Random, at -
At random is selecting people or objects and dividing them into groups without making a pre-selection on the characteristics of the people or objects.
In a research the groups should be basically the same before they can be compared on another aspect. If they are not, the outcome of the research might be unclear due to interfering aspects of the respondents or objects.
For example, in a study into the effect of gym on the mobility of the elderly, a scientist collected data from two groups: people who took gym classes and people who did not. After analysing the data, he concluded that the people who attended gym classes were more mobile. However, the elderly who attended the gym classes, lived in villages and they often walked to the supermarket or took a bike when they wanted to visit friends. The elderly in the other group lived in a city and did not visit their friends that often, and when they did, they took a taxi. So taking gym classes was not the reason for the effect found, but living in a city or town was.
In this example, the place to live was a confounder. If it is known that certain characteristics of the respondents will disrupt the results of the study, it is wise to apply matching. The groups are then equal on these characteristics. However, it is not always possible to create equal groups for all characteristics. It is now best to make fairly large groups. This ensures that disturbing features are randomly distributed among the groups. Now there may still be interference, but this interference will be the same for all groups.
Related topics to Random
- Experimental variables
Deepen your knowledge and read our manual about ...
- How to create research designs and conceptual models
- Representativity Nine ways to draw a sample