Understanding Variables, Sampling Methods, and Sample Size in Research

 Lecture-06


In this lecture

Variable
Population
Sample
Sampling
Sample size


Variable

In the research field, a variable is any characteristic, attribute, or quantity that can be measured, observed, or manipulated. Variables are essential in research as they help to define relationships, test hypotheses, and draw conclusions. For example; height, weight, income, age etc. The main focus of the scientific study is to analyse the functional relationship of the variables. A variable is a quantity which can vary from one individual to another. The quantity which can vary from person to person.

Variable is a property that taken on different value”, Kerlinger

Types of Variables in Research:

Continuous Variable:

continuous variable is a type of quantitative variable that can take on any numerical value within a specific range. It can be measured to any level of precision and is not limited to whole numbers.


Example:

·         Height of individuals (e.g., 160.5 cm, 175.2 cm).

·         Temperature (e.g., 36.7°C, 98.6°F).

·         Time taken to complete a task (e.g., 12.3 seconds, 15.8 seconds).

 

Discrete Variable:

discrete variable is a type of quantitative variable that can only take on specific, distinct values with gaps between them. It often represents counts or whole numbers.


Example:

·         Number of students in a class (e.g., 25, 30, 35).

·         Number of cars in a parking lot (e.g., 10, 15, 20).

·         Number of children in a family (e.g., 1, 2, 3).

 

Dependent Variable or Criterion Variable:

The dependent variable (also called the criterion variable) is the outcome or result that is influenced by another variable. It is the variable that researchers measure to assess the effect of the independent variable.

Example:

·         In a study examining the effect of study time on exam scores, exam scores are the dependent variable because they depend on the amount of study time.

·         In a drug trial, the recovery rate of patients is the dependent variable, as it depends on the type of drug administered.

 

Independent Variable or Experimental Variable:

The independent variable (also called the experimental variable) is the variable that is manipulated or changed by the researcher to observe its effect on the dependent variable. It is the presumed cause in a cause-and-effect relationship.

Example:

·         In the study, on study time and exam scores, study time is the independent variable because it is manipulated to see its effect on exam scores.

·         In a drug trial, the type of drug administered is the independent variable, as it is changed to observe its impact on patient recovery.

 

Controlled Variable:

controlled variable is a variable that is kept constant or unchanged throughout an experiment to ensure that it does not influence the outcome. It is used to isolate the effect of the independent variable on the dependent variable. By controlling these variables, researchers can accurately determine whether changes in the dependent variable are solely due to the manipulation of the independent variable.

Example: 

·        In an experiment to test the effect of fertilizer on plant growth, factors like amount of watersunlight, and soil type are controlled variables. These are kept constant to ensure that any changes in plant growth are solely due to the fertilizer (independent variable) and no other factors.

Confounding Variable:

confounding variable is an external factor that influences both the independent and dependent variables, making it difficult to determine the true relationship between them. Its effect is often confused with the effect of the independent variable. Confounding variables can be of two types: intervening variables and extraneous variables.

Example: 

·        In a study examining the relationship between exercise (independent variable) and weight loss (dependent variable), diet could be a confounding variable. If participants change their eating habits while exercising, it becomes unclear whether weight loss is due to exercise or diet.

 Intervening Variable:

An intervening variable is an abstract, unobservable factor that indirectly affects the relationship between the independent and dependent variables. It explains the process or mechanism through which the independent variable influences the dependent variable. Intervening variables are often related to internal states, such as emotions or psychological factors, and are difficult to measure directly.

Example: 

·        In a study on the impact of study time (independent variable) on exam performance (dependent variable), motivation could be an intervening variable. Motivation influences how effectively study time translates into better exam performance, but it is not directly observable.

Extraneous Variable:

An extraneous variable is an independent variable that is not the primary focus of the study but may still affect the dependent variable. If not controlled, extraneous variables can introduce errors into the experiment, making it difficult to determine the true effect of the independent variable. Researchers often control extraneous variables through randomization or by holding them constant.

Example: 

·        In a study investigating the relationship between self-concept (independent variable) and social studies achievement (dependent variable), intelligence could be an extraneous variable. Intelligence may influence social studies achievement, but since it is not the focus of the study, it must be controlled to avoid distorting the results.

Organismic Variable:

An organismic variable is a characteristic of the participants that cannot be manipulated or changed by the researcher. These variables are inherent to the individuals being studied, such as age, gender, intelligence, or socioeconomic status. While they cannot be altered, they can be measured and accounted for in the research design.

Example: 

·        In a study comparing the academic performance of boys and girls (organismic variable), differences in performance might be influenced by factors like intelligencemotivation, or socioeconomic background. These factors, rather than gender itself, could explain any observed differences. Since gender cannot be manipulated, it is treated as an organismic variable.

Population:

  • Population refers to the entire set of individuals, items, or data that share a common characteristic and are of interest to the researcher. It is the complete group about which the researcher wants to draw conclusions.
  • Example: If a researcher is studying the academic performance of college students in a country, the population would include all college students in that country.

