Chapter 11: Sampling
Judgement is generally based on our experience, or those of others.
Sampling is a technical accounting device to rationalise the collection of
information, to choose an appropriate way in which to restrict the set of
objects, persons or events from which the actual information will be drawn.
Basically, it's when a small group of people are chosen from a big group of
people.
The population is the entire set of people or objects for which the researcher is
trying to determine some characteristic. The sample is a subset, or small
group, that the researcher will perform their research on in order to draw a
conclusion on the entire population.
Each individual, group or object that is within the sample is called a unit of
analysis. Population parameters are values that relate to the population as a
whole, e.g the average age of the participants. Statistics, or sample statistics,
are when the population parameters (or values) are connected from the sample
to the general population. So, statistics are just the estimate of population
parameters.
Quantitative research usually aims at testing a hypothesis on a
representative sample – a sample that reflects the demographics of the
general population. This is deductive and accurate. Qualitative research aims
to understand phenomena, and is therefore inductive and holistic.
Quantitative sampling theory is the study of the relationship between a
population and the samples which are drawn from it. Statistical inference is
the process of making generalisations about the population from the findings
from the sample. In other words, this is when we draw conclusions on the
population based on the results of the sample. When researchers present their
findings, they usually state what is probable (likely to happen) rather than
certain.
The main advantages of sampling (taking data from a small sample instead of
taking data from the entire population) are as follows:
1. Gathering data using a sample is less time consuming because you don't
have to interview every single person that belongs to that group, e.g. It is
easier to choose 200 pre-school teachers than to interview 5000 across the
country.
2. Sampling is less costly. This is also because sampling takes less time than
interviewing everyone, and time is money.
3. Sampling is often the only practical method of data collection. It is better to
test the lifespan of 10 light bulbs, rather than all the light bulbs in the country
as this is not practical.
4. It is a practical way of collecting data when the population is infinite or very
big.
Sampling must satisfy certain criteria for qualitative and quantitative research:
1. Quantitative research: The sample is selected before data collection
happens. It is considered a good sample if it represents the general population
well in terms of diversity, and gives results that tell you something about the
general population.
2. Qualitative research: The sample is often selected before and during the
research. It is considered a good sample if it allows all possibilities of the
, research topic or phenomenon to be investigated. Data saturation is when
the researcher has enough information, and claims that collecting more data
would not increase their knowledge or tell them something they don't already
know.
In both types of research, one has to have a well-defined population and an
adequate sample.
The target population is the set of elements that the research focuses upon.
In other words, they are the subjects of the research. The population will be
well-defined if the target population can be described accurately, and the
researchers can make a list of the different factors that the population bring to
the research. An operational definition must be given and there must be certain
criteria that the population must adhere to to belong to this operational
definition.
For quantitative sampling, the sample must have all the same characteristics
as the general population. Thus, a representative sample is used. The first
way to ensure that you are using a representative sample is by using a correct
sampling frame. This is the list of all units from which the sample could be
taken, e.g. a list of people or groups that could potentially be used. An
inadequate sampling frame would discard some part of the population, e.g. If a
researcher selected their participants using the phone book, he is discarding all
those who don't own a phone. Therefore, the phone book is an inadequate
sampling frame.
The sampling theory distinguishes between probability sampling and non-
probability sampling:
Probability (or random) sampling is when you know the chance of each
person or unit of the population being selected. In other words, it is when the
probability of choosing each person or element can be determined. E.g. If you
were doing random sampling and selecting 10% of the population, then you
know that each person or element has a 1 in 10 chance of being chosen.
Non-probability sampling is when the probability of choosing each person or
element within the population is unknown. Here, it is not possible to determine
the chance of being selected, because there is a certain criteria or reason for
being selected. This makes it difficult to see how well the sample represents
the general population, thus making it difficult to draw conclusions about the
population from the sample. However, this also has some advantages. For
example, this is the only practical solution when the population lists are not
available. It is also always cheaper, faster, and often good enough for
homogenous populations (populations where most people are similar regarding
a certain trait). Most disadvantages of non-probability sampling could be
overcome just by using a larger sample size.
The most common sampling procedures are explained below, and fall under the
categories of probability and non-probability sampling.
1. Probability sampling (for quantitative research) includes:
a) Simple random sampling: In this case, random is simply the criteria for being
selected, it does not mean “accidentally chosen”. An example of this would be
selecting the first ten students to enter the lecture hall. You did not actively
choose them according to any individual criteria, but you did find a way to
select them in a random way. This method provides each person with an equal
chance at being selected. A method that is often used is the lottery method –
each member of a population is given a number, and then a certain amount of
numbers are chosen, and the people who represent those numbers are chosen
Judgement is generally based on our experience, or those of others.
