Critical Reading and Analysis

As a critical reader, you need to be able to develop your own ideas about what you read. Don’t just accept that everything you read is completely accurate or the only way of discussing an idea.

Here are some ways you can improve your critical reading skills:

Question as you read

Approach reading with a list of questions you can use to evaluate the information.

Questions you can ask include:

• Is the research objective? Is there any evidence of bias?
• What do the numbers that are quoted actually mean? Consider absolute numbers as percentages and vice versa (e.g. does 8 out of 10 sound as impressive as 80%?).
• Are the results meaningful and useful, or is it difficult to see how the results could be used or applied?
• Have other writers or researchers found similar patterns? Have the results of research been replicated?
• Did the study look at long-term effects or were only short-term results reported?
• Could there be other reasons for the findings other than those the researcher states?
• Are there any parts of the research process that were not well described or were not considered? Are there any omissions or gaps in the research process or thinking?

Look for links

Think about how the information you’re reading relates to:

• your existing knowledge, or
• other texts or authors.

This will help you to:

• identify similar ideas even when they sound different
• contextualise the information so you’re better able to form opinions on what you’ve read.

Evaluate the argument

To critically evaluate the content of an argument, you can:

• assess the strengths and weaknesses of the argument
• analyse the components of the argument (e.g. facts, ideas and claims)
• evaluate how the text has been organised
• examine the credibility of the evidence used to develop the arguments, how the author has analysed the material, and whether their conclusions are logical
• consider how the ideas can be applied in practice or what the problems may be.

Assess the source of the information

To evaluate the text’s appropriateness and quality as a research source consider the:

• author’s credentials, such as their institutional affiliations, educational background, past work and experience
• publication date, including whether there are more recent editions of the text, or other texts that provide more up-to-date information
• scope
• intended audience
• writing style and general quality.

Identify gaps and bias

To help identify gaps and bias in a text, identify information that is presented with little or no explanation or evidence. Look out for:

• generalisations
• assumptions
• opinions (even from experts)
• selective inclusion of evidence that only supports a particular point of view
• words that may indicate the author is overstating or making unjustified assumptions (e.g. plainly, obviously, undeniably, naturally, as you will agree, there is no doubt, it has to be admitted, clearly).

Part of becoming a successful critical reader is being able to take your thoughts about different texts and translate them into critical analysis in your writing.

Here are some examples of issues a critical reader would identity and the analysis they would make in their writing

Overgeneralisations and assumptions

Researchers often make simplifying assumptions when tackling a complex problem. This might provide some insight, but will also likely have limitations.

Example:

Students responded well to the teaching strategy, and measures of performance and motivation showed significant improvements. These improvements may in part be the results of small class sizes in the study and may not necessarily occur in larger mixed-ability classes.

Methodological limitations

Researchers may simplify the conditions under which an experiment occurs (compared to the real world) to be able to more easily investigate what’s going on.

Example:

While studies by Smith (1999), Brown (2000) and Green (2003) generally claim that women are superior to men at understanding body language, Wright (1998) has found no difference. Furthermore, methodological problems raise questions about the positive results. For example, Brown’s (2000) work looked solely at facial expressions, asking participants to make judgements by looking at photographs. Whether these findings would be valid in real-life situations was not explored.

Objectivity of research

Some research may be biased in its structure.

Example:

These findings suggest that property developers are primarily concerned with land use issues. However, this may be more a reflection of the questions asked in the survey than of developers’ primary concerns, as no open-ended questions were asked.

Limitations due to sample group

Some studies use a limited number or range of participants.

Examples:

The fact that 80% of students were satisfied with this mode of teaching is significant; however, it is important to note that only 20 of the 150 students in the class completed the questionnaire.

The study found that the average height of the adult male was 1.8 metres; however, this is questionable data as the sample for this study was taken at the local basketball court.

Results not replicated

If results have not been replicated in any other study, it indicates that the results are suggestive, rather than conclusive.

Example:

The use of band aids as an effective treatment for leg ulcers has the advantage of being cost-efficient; however, since this finding has not been replicated in other studies, practitioners may need to be cautious about applying this treatment until there is more evidence to support the finding.

Long-term effects unknown

If long-term effects have not been tested it may limit the applicability of the study.

Example:

Clients showed improvements in self-esteem after treatment; however, no long-term follow-up measures were taken.

Omissions

Look for topics that have not been discussed to determine whether this would limit the applicability of the results.

Example:

Brownlee and Whitely (2005) and Longley (2004) showed that economic buoyancy is related to debt-free spending; however, neither of these studies considered the impact of employment rates on economic buoyancy. It seems likely that employment rates could significantly affect this aspect of the economy.

Correlation versus causation

Just because two variables have a statistical relationship (correlation), it doesn’t necessarily mean that one causes the other.

Example:

Research suggests that dog owners are less likely to die of heart disease than people who don’t own a dog. Is this because owning a dog has health-giving benefits, or are people living with heart disease less likely to choose to get a dog?