1. Describe two sampling techniques.
(4 marks)
(a) Random (2m)
- Simple random sampling allows every member of the sampling frame and equal chance of being selected.
A list of random numbers may be used to select each name to avoid bias. The names are randomly
selected but may not necessarily be representative. Some groups might be over or under-represented.
- Stratified random sampling offers a possible solution to the problem of representativeness. This sampling
frame is divided into groups which reflect wider population for example age, gender, ethnic and class
group. For instance, if the sampling frame is based on men In Malaysia, the researcher might divide the
women into groups according to age or ethnicity and then draw a random sample from each of these
group.
(b) Non-Random (2m)
- Snowball sampling starts off with one member first before using his/her network to build up the sample
size and add more like-minded or like-situated individuals. This is usually helpful when researchers have
difficulties obtaining people for their samples. Since it is not random and relies on personal contacts, the
sample is likely to be low in representativeness.
- quota sampling is a non-random, non-probability sampling technique. It is often used for opinion polls and
market research.it is simpler, quicker, and cheaper than stratified random sampling. However, it is likely to
produce a representative sample.in quota sampling, people In same stratum do not have an equal chance of
being selected.
2. Explain two reasons why sociologists use different sampling techniques. (8 marks)
(a) 2 reasons why use random (4m)
-when using random sampling it relies on statistical probability to ensure the sample is representative of the
whole population of the whole population. Sociologist prefer representative samples that are representative
of the larger population because it allows or possible for them to make generalization for example the British
crime survey which is used to measure crime in England and wales, uses a nationally representative sample
of some 20,000 households.
- sociologist use this technique to avoid bias in selection as this technique are based on chance distribution
for example every name is given a number and then a list of random numbers is used to select a sample. This
technique also prevent the researcher from selecting a sample that provides a result that fits their theory or
supports their hypothesis. Therefore data produced are more valid
(c) 2 reasons why use non-random (4m)
When using non-random techniques, the sample is not considered as representative. This means the data
cannot be used to make generalisations about the wider population. Some sociologists may choose non-
random sampling because they are simply interested in the behaviour of that specific group itself, rather than
what it represents. For example, Venkatesh’s in-depth study of Chicago gangs does not claim to represent all
gangs in general.
Some researchers may also use non-random techniques because it is the only option in certain situations. The
researcher may not have enough information about the characteristics of the target population (sampling
frame not available/accessible) and therefore may need to rely on volunteers who are willing to participate
for instance. This means that the sample is likely to be biased because some groups may be left out from the
study, because they chose not to participate.
3. Explain one strength and one limitation of using longitudinal surveys as a research method. (6 marks)
(a) Advantage (3m)
A longitudinal study involves the researcher studying the same group of people over an extended period of
time. The researcher revisits the same sample at regular intervals to gather data. This means that the
researcher is able to build a picture of how their lives change over time when they compare their findings
from a previous stage of their lives. This is an advantage over the ‘snapshot’ picture that only shows what is
happening at a particular point in time. Thus, compared to cross-sectional surveys, longitudinal surveys offer
a more in-depth account of the group being studied. (high in validity)
Longtudinal studies allowed the researcher to identify and track personal and social changes over long
periods, revealing trends that would otherwise remained hidden .people don’t have to report on events
retrospectively so problems of faulty memory or reinterpretation of events reduced.
(b) Disadvantage (3m)
Longitudinal surveys are carried out at intervals (also known as ‘waves’) over a significant period of time
and can last for many years. This requires long-term commitment from both members of the sample and the
researcher(s) themselves. This may lead to practical difficulties in obtaining funding from sponsors who are
unwilling to commit to an expensive project whose results are unknown for many years. These studies tend
to be large scale as well which also adds to the high expenses involved in tracking development of the
sampling units. For example, the UK Millenium Cohort Study (MCS) involves a total of 18 552 families (18
827 children) with data collected at ages 9 months and 3, 5, 7 and 11 years.
Sample attrition, or the number of people who drop out from the original sample over time, is a major
limitation of these surveys. People drop out of the research for a range of reasons ; they may lose interest ,
move away without leaving contact details or may die. High levels of attrition can reduce the
representativeness of the sample over time- a problem that grows the longer the study lasts.
