Content analysis as a research method
Content analysis is a research method used to measure the number of times a behaviour or event occurs within one or several forms of media. Content analysis can be done on many forms of media. For example:
- Song lyrics
- Computer games
Conducting content analysis
Researchers who wanted to see how much aggression occurred on television would use content analysis as a research method. They would have to take a number of steps to do this:
1) Decide what aggressive behaviour is.
2) Develop a list of behaviours or categories that could be measured as aggressive.
3) Decide on the sample they need to study (e.g. which television programmes or adverts, what times of day)
4) Tally (count) the times aggression occurred.
5) Assess the reliability of their results.
The table below is an example of a tally chart.
Once they have selected a sample of programmes to watch, the researcher uses a tally to record how many times they witnessed the aggressive acts on their list. A tally is a mark that shows a behaviour occurring once. These are added up to give the total number of aggressive acts so that they can decide how much aggression is on television.
This process is quite straightforward, but there are a few things you need to be aware of.
The choice of categories
The list of behaviours needs to be a good example of what is being measured. So, putting a category such as kissing as a measure of aggression would clearly not be appropriate. It could even be more subtle than this. For example, if a category of pushing was on the list, it might not be a good measure of aggression if you are watching a game of tag in a playground as it would appear as if children were more aggressive than they actually were.
Good content analysis depends on looking at a good sample of programmes, books or other forms of media to study. Even the time of day or type of programme can lead to a biased sample. For example, if you only look at television programmes after the watershed (9 p.m.) they are more likely to include violence than before the watershed. This may lead you to conclude that more television is violent than is the case. A good sample of television programmes should be at different times of the day, on different channels and different days of the week.
A poor sample means that the study findings will be unrepresentative. This means that the programmes in the sample cannot be said to be similar to all television programming or media forms being analysed.
Reliability of content analysis
When a researcher does a content analysis they might record tallies that other researchers would not, like the example of pushing in a playground. Another researcher might see this as playful and not give it a tally.
Each researcher has their own views, and this means that the results of a study might not be reliable. This could lead to researchers coming to different conclusions.
One way of overcoming this is by getting two or more researchers to do the same study. Everyone's results can be compared and only those that are agreed upon are used as a result. If more than two people agree on them then we can assume that most people will too.
content analysis: a research method used to measure the number of times something comes up in a book, newspaper article, television programme, etc.
tally: a single mark on a chart to show that a behaviour/category has been found during a content analysis.
unrepresentative: limited so that it might not apply to everyone
reliability: refers to whether findings from a study would be found again if the study was repeated.