Natural Science Testing Scientific Ideas. TestingTesting hypotheses and theories is at the core of the process of science. Any aspect of the natural.

Natural Science

Testing Scientific Ideas

Testing hypotheses and theories is at the core of the process of science. Any aspect of the natural world could be explained in many different ways.

It is the job of science to collect all those plausible explanations and to use scientific testing to filter through them, retaining ideas that are supported by the evidence and discarding the others.

testIn science, an observation or experiment that could provide evidence regarding the accuracy of a scientific idea. Testing involves figuring out what one would expect to observe if an idea were correct and comparing that expectation to what one actually observes.

hypothesisA proposed explanation for a fairly narrow set of phenomena, usually based on prior experience, scientific background knowledge, preliminary observations, and logic.

theoryIn science, a broad, natural explanation for a wide range of phenomena. Theories are concise, coherent, systematic, predictive, and broadly applicable, often integrating and generalizing many hypotheses. Theories accepted by the scientific community are generally strongly supported by many different lines of evidence-but even theories may be modified or overturned if warranted by new evidence and perspectives.

scienceOur knowledge of the natural world and the process through which that knowledge is built. The process of science relies on the testing of ideas with evidence gathered from the natural world. Science as a whole cannot be precisely defined but can be broadly described by a set of key characteristics.

natural worldAll the components of the physical universe — atoms, plants, ecosystems, people, societies, galaxies, etc., as well as the natural forces at work on those things. Elements of the natural world (as opposed to the supernatural) can be investigated by science.

evidenceTest results and/or observations that may either help support or help refute a scientific idea. In general, raw data are considered evidence only once they have been interpreted in a way that reflects on the accuracy of a scientific idea.

You can think of scientific testing as occurring in two logical steps: (1) if the idea is correct, what would we expect to see, and (2) does that expectation match what we actually observe?

expectationIn science, a potential outcome of a scientific test that is arrived at by logically reasoning about a particular scientific idea (i.e., what we would logically expect to observe if a given hypothesis or theory were true or false). The expectations generated by an idea are sometimes called its predictions. Observations that match the expectations generated by an idea are generally interpreted as supporting evidence. Mismatches are generally interpreted as contradictory evidence.

observeTo note, record, or attend to a result, occurrence, or phenomenon. Though we typically think of observations as having been made "with our own eyes," in science, observations may be made directly (by seeing, feeling, hearing, tasting, or smelling) or indirectly using tools.

Ideas are supported when actual observations (i.e., results) match expected observations and are contradicted when they do not match.

SummaryTesting ideas with evidence is at the heart of the process of science. Scientific testing involves figuring out what we would expect to observe if an idea were correct and comparing that expectation to what we actually observe.

Science neither proves nor disproves. It accepts or rejects ideas based on supporting and refuting evidence, but may revise those conclusions if there are reasons such as new evidence or perspectives.

TESTING IDEAS ABOUT

CHILDBED FEVER.

A doctor worked on a maternity ward in the 1800s.

In his ward, an unusually high percentage of new mothers died of what was then called childbed fever. The doctor considered many possible explanations for this high death rate.

In the late 1840's, Dr. Ignaz Semmelweis was an assistant doctor.

One thing that he took care of was the delivery rooms. (rooms where women give birth to babies)

23

Semmelweis found that the death rate in a delivery room staffed by medical students was up to three times higher than in a second delivery room staffed by midwives.

25

In fact, women were terrified of the room staffed by the medical students.

26

First clinic Second clinic

Year Births Deaths Rate (%) Births DeathsRate (%)

1841 3,036237 7.8 2,44286 3.51842 3,287518 15.8 2,659202 7.61843 3,060274 9.0 2,739164 6.01844 3,157260 8.2 2,95668 2.31845 3,492241 6.9 3,24166 2.01846 4,010459 11.4 3,754105 2.8

Puerperal fever mortality rates for the First and Second Clinic at the Vienna General Hospital 1841–1846. The First Clinic has the larger mortality rate.

Semmelweis observed that the students were coming straight from their lessons in the autopsy room to the delivery room.

29

Autopsy – cutting up dead bodies to find out how they died.

30

Two of the many ideas that he considered were (1) that the fever was caused by mothers giving birth lying on their backs (as opposed to on their sides) and (2) that the fever was caused by doctors' unclean hands (the doctors often performed autopsies immediately before examining women in labor)

He tested these ideas by considering what expectations each idea generated.

Please go to www.socrative.com Student log inRoom number 7318a Semmelweis considered that the fever

might be caused by mothers giving birth lying on their backs (as opposed to on their sides). What expectations could this idea generate?

www.socrative.com Semmelweis considered that the fever was

caused by doctors' unclean hands (the doctors often performed autopsies immediately before examining women in labor). What expectations could this idea generate?

