What is the Kish Selection Procedure?

There are many different ways to select someone to interview for a survey. These selection procedures can be very simple, such as calling a home and asking whoever picks up, or more complex. An example of a complex method of selecting people for surveys is known as the Kish Selection Method. It was developed in 1949, and is designed to select respondents “within the unit.”

What is “Within Unit?”

Most people remember being called on the phone by a research or polling company at some point in their life. What they may not remember is that often the person asks to speak with the head of the household, or the person in charge of making decisions.

If a sample was truly random, then the person picking up the phone would be the one to speak to. But these researchers are looking for someone “within the unit” – someone within the household that they want to speak to more than the person that picked up the phone.

The Kish Selection Method has a similar style to it, except it’s a bit more complicated. It’s completed as follows:

  • First, everyone that fits the eligibility criteria, such as being over the age of 18, is gathered together. If there is only one person, that person is the primary survey respondent.
  • The interviewer collects the age and gender of everyone that is eligible for the survey.
  • Those in the household are placed in a selection grid.
  • The researchers then choose a respondent based on their place in the grid, using some form of random selection so that the person selected to take the survey didn’t have some commonalities that caused them to introduce bias into the data.

The Kish Selection Method is essentially designed to ensure that the person most likely to answer a less invasive survey (one where the age and gender isn’t asked, and the person that answers the phone can answer the questions) isn’t different in a specific way from the rest of the population, such as if women are more likely to answer phones than men, or something similar.

Unfortunately, the Kish Selection Method has since lost popularity, because the response rate when forced to answer personal information is very low, especially these days when people are more averse to filling out surveys. But it was once a more effective way of ensuring that you’re not introducing bias when you select randomly via the phone, because there is a chance that those answering the phone and answering the questions aren’t necessarily “random.”

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>