Download Benchmark- Data Collection Design and more Exams Business Research Methods for Managers in PDF only on Docsity!
Sampling Methods and
Measurement Levels in Data
Collection Design
Benchmark- Data Collection Design
Probability and Non-Probability Sampling
Research and data sampling can be conducted in many different ways. The two most commonly used methods are probability and non-probability sampling.
Probability Sampling
Probability sampling is the process of selecting elements from a population in order to reflect the total population within the statistics (Curan, 2019). This representative method allows for an equal chance of being chosen as a sample to be given to each individual within a population. These samples will be able to represent the population as a whole without actually having to sample the total population. There are four main methods to probability sampling:
Simple Random Sampling : This is done by creating a master list of everyone within a population and then randomly selecting subjects. This method is seen as the easiest, but can become time-consuming with larger sample areas.
Stratified Random Sampling : This places individuals of a population within certain classifications based on age, race, education, etc. The goal of this type of sampling is to create a well-rounded and diverse sample, more precise than simple random sampling.
Systematic Random Sampling : This is based on the choosing of a number that will be a continuous separation between subjects. This type of sample is useful, but not as precise as it would be if the researcher used a computer generator to randomize subjects.
Cluster Random Sampling : This is used for larger populations that are difficult to randomize. This type of sample is based on boundaries that are placed by the researcher in order to make randomizing subjects easier for a larger population.
Non-Probability Sampling
Non-probability sampling methods do not give all individuals within a population a fair chance of being chosen. Subjects under this method are
usually chosen based on availability, or their judgment of the researcher (Explorable.com, 2009). The types of non-probability sampling include:
Convenience Sampling : This is based on the accessibility of subjects to the researcher. This method is seen as the easiest and least time- consuming.
Consecutive Sampling : This is similar to convenience sampling; however, it seeks to include all accessible subjects, rather than just the ones that are easy to recruit.
Quota Sampling : This is based on classifications of individuals based on race, education, gender, age, and socioeconomic status. Under the chosen quota trait, the researcher ensures that there is equal representation of subjects.
Judgmental Sampling : This is a more purposeful method and is based on the fitness of an individual with respects to the research being conducted in comparison to other possible subjects.
Snowball Sampling : This is used in smaller population sizes. This method uses one initial subject and asks them to recommend other potential subjects that would be fit for the researcher.
Non-probability methods are ideal to use when randomization is difficult within larger populations.
Levels of Measurement
Levels of measurement are used to describe the relationship between values that are assigned to specific variables. It is useful to know the measurement value of a sample because it helps the researcher understand how to interpret the values and what statistical analysis to use (Trochim, 2006). Levels of measurement are broken up into four categories:
Nominal Measurement : This uses numbers and symbols to classify variables, rather than put them into order. For example, if the data represents people from a gender group, males would be represented as M and females would be represented as F in a data set.
Ordinal Measurement : This shows an ordered relationship between the variables of a data set.
Interval Measurement : This specifies the distance between each variable on a scale, while also placing them into order.
Ratio Measurement : In this measurement, variables have an equal interval distance between them on a scale. This measurement also allows for the value of zero to be used, unlike any other level of measurement.
