Cochran’s Formula for Sample Size

Cochran’s formula is one of the most widely used methods for determining sample size in research. It helps researchers estimate the minimum number of respondents required to obtain reliable results from a large population. In this guide, we explain the Cochran sample size formula, define its components, and demonstrate how to calculate sample size step-by-step with examples.

What is Cochran’s Formula?

Cochran’s formula is a statistical method used to determine the sample size needed for a research study or survey. The formula was developed in 1977 by William G. Cochran, a well-known statistician who made important contributions to sampling theory. It helps researchers estimate the minimum number of respondents they need to obtain reliable results when studying a large population.

The formula is popular in various fields, including survey research, social sciences, public health, and market research. It allows researchers to calculate a sample size that represents the population with a chosen confidence level and margin of error. By using Cochran’s formula, researchers can collect data from a manageable number of participants while still producing statistically reliable estimates about the larger population.

Cochran’s Sample Size Formula

Cochran’s formula provides a mathematical way to determine the number of respondents needed in a study. It uses information about the confidence level, the estimated population proportion, and the acceptable margin of error to calculate the required sample size.

Cochran’s sample size formula is n₀ = Z².p(1-p)/e²

Where:

• n₀ is the minimum sample size you want to estimate, assuming an infinite population
• Z is a z-value that corresponds to a chosen confidence level. For instance, if you want to be 95% confident, then z = 1.96.
• p is an estimate of the population proportion that has the characteristics of interest
• 1-p is the proportion of the population without the characteristic of interest and is often denoted by the letter q
• e is the margin of error. It represents the maximum acceptable difference between the sample estimate and the true population value.

Note. Some authors replace 1−p with the letter q. In this case, the Cochran’s formula can also be written as n₀ = Z².pq/e². This does not change the formula since q = 1-p.

Calculating Sample Size Using Cochran’s Formula: An Example

A public health researcher wants to estimate the proportion of households that have access to clean drinking water in a large region. The researcher plans to conduct a survey with a 95% confidence level and a 4% margin of error. Because the true population proportion is unknown, the researcher assumes p = 0.5. Calculate the minimum sample size for the study.

Solution

From the question, we know that:

• Confidence level = 95%. The z-score for 95% confidence level is 1.96
• Margin of error, e = 0.04 (4%)
• p = 0.5

Substituting the values in the Cochran’s formula, we have:

n₀ = 1.96² * 0.5(1-0.5)/0.04²

= 0.9604/0.0016

=600.25

Rounding up to the nearest whole number, n₀ = 601

Therefore, the researcher would need to randomly sample at least 601 households to obtain reliable results that are representative of the target population. In other words, surveying 601 households will allow the researcher to estimate the proportion of households with access to clean drinking water with 95% confidence and a 4% margin of error. Using this sample size ensures that the survey results will provide a reliable estimate of the true population proportion.

Key Parameters in Sample Size Determination Using Cochran’s Formula

When applying Cochran’s formula, researchers must specify several important parameters. These parameters determine how large the sample size should be in order to obtain reliable results.

1. Confidence Level

The confidence level indicates how confident the researcher wants to be that the sample results represent the population. Each confidence level corresponds to a Z-value from the standard normal table.

Common examples include:

• 90% confidence level → Z = 1.645
• 95% confidence level → Z = 1.96
• 99% confidence level → Z = 2.576

Among these, the 95% confidence level is the most commonly used in research studies. However, this is not to say that your study should be tied to the 95% confidence level. You can also use any other confidence level, such as 96%, 97%, etc.

2. Population Proportion (p)

The parameter p represents the estimated proportion of the population that has the characteristic being studied. For example, it may represent the proportion of people who prefer a certain product or support a particular policy.

If the population proportion is unknown, researchers often use p = 0.5. This value produces the maximum possible sample size, which ensures the sample will be large enough for reliable estimates.

3. Margin of Error (e)

The margin of error represents the maximum difference allowed between the sample estimate and the true population value. It reflects the level of precision desired in the study results.

Common choices include:

• 5% (0.05) — commonly used in many surveys
• 3% (0.03) — used when higher precision is required

A smaller margin of error requires a larger sample size.

Cochran’s Formula for Finite Population

Cochran’s formula was originally developed for situations where the population is very large or effectively infinite. In such cases, the calculated sample size n_0​ provides a reliable estimate of the required number of respondents. However, when the population size is known and relatively small, the initial sample size may be larger than necessary.

To address this issue, researchers apply the finite population correction (FPC). This adjustment reduces the sample size because sampling a large proportion of a small population provides more information about the population. As a result, you need fewer observations to achieve the same level of precision.

The finite population correction formula is n =n_0​ /[1+(n_0​ -1)/N]

Where:

• n is the adjusted sample size
• n₀ is the initial sample size you get after applying Cochran’s formula
• N is the known population size

The adjustment is necessary because the original formula assumes that the population is extremely large. When the population is finite, the sampling variability decreases as a larger portion of the population is included in the sample. The finite population correction accounts for this effect and produces a more realistic sample size for smaller populations.

Example with Finite Population

Suppose a researcher wants to conduct a survey among university students at a particular college. The total number of students enrolled at the college is 2000. Using Cochran’s formula for large populations, the researcher initially calculated a required sample size of 277 respondents. Apply the finite population correction to determine the minimum sample size for this study.

Solution

From the question, we know that:

• N = 2000
• n₀ = 277

Because the population size is known and finite, the researcher must apply the finite population correction to adjust the sample size.

Applying the finite population correction formula and substituting the values, we have:

n =277/[1+(277 -1)/2000]

=277/[1+276/2000]

=277/1.138

= 243.41

Since the sample size must be a whole number, we round up to the nearest whole number.

Thus, the adjusted sample size, n = 244 respondents

This means the researcher only needs to survey 244 students instead of 277 when the population size is 2000. The finite population correction reduces the required sample size because the population is relatively small and known. Sampling 244 students is sufficient to estimate population characteristics with the same confidence level and margin of error.

When to Use Cochran’s Sample Size Formula

Cochran’s sample size formula is commonly used when researchers need to determine an appropriate sample size for studies that involve large populations. It is especially useful when collecting data from every member of the population is not practical. By using this formula, researchers can select a sample that provides reliable estimates of population characteristics.

Here are the common fields that use the formula for sample size determination

• Survey research – to determine how many respondents should be selected from a large population.
• Questionnaire studies – to estimate the number of participants needed to produce reliable responses.
• Social science research – when studying attitudes, behaviors, or opinions within a population.
• Market research – to estimate consumer preferences, product demand, or customer satisfaction.
• Public health studies – when estimating the prevalence of health conditions or behaviors in a population.

Limitations of Cochran’s Formula

While the Cochran’s formula produces a precise sample size, it has several limitations, which include:

• Assumes random sampling. The formula assumes that the sample will be selected using random sampling techniques. This means every member of the population has an equal chance of inclusion. Thus, if you decide to use non-random sampling techniques such as convenience sampling, the calculated sample size may not produce reliable results.
• Requires an estimate of the population proportion (p). The formula requires an estimate of the population proportion p. In many studies, this value is unknown, so researchers often use p = 0.5. While this approach ensures the largest and safest sample size, it may not always reflect the true population proportion.
• Less suitable for small populations without adjustment. The formula was originally designed for very large populations. When the population size is small or known, the formula may produce a sample size that is unnecessarily large. In such cases, researchers should apply the finite population correction to obtain a more appropriate sample size.

Frequently Asked Questions