Sample:

  • sample is a subset of the population that is selected for the actual study. It is a smaller, manageable group that represents the population. The goal of sampling is to draw conclusions about the population based on the analysis of the sample.
  • Good and Hatt, “A sample as the name implies, is a smaller representation of a larger whole.
  • Example: From the population of all college students in a country, a researcher might select 500 students from different colleges as a sample.

Sampling:

  • Sampling is the process of selecting a sample from the population. It involves choosing a group of individuals, items, or data points that accurately represent the larger population. Sampling is crucial because studying the entire population is often impractical due to time, cost, or logistical constraints.
  • David S. Fox, “In the social sciences, it is not possible to collect data from every respondent relevant to our study but only from some fractional part of the respondents. The process of selecting the fractional part is called sampling.”
  • Example: A researcher might use random sampling to select 500 college students from a list of all college students in the country.

Types of Sampling Methods:

Sampling methods are broadly categorized into two types:

·        Probability Sampling

·        Non-Probability Sampling

Probability Sampling

Probability sampling is a sampling technique in which every member of the population has a known, non-zero chance of being selected in the sample. This method relies on random selection, ensuring that the sample is unbiased and representative of the population.

Types of Probability Sampling:

1.     Random Sampling

2.     Systematic Sampling

3.     Stratified Sampling

4.     Multistage Sampling

5.     Cluster Sampling

Random Sampling

·        Random sampling (also referred to as Simple random sampling) is the most straightforward probability sampling strategy.

·        It is the most popular method for choosing a sample among population for a wide range of purposes.

·        It is one in which each element of the population has an equal and independent chance of being included in the sample i.e. a sample selected by randomization method is known as simple random sample and this technique is simple randomizing.

Randomization is done by using the following techniques:

·         Tossing a coin

·         Throwing a dice

·         Lottery method

·         Blind folded method

 

Systematic Sampling

Systematic sampling is an efficient and improved method over simple random sampling. It requires a complete list of the population, arranged in a systematic order (e.g., alphabetical, numerical, or any logical sequence). The process involves selecting every nth individual from the list after a random start, where:

·         Sample size (n) = Desired number of individuals in the sample.

·         Population size (N) = Total number of individuals in the population.

·         Sampling interval (k) = N/n (every kth individual is selected).

Example:

If the population size (N) is 1,000 and the sample size (n) is 100, the sampling interval (k) is 10. A random start is chosen between 1 and 10, and every 10th individual is selected from the list.


Stratified Sampling

Stratified sampling is a probability sampling technique where the population is divided into distinct subgroups, called strata, based on specific characteristics (e.g., age, gender, income, education level). A random sample is then drawn from each stratum. 


Multistage Sampling

Multistage sampling is a complex form of cluster sampling where the population is divided into multiple stages before selecting the final sample. It is used when dealing with large and geographically dispersed populations, making data collection more feasible and cost-effective.

How It Works:

1.     First Stage: The population is divided into large groups or clusters.

2.     Second Stage: A random selection of clusters is made.

3.     Third Stage: Further sub-clusters are selected within the chosen clusters.

4.     Final Stage: Individuals or elements from the sub-clusters are randomly selected for data collection.

Example

Studying Educational Performance in a Country

Suppose a researcher wants to study the academic performance of high school students across an entire country.

1.     First Stage: The country is divided into regions (North, South, East, and   West).

2.     Second Stage: A random selection of some regions (e.g., North and South) is made.

3.     Third Stage: Within each selected region, a random selection of cities is made.

4.     Fourth Stage: From each selected city, a few schools are randomly chosen.

5.     Final Stage: A sample of students is randomly selected from the chosen schools.

Cluster Sampling

Cluster sampling is a probability sampling technique where the population is divided into clusters (groups) that are naturally occurring or logically defined (e.g., schools, neighborhoods, cities). Instead of sampling individuals, entire clusters are randomly selected, and all individuals within the chosen clusters are included in the sample. This method is useful when the population is large, spread out, or difficult to access.

Steps in Cluster Sampling:

1.     Divide the Population into Clusters:

·        Split the population into smaller, mutually exclusive groups (clusters) that are heterogeneous within themselves but homogeneous between each other.

 

2.     Randomly Select Clusters:

·        Use random sampling to select a specific number of clusters.

 

3.     Include All Individuals in Selected Clusters:

·        All members of the chosen clusters are included in the sample.

Example:

A researcher wants to study the academic performance of high school students in a country with 10,000 schools. Instead of sampling individual students, the researcher divides the population into clusters based on schools. Suppose the researcher selects 50 schools (clusters) randomly from the total list of schools. All students within the selected 50 schools are included in the sample.

Non-Probability Sampling

Non-probability sampling is a sampling technique where individuals are selected based on non-random criteria. In this method, not every individual in the population has an equal chance of being selected. It is commonly used in qualitative research, exploratory studies, and when random sampling is impractical or unnecessary.

Types of Non-Probability Sampling

·        Convenience Sampling

·        Judgment Sampling

·        Quota Sampling

·        Snowball Sampling

Convenience Sampling

·        Participants are chosen based on their availability and willingness to participate.

