Pilot Studies & Design

Test your knowledge with free interactive questions on Seneca — used by over 10 million students.

Pilot Studies

Many research studies begin with a small-scale investigation known as a pilot study. These help researchers to test out their methodology and make minor changes.

Formation of pilot study

A pilot study involves running the planned methodology but with a much smaller number of participants.
The participants will often be a small opportunity sample such as classmates or colleagues.

Function

A pilot study allows for modification of methodology where necessary.
- For example, researchers may decide that an experimental memory task is too easy, and needs to be made more difficult to avoid a ceiling effect (where everyone gets a high score, skewing the data).

Advantages of pilot studies

A pilot study allows a researcher to decide whether or not it will be worthwhile to conduct a planned study on a larger scale.
A pilot study also provides the researchers with practice of running the study before the full data gathering begins, allowing all aspects of the study to go more smoothly.

Experimental Design

When designing an experiment, the researcher must make a decision - whether to divide participants into groups for separate conditions, or to have a single group complete more than one condition.

Basic design

Every experiment has at least two experimental conditions (or one experimental condition plus one control condition).
This allows researchers to make a comparison between different values of the independent variable.

Repeated measures

When allocating participants to conditions, one option is to have every participant complete every condition. This is known as a repeated measures design.

Independent groups

Alternatively, the researcher may split participants into groups to complete the different conditions of the research study.
This is known as an independent groups design.
Each participant only takes part in a single condition.

Evaluation of repeated measures

A repeated measures design minimises participant variables because the same people are being studied in every condition.
But it means that participants may guess the hypothesis of the study, and they may get better or worse at a task as the experiment goes on (‘order effects’).

Evaluation of independent groups

An independent groups design suffers from participant variables because different people are being studied in the different conditions.
But it avoids order effects and makes it harder for participants to perceive the hypothesis of the study.

Matched pairs

A matched pairs design tries to get the best of both worlds: participants are in different groups, but they are matched up on age or abilities to minimise the role of participant variables.
This is done by matching participants into pairs, and then randomly allocating a member of each pair to each condition.

Observational Design

When designing an observation study, the researcher must consider how to gain data from the situation. The main choices relate to the use of behavioural categories, event sampling, and time sampling.

Prior planning

Observation studies involve prior planning to consider what behaviour is likely to be observed.

Observation schedule

To simplify things for the observer, different behaviours are combined into categories (for example, punching and kicking could both be described as ‘aggression’).
These behavioural categories are listed on an observation schedule.

Event sampling

The observer also has a choice between event sampling and time sampling:
Event sampling involves recording an event every time it happens over a period of time.
- For example, if a researcher was observing a child in a school playground, they may record every example of aggression over a 20-minute period (based on the chosen behavioural categories),
- They may also record what else was happening at the same time.

Time sampling

Time sampling involves recording the most prominent behaviour at many different points in time.
The points in time could be selected randomly or systematically.
- For example, an observer may observe a child in a school playground once every two minutes, and record what they are doing at that point.

Limitations

Behavioural categories, event sampling, and time sampling all simplify the process of observing behaviour and make it more reliable.
- But they can also simplify and reduce the level of detail, and so reduce the validity of the observation.