What Is A Good Sample Size For Research

One of the near common questions I get asked by people doing surveys in international evolution is "how big should my sample size exist?". While there are many sample size calculators and statistical guides available, those who never did statistics at university (or have forgotten it all) may find them intimidating or hard to use.

If this sounds similar you, then keep reading. This guide will explain how to choose a sample size for a basic survey without any of the complicated formulas. For more easy rules of thumb regarding sample sizes for other situations, I highly recommend Sample size: A rough guide by Ronán Conroy and The Survey Research Handbook past Pamela Alreck and Robert Settle.

This commodity is a short introduction to the topic for a more than in-depth coverage of the topic consider enrolling in the free online form offered by University of Florida.

This communication is for:

Basic surveys such as feedback forms, needs assessments, opinion surveys, etc. conducted as part of a program.
Surveys that utilise random sampling.

This communication is Non for:

Research studies conducted by universities, research firms, etc.
Complex or very large surveys, such as national household surveys.
Surveys to compare between an intervention and control group or earlier and subsequently a program (for this situation Sample size: A rough guide).
Surveys that utilize not-random sampling, or a special type of sampling such equally cluster or stratified sampling (for these situations encounter Sample size: A rough guide and the United nations guidelines on household surveys).
Surveys where you plan to use fancy statistics to analyse the results, such as multivariate analysis (if you know how to practise such fancy statistics then yous should already know how to cull a sample size).

The minimum sample size is 100

Most statisticians agree that the minimum sample size to go any kind of meaningful result is 100. If your population is less than 100 then you really need to survey all of them.

A good maximum sample size is unremarkably ten% every bit long every bit information technology does not exceed 1000

A good maximum sample size is normally effectually 10% of the population, as long equally this does not exceed 1000. For example, in a population of 5000, ten% would be 500. In a population of 200,000, 10% would be 20,000. This exceeds 1000, so in this instance the maximum would be 1000.

Even in a population of 200,000, sampling 1000 people will normally give a adequately accurate consequence. Sampling more 1000 people won't add much to the accuracy given the actress time and money it would toll.

Choose a number betwixt the minimum and maximum depending on the situation

Suppose that you lot want to survey students at a school which has 6000 pupils enrolled. The minimum sample would be 100. This would requite you a crude, but all the same useful, idea almost their opinions. The maximum sample would be 600, which would requite y'all a fairly accurate idea about their opinions.

Choose a number closer to theminimum if:

You have limited time and coin.
Yous but demand a rough estimate of the results.
You lot don't programme to carve up the sample into different groups during the analysis, or you only plan to use a few large subgroups (e.chiliad. males / females).
Yous call back most people will give like answers.
The decisions that volition be made based on the results do non have meaning consequences.

Choose a number closer to themaximum if:

You accept the time and coin to do it.
It is very of import to become authentic results.
You plan to divide the sample into many unlike groups during the assay (e.g. unlike age groups, socio-economic levels, etc).
Yous think people are likely to give very different answers.
The decisions that will be made based on the results of the survey are important, expensive or have serious consequences.

In exercise most people normally desire the results to exist as accurate as possible, so the limiting cistron is commonly time and money. In the instance in a higher place, if y'all had the fourth dimension and money to survey all 600 students then that volition give you a fairly accurate consequence. If you don't accept enough fourth dimension or coin and then simply choose the largest number that you can manage, as long as it'southward more than 100.

If you would like to learn more about Survey Information Collection consider taking the free course offered past University of Michigan and Academy of Maryland. Enroll here.

If you lot want to be a chip more scientific then utilize this tabular array

While the previous rules of thumb are perfectly acceptable for almost basic surveys, sometimes you need to audio more "scientific" in lodge to be taken seriously. In that case you can utilise the following table. Simply choose the column that most closely matches your population size. Then choose the row that matches the level of mistake you're willing to have in the results.

Sample size table

You will see on this tabular array that the smallest samples are all the same around 100, and the biggest sample (for a population of more than 5000) is withal around 1000. The same general principles utilise as before – if you plan to divide the results into lots of sub-groups, or the decisions to exist fabricated are very important, you should pick a bigger sample.

Annotation: This table can only be used for basic surveys to measure out what proportion of the population have a item feature (e.g. what proportion of farmers are using fertiliser, what proportion of women believe myths near family planning, etc). It tin't be used if y'all are trying to compare two groups (e.yard. control versus intervention) or ii points in time (eastward.g. baseline and endline surveys). See Sample size: A crude guide for other tables that can exist used in these cases.

Relax and finish worrying about the formulas

It'southward a muddy piddling secret among statisticians that sample size formulas often require you to have information in advance that you don't commonly have. For example, you typically need to know (in numerical terms) how much the answers in the survey are likely to vary between individuals (if you knew that in advance then you wouldn't be doing a survey!).

So even though it'south theoretically possible to summate a sample size using a formula, in many cases experts still end upwardly relying rules of thumb plus a good deal of common sense and pragmatism. That means you shouldn't worry too much if yous tin can't use fancy maths to cull your sample size – you're in skillful company.

Once you've called a sample size, don't forget to write proficient survey questions,blueprint the survey course properly and pre-test and pilot your questionnaire.

Photo by James Cridland