Research Methods: Sampling Techniques Quiz

Sampling Techniques & Sample Size Estimation

• Sampling is the process of selecting a group of items from a population.
• A census involves a complete enumeration of all items in the population.
• A census offers the highest accuracy but requires significant time, money, and energy.
• Sampling is more practical for large populations because it reduces costs, time, and possible bias.
• Statistical inference is based on the process of sampling to draw conclusions about the population.

Population & Sample

• Population: The collection of all responses, measurements, or counts of interest to the researcher.
• Sample: A subset of the population.
• A population could be composed of individuals, families, groups, organizations, events, objects, or items. Examples include: presidents, professors, books, students, and red blood cells (RBCs).

Sampling: Process and Elements

• Sampling involves these three elements: selecting the sample, collecting information, and making inferences about the population.

Importance of Sampling

• Sampling allows researchers to gather information about large populations.
• It is more economical than studying the entire population.
• It reduces the time required for data collection.
• It can improve accuracy in certain situations.
• Sampling is often necessary due to the large sizes of many populations when it isn't feasible to study the entire population.

Statistical Terms

• Target Population: The population the researcher wants to generalize his results to.
• Sampling frame: A list of all the sampling units from which the sample is drawn.
• Sampling unit: The people or items that have been sampled; the target population is the collection of sampling units.
• Sampling fraction: The ratio of the sample size to the population size.
• Sample criteria: These include the criteria for inclusion and exclusion criteria needed for the study.
• Sampling scheme: The method of selecting sampling units from the sampling frame.

Steps of the Sampling Process

• Identify the target population.
• Identify the accessible population to be sampled (N).
• Determine the sample size needed (n) and the sample criteria.
• Select the sampling technique.
• Implement the plan.

Sampling Methods

• Probability samples: Each member of the population has a known chance (probability) of being selected in the study.
• Non-probability samples: The chances (probability) of selecting members from the population are unknown.
• Examples of probability sampling methods include: simple random sampling, systematic sampling, stratified sampling, and cluster sampling.
• Examples of non-probability sampling methods include: convenience sampling, judgmental sampling, quota sampling, and snowball sampling.

Probability Samples: Advantages

• Probability samples allow generalization of results.
• They ensure representativeness and precision.

Simple Random Sampling

• Procedure involves defining the population, choosing the sample size, creating a sampling frame by numbering all units, and randomly selecting units.
• The lottery method, table of random numbers and computer-generated random numbers can be used.

Simple Random Sampling: Advantages and Disadvantages

• Advantages: known and equal chance of selection, simple, easy process.
• Disadvantages: requires the complete sampling frame, not great for diverse groups.

Systematic Sampling

• Procedure involves selecting a suitable sampling frame, numbering each element, determining the sampling interval (I = N/n), selecting a random number (s) between 1 and k, and using the formula s, s+k, s+2k,..., s+(n-1)k to select the sample.

Systematic Sampling: Advantages and Disadvantages

• Advantages: known and equal chance of selection, less expensive and faster.
• Disadvantages: possible small loss in sampling precision

Stratified Sampling

• Principle: The population is partitioned into subgroups (strata) and elements are randomly selected from each stratum.
• Procedure includes identifying and defining the population, determining the desired sample size, identifying the variable.
• Randomly selecting elements from each stratum.

Stratified Sampling: Advantages and Disadvantages

• Advantages: more accurate.
• Disadvantages: more complex, sampling plan requiring different sample sizes for each stratum.

Cluster Sampling

• Principle: The population is divided into subpopulations (clusters), a random sample of clusters is selected.
• Procedure includes identifying and defining the population, determining the desired sample size, identifying and defining a logical cluster, listing all clusters, estimating the average number of population members per cluster, determining the number of clusters needed by dividing sample size by estimated size of a cluster.
• Randomly selecting the needed number of clusters, including all population members in each selected cluster.

Cluster Sampling: Advantages and Disadvantages

• Advantages: time and cost-efficient, particularly for samples that are geographically spread, high external validity.
• Disadvantages: more complex to plan, cluster members are more likely to be alike.

Multi-Stage Sampling

• Principle: Sampling is repeated at several levels, e.g. Country → State → City → School.

Non-probability Samples

• These sampling methods do not guarantee a known chance of selection for each member of the population.
• Examples include convenience sampling, judgmental sampling, quota sampling, and snowball sampling.

Convenience Sampling

• A sample is selected from population elements that are readily available.
• Often, respondents are selected because they happen to be in the right place at the right time.
• This type of sampling is also known as accidental sampling.

Judgmental/Purposive Sampling

• Participants are intentionally selected based on their characteristics, experiences, or other criteria.

Quota Sampling

• The population is divided into strata (e.g., male and female students).
• A proportion from each stratum is selected, but not randomly.

Snowball Sampling

• Participants are recruited through referrals from initial respondents (chain referral, friend-to-friend).

Non-probability Samples: Advantages and Disadvantages

• Advantages: simple, easy to use, helpful for pilot studies, short duration, cost-effectiveness.
• Disadvantages: highly vulnerable to selection bias, unclear generalizability, high sampling error.

Sampling Errors

• Random errors (sampling error): The difference between the sample and the population from which the sample is drawn. The size can be measured, expressed as the standard error of mean/proportion.
• Standard error depends on the sample size and distribution of the measured variable.
• Systematic errors (bias): Inaccurate response (info bias) or selection bias.

Sample Size Calculation

• Why calculate sample size? Small samples can lead to inaccurate conclusions.
• Undersized studies can expose participants to potential harm without providing meaningful conclusions.
• Oversized studies waste resources and possibly expose participants unnecessarily.
• Factors considered when determining minimum sample size include: significance or confidence level, power (1-β), effect size, variability, and the alternative study hypothesis (one-tail, two-tail).

Factors Determining Sample Size

• Significance level (alpha).
• Power (1 - beta).
• Effect size.
• Variability.
• Desired Confidence Level.

Significance Level (Alpha Level)

• It is the probability of making a Type I error (incorrectly rejecting a true null hypothesis).
• Most studies set alpha at 0.05, meaning they accept a 5% chance of making a Type I error.
• Higher confidence levels (e.g., 99%) will result in a larger required sample size.

Power (1 – Beta Level)

• It is the probability of detecting a true difference if one exists in the population.
• A higher power level (e.g. 0.80) means that the study is more likely to detect a difference if one exists in the population.

Effect Size

• The minimum difference a study should detect.
• Clinically significant differences.
• Generally, smaller effect sizes require larger samples.

Variability

• Measures how spread out the data is in a population.

Alternative Study Hypothesis

• One-tailed easier to identify differences (so smaller samples needed).
• Two-tailed more difficult to identify, so a larger sample size is required

Tools for Calculating Sample Size

• Formulas (specific to each study design).
• Nomograms.
• Prepared tables.
• Computer software.

Descriptive Equation for a Single Proportion

• Used in calculating sample size for proportions. The equation needs specific values for Z, P, Q, confidence level, and degree of precision.