# Qualitative and Quantitative sampling

June 9, 2018 | Author: Nosipho Dlamini | Category: Documents

#### Description

Cluster sampling

A type of random sample that uses multiple stages and is often used to cove large geographic areas. In which aggregated units are randomly selected and then samples are drawn from the sampled aggregated units or clusters.
According to Singleton et al., (1988, p. 147) "Unlike stratified sampling , which draws cases from each stratum, cluster sampling draws cases only from those clusters selected for the sample". E.g. we have 9 suburbs in a city which is part of the investigation. Out of those, 3 are homogeneous with regards to the age of the residents. Then only one out of the three is selected.
NB: We use it to solved two problems i). The lack of a good sampling frame for a dispersed population. ii). The high cost of reaching a sampled element.

Simple RANDOM SAMPLING
Each individual case in the population theoretically has an equal chance of being selected for the sample. The following may serve as an example:
i). List all the street numbers of all respondents. ii). Decide on the sample size e.g. 50. No duplicate numbers should be included.
iii). Use a random table, and starting from any chosen point in the table, taking e.g. three digits at a time, reading vertically or horizontally, sets of three digits are chosen and written down.

Probability sampling
It is based on randomisation
A probability sample is one in which each sample (or other sampling unit) in the population has the same known probability of being selected (Seaberg, 1988; Kirk, 1999).
In probability sampling, the odds of selecting a particular individual are known and can be calculated (Gravetter & Forzano, 2003).

Probability vs non-probability sampling
Simple random sampling
Systematic
Stratified random sampling
Cluster sampling
Panel sampling
Accidental sampling
Purposive sampling
Quota sampling
snowball sampling
Target sampling
How do we ensure representativeness?
The answer offered by most methodologists is that random sampling is the only technique available that will ensure that an optimal chance of drawing a sample that is representative of the population from which it was drawn.
this then leads us to consider the two kinds of sampling available to researchers, probability sampling, which is based on randomisation ; and non-probability sampling which does not implement randomisation.
Cont…
When we talk about representativeness, it means the sample must have exactly the same characteristics as the population relevant to the research in question . E.g. if gender and socio-economic class are variables (characteristics), relevant to the research, a representative sample will have approximately the same proportions of men and women and middle-class and working-class individuals as the population.
Representativeness of samples
Representativeness is a very important aspect of sampling (Kerlinger, 1986).
Representativeness is important when we want to generalise from the sample to the larger population., that is we study a sample in order to draw conclusion about the population from which the sample was drawn (Reid & Smith, 1981).
BUT…What exactly does the above mean?

Reasons for the use of Samples
The major reason for sampling is feasibility (Smith & Reid, 1981; Sarantakos, 2000).
The observation or study of a phenomenon in its entirety would be tedious and time-consuming. It would also produce gigantic amounts of data, which by implication would be difficult to process, analyse and interpret (Arkava & Lane, 1983).

Systematic sampling
Here the first case is selected randomly, preferably from a random table (Hoinville, 1978).
All subsequent cases are selected according to a particular interval e.g. each fifth or tenth case on a list of names depending on the percentage sample needed (Van der Walt, 1984).
Effort is saved by this method. According to Babbie (1990), Systematic sampling is considered as having a higher value than simple random sampling.
The END
NOTE: Please read the recommended (Neuman, 2014) text to beef-up these slides.
Ngiyabonga, Ndatenda, Thank you!

Target sampling

Watters and Biernacki (1989, p. 420) define a target sample as "a purposeful , systematic method by which controlled lists of specified populations within geographic districts are developed and detailed plans are designed to recruit adequate numbers of cases within each of the targets"
Panel sampling
A panel sample means that a fixed panel of persons is selected from the population of persons involved in a particular issue. Naturally the panel must be proportionately representative. Of the relevant population.
E.g. OK Stores wants to select a panel from their clients to test a certain product, this method of sampling can be used. If 80% of the clients of the product are women, four women for each man should be included in the panel.
Non-probability sampling

In non-probability sampling, as the name suggest, the odds of selecting a particular individual are not known because the researcher does not know the population size, or the members of the population (Gravetter & Forzano, 2003). This section focuses on;
Accidental sampling
Purposive sampling
Quota sampling
Snowball sampling
Target sampling

Accidental sampling

Any case which happens to cross the researcher's path and has anything to do with the phenomenon is included in the sample until the desired number is obtained (Singleton et al., 1988).
Thea so-called man-in- the-street interviews conducted by television teams are a case in point.
Other authors call this type of sampling a convenient, availability or haphazard sample (Nachmias & Nacmias, 1981; McBurney, 2001).
"We simply reach out and take the cases that are at hand, continuing the process until the sample reaches a designated size" (Judd et al., 1981).

Purposive sampling
This type of sample is based entirely on the judgment of the researcher in that a sample is composed of the elements that contain the most characteristic , representative or typical of the population.
See Neuman 2014, p.273.
Quota sampling

Quota samples are often used by market researchers. Their main purpose is to draw a sample that is a close replica of the population.
E.g. in social research, to sample first year UKZN students, a researcher using Quota sampling will take appropriate %s of White, India, and Black African students. Each race will provide a relevant quota.
See Neuman 2014, p.273.

Snowball sampling

It involves approaching a single case which is involved in the phenomenon to be investigated in order to get information on other similar persons.
In turn, this person is requested to identify other people who could make up the sample.
Similar to a snowball that gathers snowflakes as it rolls, so does the researcher who starts off with a single participant, and gathers more participants as each refers him/her to the next one.
Stratified random sampling

It is suitable for heterogeneous populations because the small subgroups percentage-wise can be ensured (Van der Walt, 1984).
Stratification consists of the universe (potential subjects) divided into a number of strata which are mutually exclusive, and the members of which are homogeneous with regard to some characteristics such as gender, home language or age (Glicken, 2003).
The number of desired persons is proportionally selected within each of the different strata.
What is sampling?
Selection of elements of the population considered for actual inclusion in the study (Arkava & Lane, 1983, p.27). Alternatively, a sample is a small portion of a total set of objects, events or persons which together comprise the subject of our study (Seaberg, 1988, 240). We study the sample in order to understand the population from which it was drawn.
Qualitative and Quantitative sampling
BY Shumba, k.
Dept. of Psychology, SAHS
H.C, UKZN

Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
7/31/2017

#
Click to edit Master title style
Click to edit Master subtitle style
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
Click to edit Master text styles
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
7/31/2017

#
7/31/2017

#
Click to edit Master title style
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
Click to edit Master text styles
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
Click to edit Master text styles
7/31/2017

#
"
"
Click to edit Master title style
Click to edit Master text styles
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
Click to edit Master text styles
Click to edit Master text styles
Click to edit Master text styles
Click to edit Master text styles
Click to edit Master text styles
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles
7/31/2017

#
Click to edit Master title style
Click to edit Master text styles