He tested these ideas by considering what expectations each idea generated. If it were true that childbed fever were caused by giving birth on one's back, then changing procedures so that women labored on their sides should lead to lower rates of childbed fever. The doctor tried changing the position of labor, but the incidence of fever did not decrease; the actual observations did not match the expected results.

www.socrative.com The doctor tried changing the position of labor,

but the incidence of fever did not decrease; the actual observations did not match the expected results. Do you think that the idea was supported by the evidence?

If childbed fever were caused by doctors' unclean hands, having doctors wash their hands thoroughly with a strong disinfecting agent before attending to women in labor should lead to lower rates of childbed fever. When the doctor tried this, rates of fever fell; the actual observations matched the expected results, supporting the second explanation.

www.socrative.com When doctors washed their hands thoroughly

with a strong disinfecting agent before attending to women in labor , rates of fever fell. Do the actual observations match the expected results?

DOES HAND WASHING WORK?SEMMELWEIS - 1847

MonthMonth BirthsBirths DeathsDeaths % Mortality% Mortality

AprilApril 312312 5757 18.318.3

MayMay 294294 3636 12.212.2

JuneJune 268268 66 2.42.4

JulyJuly 250250 33 1.21.242

www.socrative.com Describe what happens in the slide labeled

“Does Hand Washing Work?”

www.socrative.com Can you develop different explanations for

what happens in the slide labeled “Does Hand Washing Work?”

Here is some of his evidence

Puerperal fever monthly mortality rates for the First Clinic at Vienna Maternity Institution 1841–1849. Rates drop markedly when Semmelweis implemented chlorine hand washing mid-May 1847

In his 1861 book, Semmelweis presented evidence to demonstrate that the advent ofpathological anatomy in Wien (Vienna) in 1823 (vertical line) was accompanied by the increased incidence of fatal childbed fever. The second vertical line marks introduction of chlorine hand washing in 1847. Rates for the Dublin maternity hospital, which had no pathological anatomy, are shown for comparison.

Over time doctors began to accept the ideas of Semmelweis. Doctors and nurses began to wash their hands more often and more carefully, using strong soaps and disinfectants.

Now the ideas are strongly accepted and encouraged by major health groups.

(Semmelweiss’ story at Encyclopedia Britannica)

In 1848 a liberal political revolution happened in Europe. Semmelweis took part in the events in Vienna. The revolution was crushed. Semmelweis’ political activities had increased the obstacles to his professional work. In 1849 he lost his post at the clinic. He then applied for a teaching post at the university in midwifery but was turned down.

Soon after that, he gave a successful lecture at the Medical Society of Vienna entitled “The Origin of Puerperal Fever.” He applied once more for the teaching post, but, although he received it, there were restrictions attached to it that he considered humiliating. He left Vienna and returned to Pest in 1850.

He worked for the next six years at the St. Rochus Hospital in Pest. An epidemic of puerperal fever had broken out in the obstetrics department, and, at his request, Semmelweis was put in charge of the department. His work quickly reduced the death rate, and in his years there it averaged only 0.85 percent. In Prague and Vienna, meantime, the rate was still from 10 to 15 percent.

In 1855 he was appointed professor of obstetrics at the University of Pest. He married, had five children, and developed his private practice. His ideas were accepted in Hungary, and the government sent a circular to all district authorities ordering the introduction of the methods of Semmelweis.

In 1861 Semmelweis published his principal work, Die Ätiologie, der Begriff und die Prophylaxis des Kindbettfiebers (The Etiology, Concept, and Prophylaxis of Childbed Fever). He sent it to all the prominent obstetricians and medical societies abroad, but the general reaction was adverse. The weight of authority stood against his teachings.

He addressed several open letters to professors of medicine in other countries, but to little effect. At a conference of German physicians and natural scientists, most of the speakers—including the pathologist Rudolf Virchow—rejected his doctrine.

The years of controversy gradually undermined his spirit. In 1865 he suffered a breakdown and was taken to a mental hospital, where he died. His illness and death were caused by the infection of a wound on his right hand, apparently the result of an operation he had performed before being taken ill. He died of the same disease against which he had struggled all his professional life.

• For example here are some prezis about Semmelweis.

http://prezi.com/explore/search/?csrfmiddlewaretoken=d9fc3764ef3ab9971bf9a99dd257ff04&search=Semmelweis#search=Semmelweis&reusable=false&page=1&users=less

–Hands are the most common vehicle to transmit health care-associated pathogens

HANDS ARE THE MAJOR SOURCE OF PATHOGENS

57

• Most common mode of transmission of pathogens is via hands!

SO WHY ALL THE FUSS ABOUT HAND HYGIENE?

58

59

GLOBAL HAND WASHING DAY

60

Global Hand washing Day is a campaign to motivate and mobilize millions around the world to wash their hands with soap. The campaign is dedicated to raising awareness of hand washing with soap as a key approach to disease prevention.