Sampling is a technical accounting device to rationalise the collection of
information, to choose an appropriate way in which to restrict the set of
objects, persons or events from which the actual information will be drawn.
Basically, it's when a small group of people are chosen from a big group of
people.
The population is the entire set of people or objects for which the researcher is
trying to determine some characteristic. The sample is a subset, or small
group, that the researcher will perform their research on in order to draw a
conclusion on the entire population.
Each individual, group or object that is within the sample is called a unit of
analysis. Population parameters are values that relate to the population as a
whole, e.g the average age of the participants. Statistics, or sample statistics,
are when the population parameters (or values) are connected from the sample
to the general population. So, statistics are just the estimate of population
parameters.
Quantitative research usually aims at testing a hypothesis on a
representative sample – a sample that reflects the demographics of the
general population. This is deductive and accurate. Qualitative research aims
to understand phenomena, and is therefore inductive and holistic.
Quantitative sampling theory is the study of the relationship between a
population and the samples which are drawn from it. Statistical inference is
the process of making generalisations about the population from the findings
from the sample. In other words, this is when we draw conclusions on the
population based on the results of the sample. When researchers present their
findings, they usually state what is probable (likely to happen) rather than
certain.
The main advantages of sampling (taking data from a small sample instead of
taking data from the entire population) are as follows:
1. Gathering data using a sample is less time consuming because you don't
have to interview every single person that belongs to that group, e.g. It is
easier to choose 200 pre-school teachers than to interview 5000 across the
country.
2. Sampling is less costly. This is also because sampling takes less time than
interviewing everyone, and time is money.
3. Sampling is often the only practical method of data collection. It is better to
test the lifespan of 10 light bulbs, rather than all the light bulbs in the country
as this is not practical.
4. It is a practical way of collecting data when the population is infinite or very
big.
Sampling must satisfy certain criteria for qualitative and quantitative research:
1. Quantitative research: The sample is selected before data collection
happens. It is considered a good sample if it represents the general population
well in terms of diversity, and gives results that tell you something about the
general population.
2. Qualitative research: The sample is often selected before and during the
research. It is considered a good sample if it allows all possibilities of the
, research topic or phenomenon to be investigated. Data saturation is when
the researcher has enough information, and claims that collecting more data
would not increase their knowledge or tell them something they don't already
know.
In both types of research, one has to have a well-defined population and an
adequate sample.
The target population is the set of elements that the research focuses upon.
In other words, they are the subjects of the research. The population will be
well-defined if the target population can be described accurately, and the
researchers can make a list of the different factors that the population bring to
the research. An operational definition must be given and there must be certain
criteria that the population must adhere to to belong to this operational
definition.
For quantitative sampling, the sample must have all the same characteristics
as the general population. Thus, a representative sample is used. The first
way to ensure that you are using a representative sample is by using a correct
sampling frame. This is the list of all units from which the sample could be
taken, e.g. a list of people or groups that could potentially be used. An
inadequate sampling frame would discard some part of the population, e.g. If a
researcher selected their participants using the phone book, he is discarding all
those who don't own a phone. Therefore, the phone book is an inadequate
sampling frame.
The sampling theory distinguishes between probability sampling and non-
probability sampling:
Probability (or random) sampling is when you know the chance of each
person or unit of the population being selected. In other words, it is when the
probability of choosing each person or element can be determined. E.g. If you
were doing random sampling and selecting 10% of the population, then you
know that each person or element has a 1 in 10 chance of being chosen.
Non-probability sampling is when the probability of choosing each person or
element within the population is unknown. Here, it is not possible to determine
the chance of being selected, because there is a certain criteria or reason for
being selected. This makes it difficult to see how well the sample represents
the general population, thus making it difficult to draw conclusions about the
population from the sample. However, this also has some advantages. For
example, this is the only practical solution when the population lists are not
available. It is also always cheaper, faster, and often good enough for
homogenous populations (populations where most people are similar regarding
a certain trait). Most disadvantages of non-probability sampling could be
overcome just by using a larger sample size.
The most common sampling procedures are explained below, and fall under the
categories of probability and non-probability sampling.
1. Probability sampling (for quantitative research) includes:
a) Simple random sampling: In this case, random is simply the criteria for being
selected, it does not mean “accidentally chosen”. An example of this would be
selecting the first ten students to enter the lecture hall. You did not actively
choose them according to any individual criteria, but you did find a way to
select them in a random way. This method provides each person with an equal
chance at being selected. A method that is often used is the lottery method –
each member of a population is given a number, and then a certain amount of
numbers are chosen, and the people who represent those numbers are chosen
