Lesson 3.03: Using Evidence

Welcome to Lesson 3.03, where we will explore the critical skill of using evidence. Evidence is fundamental to many activities, from academic research to policy-making and popular journalism. Understanding how evidence is used, and how to evaluate it, is a vital part of critical thinking.

The many ways evidence is used can be broadly categorized into two main areas: explaining evidence and drawing inferences

Explaining Evidence

When we explain evidence, we are essentially looking backward. We are trying to make suggestions about what happened before the evidence to bring it about. For instance, astrophysicists observe phenomena in the universe and then try to come up with explanations for the birth, life, and death of stars, planets, and galaxies based on their observations

Drawing Inferences

In contrast, drawing inferences involves moving forward, beyond the evidence. This means using the evidence to predict future occurrences or to recommend policies or courses of action.

You may be asked to assess an explanation or inference that has already been presented, or you might be asked to suggest your own explanation or inference. It's important to recognize that there's a strong connection between these two tasks: if you can suggest alternative explanations or inferences, it often highlights a weakness in the explanation or inference that was initially provided.

Assess Explanation for Evidence

When evaluating explanations, keep in mind that they may be weak if:

They fail to account for the whole of the evidence they claim to explain. A strong explanation should cover all available facts.
They rely on speculative additional information or unstated assumptions. If an explanation requires you to assume too much without support, it's likely weak.
Other explanations are at least equally plausible. If there are multiple credible ways to explain the same evidence, the initial explanation might not be the strongest or only one.

Understanding Correlation

A common mistake in using evidence is confusing correlation with causation.

Correlation is defined as a statistical relationship between sets of data, or the tendency for phenomena to occur together. For example, serious researchers, science popularizers, and journalists often base claims on correlations between statistics

It's tempting to claim that eating processed meat causes heart disease if you observe that heart disease occurs more frequently in people who eat a lot of processed meat. However, simply because two things happen together (correlate) does not automatically mean one causes the other. As some commentators humorously point out about diet and health claims, it often seems like everything either causes cancer or cures it, or sometimes both, highlighting the overuse of such claims

Why Correlation Doesn't Always Mean Causation:

Coincidence: With billions of data sets available today, and a limited number of patterns, it's inevitable that many unrelated sets of data will resemble each other. This leads to "spurious correlations". For example:

US spending on science, space, and technology closely correlates with the number of suicides by hanging, strangulation, and suffocation

The number of films starring Nicolas Cage between 1999 and 2009 correlates closely with the number of people who drowned by falling into a pool during the same period

Per capita consumption of mozzarella cheese between 2000 and 2009 correlates closely with the number of doctorates awarded in Civil Engineering

It is virtually certain that none of these correlations has any causal basis whatsoever. They are simply coincidental.

2. Complex Causal Relationships: If there is a causal basis for a correlation between two sets of data, the relationship might not be as simple as A causing B.

For example, if the Divorce Law Reform Act in Westland made it easier to obtain a divorce, and if the number of divorces in the country rose significantly following the introduction of the Act, it is likely that the change in the law was partly responsible for the rise in the number of divorces. But the change in the law was probably in part a response to a change in public opinion, which made divorce more acceptable. In addition, people who knew the new law was likely to be introduced may have delayed applying for a divorce until it became easier, following the introduction of the new Act.

Sample Questions

A class of high school pupils took an intelligence test. Their mean score was 47 per cent. They then took vitamin tablets every day for a year. Their mean score on an intelligence test at the end of that period was 53 per cent.

Think about this example. What is it trying to prove? Does it do it?

Source A: The hidden cost of economic success
The current economic boom is good news for most people, but there is another side. The number of cases of depression, as measured by the number of prescriptions issued for antidepressant medication, has been rising steadily over the last few years.
High-achievers in business are currently receiving high rewards and a luxurious lifestyle which most other people envy. What they don't see is the stress and insecurity caused by the fear of losing this material prosperity.
‘Money doesn't buy happiness' is an old cliché, but we now have the evidence which proves it.

Source B: The cost of the recession
Figures released today reveal the psychological cost of the current economic recession. Rising unemployment and reductions in public expenditure have created a rise in mental health problems, especially depression. This is shown by the number of prescriptions for antidepressant medication, which rose last year by comparison with the previous twelve months.

Source A claims that economic prosperity causes depression, while Source B claims that economic hardship has the same effect. Do these claims contradict one another? Briefly explain your answer.

Source C: Research report
An analysis of the birth dates of players selected for school representative sports teams was carried out at a high school in England. Autumn-born children, the oldest within their year groups, were found to be overrepresented and summer-born pupils under-represented. In the selection of players for school sports teams, the autumn-born on average may be favoured due to initial advantages related to increased physical maturity.

Source D: Participation rates in English football leagues by dates of birth

Source C explains why the birthdate effect might occur in school sport, but this explanation does not apply to adult footballers, because they are not competing only against people from the same school year. Yet Source D shows that the birthdate effect does still influence players' chances of becoming a professional footballer.

Suggest two possible explanations for why the effect of birthdate on sporting ability seems not to disappear in adulthood.

Page updated

Report abuse