Lesson 3.04

In this module, you’ll build on your understanding of how to use evidence effectively. You’ll learn how to draw logical inferences from the information provided, evaluate the strength of those inferences, and use multiple sources to form and support a well-reasoned judgment. By the end of this lesson, you’ll be able to think critically about evidence and make informed conclusions based on what you observe and read.

Key Terms

Inference - a conclusion drawn from one or more reasons or from evidence.
Predictions - Anticipating future data or outcomes
Recommendations - Proposing actions or policies that should be adopted based on the evidence

When you are asked to draw inferences, you might encounter several types of tasks:

Logical Inference

Extracting a sound conclusion directly from a given piece of reasoning.

Prediction

Forecasting what is likely to occur based on the evidence provided.

Recommendation

Identifying a specific action or policy that the evidence would support.

Assessing Inferences: Evaluating Validity and Usefulness

A crucial skill in critical thinking is assessing inferences that others have made or that you yourself might make from evidence. This involves judging the strength and reliability of conclusions.

Evaluating Support: You may be given data (e.g., tables, graphs) and asked to determine how well the data supports reported claims or how accurately a claim expresses the research findings.
Reliability: To evaluate reliability, you need to answer questions like, "Can you reliably draw the conclusion...?" or "Why can you not draw the conclusion...?". This assessment can sometimes be made solely by examining the source material, or it may require you to apply your own critical thinking.
Relevance and Significance: When asked to explain the relevance or significance of evidence, you are expected to draw an inference that strengthens, weakens, or otherwise affects the interpretation of a claim, question, or topic. Your judgment should reflect whether a meaningful inference can be drawn; if not, the evidence is not relevant or significant.
Usefulness of Evidence: The usefulness of evidence has two primary components: credibility and significance. If the evidence lacks credibility, or if you cannot derive a significant inference from it, then the source's usefulness is diminished.
Responding to Claims: Other assessment tasks include:
- Identifying factors that weaken the support evidence provides to a claim.
- Judging how effectively additional evidence responds to an existing claim.

Common Pitfalls and Challenges in Inferences

It's important to be aware of situations where inferences might be overstretched or misapplied.

Inadequate Information: Inferences sometimes extend beyond what the available evidence truly justifies.

Trends

Evidence from a single year is insufficient to establish a claim about trends over time.

Comparisons

Evidence about one country is not enough to justify a claim comparing it with other countries unless evidence for those other countries is also provided.

Vague Comparative Claims

Claims such as a product being "more effective" are meaningless unless the comparison point is clearly identified. For example, a claim that users are "seven times more likely to quit" smoking using a product is meaningless without a clear baseline for comparison (e.g., compared to what?). Even if a baseline like "by willpower alone" is given, its definition can be ambiguous, potentially allowing for biased interpretation.

Outliers

These are individual cases that lie outside a general trend. They are a normal part of any investigation and generally do not invalidate research findings, unless they are so numerous that they obscure the overall pattern. For instance, if research shows the average income of women in "Eastland" is 87% of men's, a highly paid woman does not disprove this overall finding.

Fallacy of Division

This is a logical error where one infers that something true for a whole category must also be true for a particular member of that category. For example, if "many" or "most" members of a group share a characteristic, it does not imply that "all" members do. It's crucial to remember that what applies to an average or a majority doesn't necessarily apply to every single individual.

In many analytical situations, you may need to identify gaps in evidence and propose what further information would resolve them. This process involves two stages:

Identifying a problem or weakness in the existing evidence.

2. Thinking of what further information would resolve that identified problem.

Ethical Criticisms and Objective Research

Academic research aims to be objective and publicly defensible. While readers might have personal ethical disagreements with the results of some academic research, such personal disapproval does not constitute a valid criticism of the research itself. This is because researchers typically describe what they have found to be the case, not what should be the case. For example, criticizing research on the relationship between birthdate and academic achievement due to personal beliefs about equality of opportunity would not be a valid critique of the research, as the researchers are simply reporting observations.

Sample Questions

Try to answer these questions yourself before checking your answers.

UK blogger
I don’t agree with the birthdate effect research. I was born in July and I have a first-class degree and a doctorate, but my son was born in November and he never went to college. In my view, it is not true that people who are born in the autumn or early winter are more intelligent than people born in the summer.

Think about this example. Is it a valid criticism?

Children as young as 11 admit buying alcohol underage
A survey of children between the ages of 11 and 17 has revealed that more than half of them admitted to buying alcohol from supermarkets before they were old enough to do so legally.

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

Research report
Research has found that overweight people travel farther in their seats before their seat belts engage in the pelvic area during a car crash, compared to normal-weight people. The delay in seat-belt activation is due to an excess of soft tissue in the abdomen, which prevents the belt from gripping the pelvis firmly. The upper body is held firmly, but the lower body is not. This can put intolerable strain on vital organs, leading to an increased risk of serious injury or death.

From the evidence in this report, it is possible to conclude that overweight people who travel in cars should lose weight. Suggest an alternative conclusion which could be drawn from the same evidence.

Source A: News report
Driving safely and wearing seat belts are two ways to reduce the chances of dying in a car accident. Losing weight might be another one. A recent study, led by Dr Thomas R Rice, of the Division of Environmental Health Sciences at the University of California at Berkeley, looked at whether there was any link between being overweight and the likelihood of dying from injuries suffered in car accidents.
'Findings from this study suggest that overweight people are more likely to die from traffic collision-related injuries than people of normal weight involved in the same sorts of collision.' the researchers wrote.

Source B: Research report

Source A claims that losing weight may be a good way 'to reduce the chances of dying in a car accident’. How well does the graph in Source 13 support this claim?

Dr Asif
As a doctor for more than forty years and a homoeopath for the last twenty, I know that homoeopathy is a most valuable tool and nine times out often my treatment of choice. One only has to see its dramatic effect in treatment of fevers, depression, anxiety and pain, to become a convert. It does work! Furthermore, no homoeopathic remedy has ever been withdrawn or banned because it was dangerous.

‘Homoeopathy should be used for all patients.’ Can this be reliably concluded from Dr Asif’s statement? Explain your answer.

Mrs. Courtney
My daughter was diagnosed with a thyroid complaint that would have meant a lifetime on medicine, with all sorts of possible side effects and at goodness knows what expense. This medicine would never have been a cure - it would just have made her less ill. After 6 months of homoeopathic treatment, her blood tests were completely normal and she was bouncing around like she never had before. Hers is not an isolated case, and certainly not due to the placebo effect, since she was dragged most unwillingly to see the homoeopath, and proclaimed that she didn't believe in any of it!

Dr Branchflower
Unfortunately you don't say what the condition was, nor who made the diagnosis, nor what other medication was being taken, nor what other medical conditions had been diagnosed... and a whole host of other important complicating factors. Not all thyroid ‘complaints’ require lifetime treatments. Many illnesses eventually come to a natural end without being treated, and this was probably one of them. It is certain that the homoeopathic treatment did absolutely nothing, since it contained no active medical ingredient.

How effectively does Dr Branchflower reply to Mrs. Courtney’s claims? Explain your answer.

Page updated

Report abuse