Observational design

Question 1

What is inter-observer reliability?

Answer 1

It is recommended that researchers do not conduct observational studies alone as they may miss important details or may only notice events that confirm their opinions or hypothesis which introduces bias. Inter-observer reliability measures the degree of agreement between different people observing or assessing the same thing. The data from the two observers is compared to check for: consistency and reliability and this is called inter-observer reliability

Question 2

How is inter-observer reliability done ?

Answer 2

• Observers should familiarise themselves with the behavioural categories to be used
• They then observe the same behaviour at the same time, perhaps as part of a small scale-pilot study
• Observers should compare the data they have recorded and discuss any differences in interpretations
• Finally, observers should analyse the data from the study. Inter-observer reliability is calculated by correlating each pair of observations made and an overall figure is produced

Question 3

What are the two types of ways to record data?

Answer 3

1. unstructured observation
2. structured observation

Question 4

What is an unstructured observation?

Answer 4

• Writing down everything they see (qualitative data)
• Producing accounts of behaviour that is rich in detail
• This type of observation may be appropriate when observations are small in scale and evolve few participants
  however there is to much going on in a single observation so instead….

Question 5

What is a structured observation?

Answer 5

When the target behaviour is broken down into a set of behavioural categories (behaviour checklist)
For example, if the target behaviour was affection it could be broken down into hugging, kissing, holding hands etc…

• Behaviours are categorised
• Quantitative data is used
• Easier to compare and analyse
• Good for larger studies with lots of participants

Question 6

What are behavioural categories?

Answer 6

A behavioural category is a specific type of behaviour which is defined before the study takes place. It allows researchers to focus their investigation on a specific behaviour in order to gather the most valid and reliable data.

Question 7

What is a behaviour checklist/coding system?

Answer 7

This is the method used to develop behaviour categories. Using a coding system means that a code is invented to represent easy categories of behaviour

Question 8

How do structured observations and sampling link?

Answer 8

For a structured observation a researcher must used a systematic way of sampling their observations

Question 9

What are the two types of sampling for a structured observation?

Answer 9

1. Event sampling
2. Time sampling

Question 10

What is event sampling?

Answer 10

Involves counting the number of times a particular behaviour (the event) occurs in an individual or group

Question 10

What is time sampling?

Answer 10

Involves recording behaviour within a pre-established time frame. The observer decides in advance that observation will take place only during specified time periods (e.g. 10 minutes every hour, 3 hours per days) and records the occurrence of the specified behaviour during that period only.

Question 11

Adv and disadv of event sampling?

Answer 11

+ Good for infrequent behaviours
- Observer may overlook important details in complex events

Question 12

Adv and disadv of time sampling?

Answer 12

+ Reducing the number of observations needed to be made
- May miss behaviours in-between times
- May be unrepresentative

Question 13

Adv and disadv of structured observations?

Answer 13

+ Makes the recording of data easier and more systematic.
+ Likely to produce quantitative data which makes it easier to analyse and compare.
+ Smaller risk of observer bias.
- Results aren’t as in depth and detailed.
- Some behaviours the researcher records may not be important.

Question 13

Adv and disadv of unstructured observations?

Answer 13

+ More detailed results.
- There may be a greater risk of observer bias (no behavioural categories)
- Produces qualitative data which is more difficult to analyse.