the measurements you made on your condoms were taken from separate individualcondoms in two different test groups, then none of the individual condoms were subjected topaired measurements, which would mean that the same condom was used as both atreatment and a control.If this is the case in your experiment, you must use a statistical test designed to showwhether there is a significant difference between the means of two independent samples,those of your (1) treatment and (2) control groups.Use the following equation to calculate a t-statistic for your two means:...where x 1 and and x 2 are the means of your two groups, n 1 and n 2 are the numbers ofcondoms you used in each group, and s p 2 is the pooled variance, calculated as:...where s 1 2 is the variance of group 1, s 2 2 is the variance of group 1, df 1 is the degrees offreedom for group 1 (df 1 = n 1 - 1) and df 2 is the degrees of freedom for group 2 (df 2 = n 1 - 1)The variance (s 2 ) for a mean is calculated as_In which x is the mean, x is each individual value, and n is the sample size. You will learnmore about this in the lab on collecting data and using statistical tests to analyze data. But fornow, you can just plug in your numbers.The degrees of freedom for a two-sample t-test with independent means is calculated asthe sum of the degrees of freedom of each test group:What is your t-statistic?df = (n 1 - 1) + (n 2 - 1)1-10

What are your degrees of freedom?What is the P value associated with your t-statistic and degrees of freedom? (See Table 1-1.)> P >The paired sample t-testIf your experiment involved taking measurements from the same condom subjected to twodifferent treatments (essentially giving you "before" and "after" measurements of a singlecondom), then the two values you obtain will not be independent of one another, and are saidto be paired.In a paired sample t-test, the separate means of two sample populations is not measured.Instead, the difference between your first measurement and your second measurement on thesame individual is used to calculate the t-statistic. The difference between your first andsecond measurements is represented as d, and the mean of all your individual differences as_d.The standard deviation and variance are calculated as you did them before, but substituting_ _d and d for x and x in the equations. Thus, to calculate the variance of your paired samples,you will use the following equation:To calculate the standard deviation, take the square root of the variance._1. Calculate the mean difference (d) between your first and second measurements:2. Calculate the variance of the differences in between your first and second measurements(s d 2 ).3. Calculate the standard deviation (square root of (s d 2 )):You're now almost ready to plug these values into the t-test for paired samples. The last thingyou must do is to determine the number of independent quantities in your system, or degreesof freedom.4. The degrees of freedom (df) determine the significance level tied to every possible value ofa statistic (such as the t-statistic). The degrees of freedom is the number of data points thatare free to vary without changing the test statistic, and the df change depending on the type ofstatistic being calculated. For the paired sample t-test,df = (n - 1)...where n is the number of independent quantities in your system. For paired samples, n =the number of pairs in your system. In this case, the number of condoms you tested.Your team of four should now calculate a t-statistic for the difference between your twomeans, and use this to determine whether the difference between them is significant. Use thefollowing equation for the paired sample t-test:1-11

...where n is the number of pairs.What is the value of your t-statistic?What are your degrees of freedom?What is the P value associated with your t-statistic and degrees of freedom? (See Table 1-1.)> P >Table 1-1. Table of critical values for the two-sample t-test. The P levels (0.05) indicatinglevel of deviation from the expected not due to chance are shown in bold for both one-tailedand two-tailed hypotheses. (From Pearson and Hartley in Statistics in Medicine by T. Colton,1974. Little, Brown and Co., Inc. publishers.)2-tail --> 0.10 0.05 0.02 0.01 0.0011-tail --> 0.05 0.02 0.01 0.005 0.0005df1 6.314 12.706 31.821 63.657 636.6192 2.920 4.303 6.965 9.925 31.5983 2.353 3.182 4.541 5.841 12.9414 2.132 2.776 3.747 4.604 8.6105 2.015 2.571 3.365 4.032 6.8596 1.934 2.447 3.143 3.707 5.9597 1.895 2.365 2.998 3.499 5.4058 1.860 2.306 2.896 3.355 5.0419 1.833 2.262 2.821 3.250 4.78110 1.812 2.228 2.764 3.169 4.58711 1.796 2.201 2.718 3.106 4.43712 1.782 2.179 2.681 3.055 4.31813 1.771 2.160 2.650 3.012 4.22114 1.761 2.145 2.624 2.977 4.14015 1.753 2.131 2.602 2.947 4.07316 1.746 2.120 2.583 2.921 4.01517 1.740 2.110 2.567 2.898 3.96518 1.734 2.101 2.552 2.878 3.92219 1.729 2.093 2.539 2.861 3.88320 1.725 2.086 2.528 2.845 3.85021 1.721 2.080 2.518 2.831 3.81922 1.717 2.074 2.508 2.819 3.79223 1.714 2.069 2.500 2.807 3.76724 1.711 2.064 2.492 2.797 3.74525 1.708 2.060 2.485 2.787 3.72526 1.706 2.056 2.479 2.779 3.70727 1.703 2.052 2.473 2.771 3.69028 1.701 2.048 2.467 2.763 3.67429 1.699 2.045 2.462 2.756 3.65930 1.697 2.042 2.457 2.750 3.6461-12

C. Drawing ConclusionsWhen all groups have completed their trials, analyzed their data and drawn conclusionsabout their experiments, your TA will lead a class discussion. Answer the following.Does your P value indicate that your two means are significantly different from one another?Which hypothesis do your data indicate is incorrect?What is your group’s conclusion about your experiment and original question you posed aboutthe Case Study?If you were to try to design this experiment again, what changes/improvements would youmake?Based on your conclusions, what are some new hypotheses you could pose to further explorethis system?1-13

V. Peer Persuasion.Once your experiments are complete, each team will have about five minutes to present itsfindings to the rest of the class. Your team's goal is to convince your colleagues (classmates)that your original questions, hypotheses and experimental design and analysis are sound.Your colleagues' goal is to find the holes in your arguments, if they see them. Your TA will actas moderator and Omnipotent Judge.Each team will present its question, hypotheses, predictions, experimental procedures,results and conclusions to the entire class. You may either appoint a team representative, ortake turns presenting the various parts of the project. Be sure to also pose new hypothesesbased on your results and conclusions, since your outcomes are essentially new observations.When your team has finished its brief overview, your classmates will have a few minutes toask questions and evaluate what your team has done. You will have a chance to answer theirquestions and convince them that your experimental design was sound, and that yourconclusions accurately reflect reality--or to accept that changes might improve this protocol ifyou were to repeat it.Remember: the goal is to use your critical thinking skills to help your colleagues designbetter, more rigorous experiments.1-14