HAND WASHING A TRIBUTE TO DR. IGNAZ SEMMELWEIS

61

Testing ideas with evidence is at the heart of the process of science.

Scientific testing involves figuring out what we would expect to observe if an idea were correct and comparing that expectation to what we actually observe.

Fair Tests

Which restaurant serves the best, cheapest meals? What is causing your runny nose? Why won't your computer work? If you want to answer questions like these, you'll probably need to do some testing.

To find the real answers to such questions, you'll need to test your ideas in a fair way.

The parts of a fair test are the same:Comparing outcomes. Controlling variables.Avoiding bias. Distinguishing chance from real differences.

Comparing outcomes. To be confident in test results, it's generally important to have something to compare them to. In experiments, whatever you are comparing your test results to is sometimes called the control group or control treatment.

control groupIn scientific testing, a group of individuals or cases matched to an experimental group and treated in the same way as that group, but which is not exposed to the experimental treatment or factor that the experimental group is. Control groups are especially important in medical studies in order to separate placebo effects from outcomes of interest. Control groups are sometimes also called control treatments or simply controls. This can be confusing since this use of the term is slightly different from what we mean when talking about controlled variables.

Controlling variables. In most tests, we want to be confident in the relationship between cause and effect. To be able to make a strong statement about cause and effect, you'll need to control variables — that is, try to keep everything about the test comparisons the same, except for the variables you're interested in.

controlIn scientific testing, to keep a variable or variables constant so that the impact of another factor can be better understood.

Avoiding bias. No matter how hard we try to be objective, bias can effect our observations and judgments. In a sense, bias occurs because it's very difficult to "control" variables associated with human judgments.

objectiveNot influenced by biases, opinions, and/or emotions. Scientists strive to be objective in their reasoning about scientific issues.

biasAny deviation of results or inferences from the truth, or processes leading to such deviation.Bias can result from several sources: one-sided or systematic variations in measurement from the true value (systematic error); flaws in study design; deviation of inferences, interpretations, or analyses based on flawed data or data collection; etc. There is no sense of prejudice or subjectivity implied in the assessment of bias under these conditions.For example, when a researcher or patient knows what treatment is being given. To avoid bias, a blinded study may be done.

Distinguishing chance from real differences. All sorts of subtle things that you either don't or cannot control can affect the outcome of a test. All of these random factors will affect the outcome of the test — but in small ways. So how do you know if the outcomes of a test are due to random factors or to real differences?

First, sample size is important. The larger your sample size, the more likely it is that these random factors will cancel each other out and that real differences (if they exist) can be detected statistically.

Secondly, statistics can be used to analyze your raw data. The purpose of conducting such statistical tests is to tell you how likely it is that a difference in rating like the one that you observed is actually due to random factors.

sampleIn science, to collect information from part of an entity, with the aim of learning about the entity as a whole (e.g., to collect information on a subset of the members of a population or on cores of ice from the Antarctic). The term sample size refers to the number of repeated measurements made (e.g., the number of individuals surveyed or the number of ice cores studied). All else being equal, the larger the sample size, the more confident we can be that our sample represents the entity as a whole and the more subtle the difference between samples that we'll be able to discriminate.

dataInformation got from observations — usually observations that are made in a standardized way. The term data generally refers to raw data — information that has not yet been analyzed. Data (multiple pieces of information) is the plural form of datum (a single piece of information).

Here is a simple example of how sample size can effect the results.

People in Europe often saw these birds. They are swans.

80

81

People from Europe thought that all swans were white.

This can be written as a hypothesis.

If it is a swan,

then it will be white.

82

People from Europe thought that all swans were white.

This can be written as a hypothesis.

What hypothesis could you write?

When Europeans went to Australia they saw these birds.

83

84

They had a hypothesis, and looked for evidence, but their sample size wasn’t large enough.

DETECTING THE DIFFERENCES: STATISTICS AND SAMPLE SIZE

What is a "large" sample size? It depends on how small a difference between groups you want to be able to detect. If you are interested in very tiny differences, you need a very large sample size, and if you only care about pretty big differences, you can get away with a smaller sample size. The appropriate sample size depends on the statistical tests you want to run and the sorts of differences you want to detect.

It is often impossible to make a test perfectly fair, and each issue may be more or less important for a particular test — but by considering each of these factors in how your test is designed, you can maximize the amount of useful information you get from the test.

These considerations can be applied to tests in everyday life, and in more scientific areas — and to tests that don't involve experiments.

Summary

Designing a fair test of an idea — in formal science or in everyday life — means deciding what results you'll be comparing, controlling variables, avoiding bias, and figuring out a way to distinguish chance differences from meaningful ones.

Controlled variables are those factors that are kept constant across a test, so that the effect of another variable can be better observed.

The larger the sample size a test employs, the smaller the difference that the test will be able to detect.