The Kish Grid: What It Is, How to Use It in the Field, and When It Is the Right Choice

One of the most rigorous within-household selection methods ever developed. Here is a practical guide to how it actually works.

Picture this. You arrive at a household for a survey. There are four adults at home. Your survey targets one respondent per household. You need to pick one. Randomly.

How do you do that without introducing bias?

The Kish Grid was developed to answer exactly this question, and it remains one of the most widely used and methodologically rigorous approaches to within-household respondent selection in field research globally.

The Problem It Solves

Without a systematic selection method, interviewers tend to interview whoever is most available or most cooperative. That person is often a woman, often older, and often not representative of the full household membership. This introduces selection bias directly into your sample, even if your household selection was perfectly random.

The Kish Grid, first developed by survey methodologist Leslie Kish in 1949, solves this by giving every eligible person within the household a known probability of being selected, regardless of who answers the door.

How It Works: Step by Step

The grid itself is a pre-printed table that interviewers carry into the field. Each household visit is assigned a sequential number (household 1, household 2, household 3, and so on).

Here is how you apply it:

01.When you arrive at the household, identify all adults who are eligible for the survey based on your criteria. For example: all adults aged 18 and above who have lived in the household for at least six months.
02.Count the total number of eligible adults. Let us say there are four.
03.Assign each eligible adult a number, starting with the oldest. The oldest gets number 1, the next oldest gets number 2, and so on. This ordering gives younger adults a slightly higher probability of selection, compensating for their well-documented tendency to be less available during interview hours.
04.Look up your household number on the Kish Grid table. Run across to the column for your household count (in this case, 4 eligible adults). The grid tells you which numbered adult to interview.
05.That is the respondent. If they are not available, make up to three additional call-back attempts before treating it as a non-contact.

Kish Grid table showing how to randomly select a survey respondent within a household by cross-referencing the household sequence number against the count of eligible adults — with sequence 7 and 4 eligible people highlighted, selecting the 3rd oldest person for interview. — How the Kish Grid works in practice: cross-reference your household sequence number with the count of eligible adults to identify which person to interview. No discretion, no bias — just a number.

The grid itself is pre-configured so that across many households, the selection probabilities distribute evenly. No individual household needs to know about the grid. The randomness is built into the table design.

The Kish Grid does not ask you to trust the interviewer's judgment. It removes judgment from the equation entirely.

Known Limitations

Research comparing the Kish Grid to alternative methods, including studies published in the Italian Sociological Review, has shown that the method can produce underrepresentation of men and overrepresentation of older respondents in contexts where household structures differ significantly from the American households of the 1950s when the grid was developed. This is particularly relevant in countries with large household sizes or where young men are frequently absent during typical interview hours.

Response rates can be lower with the Kish Grid because the household listing step (enumerating all adults) can feel intrusive to respondents.
In households with many eligible adults, the listing step takes more time and increases overall fieldwork duration.
The method requires consistent training. Interviewers who do not follow the ordering protocol precisely undermine the probabilistic integrity of the selection.

When to Use It and When to Consider Alternatives

Use the Kish Grid when:

Your study requires rigorous probability sampling within households and you can justify the added complexity to funders or ethics reviewers.
The research is on a topic where gender or age representation matters significantly, such as health, labor, or welfare studies.
You are working in large-scale national surveys or multi-country studies where methodological standardization is required.

Consider simpler alternatives (last-birthday, next-birthday, or youngest adult methods) when:

Fieldwork timelines are tight and the added household listing step would significantly reduce the number of completed interviews.
Refusal rates are a primary concern and the listing question is likely to increase household-level non-response.
The research question does not require the same level of within-household representativeness.

The Kish Grid is a rigorous, well-validated tool. Like all tools, its value depends on how well it matches the demands of the project. Understanding when to use it, and when a simpler method is equally valid, is part of what distinguishes a thoughtful researcher from one who applies methods mechanically.

Find qualified field researchers with proven household survey experience.

ProjectBist profiles show methodology specialization and past project credentials.

Browse Field Researchersarrow_forward

Sources: Leslie Kish (1949) — Journal of the American Statistical Association; SAGE Encyclopedia of Social Science Research Methods; Italian Sociological Review 2021; Statistics How To; ResearchGate