Irish Leaving Certificate Biology

What scientific knowledge is

Scientific knowledge is a body of explanations that have been tested against evidence. A fact is an observation that can be confirmed, such as the temperature at which a sample of liquid boils. A hypothesis is a proposed, testable explanation for an observation. A theory is a well-tested explanation that draws together many observations and has survived repeated attempts to disprove it. In everyday speech "theory" means a guess, but in science the theory of evolution by natural selection or the cell theory are among the best-supported ideas there are, because they explain a large amount of data and have made predictions that turned out to be correct.

Scientific ideas are modified over time. When new evidence cannot be explained by an existing idea, the idea is adjusted or replaced. The classification of living things is a good example: organisms were once placed in two kingdoms, plants and animals, and later in five and six kingdoms as microscopes and then DNA sequencing revealed differences that the older scheme could not show. The change was not an error that embarrassed science; it was the process working, with better tools producing better models.

Science is also a global activity. Scientists in different countries build on each other's results, which only works because they share conventions. SI units mean that a mass recorded in Dublin is read the same way in Tokyo. Binomial names mean that a species has one internationally agreed name whatever the local language calls it. Standard methods let another laboratory repeat an investigation.

Peer review and reproducibility

Before a scientific result is accepted it is written up with its methods and data and sent to a journal, where other scientists in the same field, the peer reviewers, judge whether the methods are sound, whether the data support the conclusions and whether the work is original. Reviewers may ask for more data or reject the paper. Peer review does not guarantee that a result is right, but it removes many errors and makes clear, open reporting the norm.

Reproducibility means that an independent team, following the published method, gets results that agree with the original ones. A claim that only one laboratory can obtain is not yet reliable knowledge. A finding is also more convincing when it has been reached by different methods, for example when the fossil record, comparative anatomy and DNA sequences all point to the same evolutionary relationship.

Two related terms are worth keeping apart. Repeatability is getting consistent results when the same person repeats the measurement with the same equipment. Reproducibility is getting consistent results when a different person, place or method is used.

Bias

Bias is a systematic leaning that distorts how evidence is collected, reported or interpreted. It can enter in several ways.

• Selective reporting, where results that support a preferred conclusion are published and the rest are left out.
• Sampling bias, where the sample does not represent the population. A survey of one school's students says little about all teenagers.
• Funding and commercial interest, where the body paying for a study gains from a particular result. This does not make the result wrong, but it is a reason to look harder at the methods.
• Confirmation bias, where a researcher gives more weight to observations that agree with what they expected.
• Wording and presentation, for example a graph whose axis starts well above zero so that a small difference looks large.

Scientists reduce bias by using control groups, random allocation, blind procedures where the person measuring does not know which sample is which, large enough samples and open publication of methods and data.

Evaluating sources

When researching a scientific issue, such as the effect of a pesticide on bees or the safety of a vaccine, the sources used will differ in quality. A useful set of questions is these.

1. Who produced the information and what are their qualifications and interests?
2. Is it a primary source (the original report of an investigation) or a secondary source (a summary or discussion of other people's work)?
3. Has it been peer reviewed, or is it an opinion piece, an advertisement or a social media post?
4. How large and how representative was the sample, and what was compared with what?
5. Is the evidence given, or only a conclusion? A source may lack detail, so that the reader cannot judge the claim.
6. Is the date recent enough, and do other independent sources agree?

Primary data are collected by the person using them, for example their own measurements of enzyme activity. Secondary data were collected by someone else, such as a national database of species records or a published table of results. Both can support a conclusion, but secondary data need checking for how they were gathered and for what is missing.

Media claims

A headline claiming that a food "causes" or "cures" something needs the same scrutiny. Correlation, two things changing together, is not evidence that one causes the other; a third factor may drive both. A result in mice, or in a test tube, may not hold in humans. A relative risk reported without the baseline risk can make a small effect sound alarming: doubling a risk of one in a million still leaves a very small risk. A claim is stronger when the study was controlled, the sample was large and the finding has been reproduced.

Science in society

Scientific knowledge makes possible advances such as vaccines, crop breeding and genetic testing, but whether and how it is applied depends on more than the science. Economic factors (cost, and who profits), social factors (public attitudes and access), sustainability (effects on resources and the environment) and ethical factors (what ought to be done, and to whom) all influence decisions. Science can say what a gene-editing technique does and how likely it is to work; it cannot alone decide whether it should be used on human embryos. That decision involves values that scientists, legislators and the public discuss together.

Science also has limits. It deals with questions that can be tested by observation and measurement, it gives conclusions that are provisional, and its data always carry some uncertainty.

Scientists contribute to society through discoveries and inventions, and their impact can be assessed by looking at what changed as a result. Examples include Edward Jenner's smallpox vaccination, which led to the eradication of smallpox in 1980, Rosalind Franklin's X-ray diffraction images of DNA, which gave the evidence for the double helix worked out by Watson and Crick, and Alexander Fleming's observation of the antibacterial mould Penicillium, developed into a medicine by Florey and Chain. In each case a discovery depended on earlier work by others, took years to apply, and had effects, good and bad, that nobody predicted at the start.