  • Example: Surveying people in a shopping mall or using social media polls.

Judgment Sampling

Judgmental sampling, also known as purposive sampling, is a type of non-probability sampling where the researcher selects participants based on their knowledge, expertise, or specific characteristics relevant to the study. Instead of selecting randomly, the researcher uses their judgment to choose individuals who best fit the study’s purpose. This method relies on the researcher’s discretion to choose participants who are most relevant or representative of the research objectives.

Example:

·        A researcher is studying the impact of a new teaching method on student performance. Using judgmental sampling, the researcher selects 10 experienced teachers who have been recognized for their innovative teaching practices. These teachers are chosen because their expertise and experience are deemed most relevant to the study’s objectives.

Quota Sampling

Quota sampling is a non-probability sampling technique where the population is divided into subgroups (called strata) based on specific characteristics (e.g., age, gender, income, education level). The researcher then selects a predetermined number of individuals (quota) from each subgroup to ensure that the sample reflects the population’s diversity. Unlike stratified sampling, quota sampling does not involve random selection within subgroups.

Steps in Quota Sampling:

1.     Define the Population and Subgroups: Identify the population and divide it into subgroups based on key characteristics (e.g., age, gender).

2.     Determine Quotas: Decide how many individuals to include from each subgroup, often proportional to their representation in the population.

3.     Select Participants: Use convenience or judgment to select individuals from each subgroup until the quota is met.

4.     Collect Data: Gather information from the selected participants.

Example:

A researcher wants to conduct a survey on consumer preferences for a new product. The population is divided into subgroups based on age and gender:

·         Age Groups: 18–25, 26–40, 41–60, 61+

·         Gender: Male, Female

The researcher decides to survey 200 people and sets quotas proportional to the population distribution:

·         18–25: 50 (25 males, 25 females)

·         26–40: 60 (30 males, 30 females)

·         41–60: 70 (35 males, 35 females)

·         61+: 20 (10 males, 10 females)

Snowball Sampling

Snowball sampling is a non-probability sampling technique used in research, particularly in social sciences, where existing study subjects recruit future subjects from among their acquaintances. This method is especially useful when studying hard-to-reach or hidden populations, such as individuals with rare diseases, members of marginalized communities, or people involved in illegal activities.

Here’s how snowball sampling typically works:

1.     Initial Recruitment: The researcher identifies and recruits a small number of initial participants who meet the criteria for the study. These participants are often referred to as "seeds."

2.     Referral Process: After the initial participants have been interviewed or surveyed, they are asked to refer others they know who also meet the study criteria. This process relies on the social networks of the initial participants.

3.     Chain Reaction: The new participants are then asked to refer others, creating a chain reaction or "snowball" effect. This process continues until the desired sample size is reached or until no new participants can be found.

Example:

A researcher is studying the experiences of undocumented immigrants in a particular city. Due to the sensitive nature of their status, undocumented immigrants are difficult to identify and reach through traditional sampling methods.

1.     Initial Contact: The researcher might start by contacting a local advocacy group that works with undocumented immigrants. Through this group, they identify and recruit a few individuals who are willing to participate in the study.

2.     Referral: These initial participants are then asked if they know other undocumented immigrants who might be willing to participate. They provide contact information or introduce the researcher to their acquaintances.

3.     Expansion: The researcher contacts these new participants, who in turn refer more individuals. This process continues until the researcher has gathered enough data or reached a saturation point where no new information is being obtained.

 

Types of snowball sampling

1.     Linear Snowball Sampling

·         In this type, the sampling process follows a single chain of referrals. The researcher starts with one or a few initial participants (seeds), who then refer others, and this process continues in a linear fashion.

2.     Exponential Non-Discriminative Snowball Sampling

·         In this approach, each participant is asked to refer multiple new participants, leading to an exponential growth in the sample size. The researcher does not discriminate or select among the referrals; all referred individuals are included in the sample.

3.     Exponential Discriminative Snowball Sampling

·         Similar to exponential non-discriminative sampling, but the researcher selectively includes only certain referrals based on specific criteria. Only participants who meet the study's requirements are included. This allows for more control over the sample composition.


Sample size

·        Unknown population

·        n = required sample size

·        Zα/2​ = critical value from the standard normal distribution corresponding to the desired confidence level (e.g., 1.96 for 95% confidence)

·        E = margin of error 

·        p = estimated population proportion (use 0.5 if unknown for maximum variability)


·        Known population

·        p = sample proportion,

·        q = 1 – p

·        N = size of population

·        n = size of sample

Example

What should be the size of the sample if a simple random sample from a population of 4000 items is to be drawn to estimate the per cent defective within 2 per cent of the true value with 95.5 per cent probability? What would be the size of the sample if the population is assumed to be infinite in the given case?

[N = 4000;

e = .02 (since the estimate should be within 2% of true value);

z = 2.005 (as per table of area under normal curve for the given confidence level of 95.5%).

p = 0.5]

 

Build Struct : Exploring Insights of Civil Engineering

Post a Comment

Previous Post Next Post