BUACC5931 - Report on Research and Statistical Methods for Business

Q1 Sample Size
Sample size plays a vital role in the determination of the features of a population. A magnitude
of proportion of the population is referred to as sample size. The decision of coming up with the
exact sample size is essential in the collection of accurate information, this is according to
(Kühberger et al, 2014). As a result therefore, for the collected data to maintain its reliability, the
accuracy of the sample size has to be considered. By applying the rightful formula for the
determination of the sample size i.e.
Sample size =
population distribution percentage pick
( margin error %
confidence level score )2
From the population of 69,000 Belgian bank workers, the size of the sample from this population
by applying the stated formula was supposed to be 383 at the 95% confidence level with a
marginal error pf 0.05. The preferred percentage pick in the calculation of the sample size was
50%. This percentage was chosen because it is conservative when it comes to calculation of a
vast sample size. By taking all the factors into account i.e. percentage pick, marginal error and
confidence level, out of the population of 69,000 bankers, 383 is the recommended sample size
that ought to have been used by the two research institutions. Bearing this in mind therefore, we
can conclude that choosing to work with the sample size of 15,000 bank workers being that it is
beyond the recommended size, it is a large sample size. Working with large sample size
normally has some advantages and disadvantages as will be discussed.
Advantages of sample size
Large sample size is importance in minimizing the Margie of error which further helps in
boosting the accuracy of the obtained results from the sample, this is one among other

advantages of large sample size. This increases the precision by which the population parameter
fall within the range of the calculated point estimator (Clearly et al, 2014). Wide coverage by the
large sample size (15,000 participants) in a population helps in acquiring more accurate
information from the participants concerning the subject under study (i.e. stress in this case) as
compared to when the smaller sample size would have been used. Furthermore, big and
increased sample size raises the representativeness of the individuals in the population which
further takes care and caters for the outliers that would be present in the population, this is
according to (Belli et al, 2014).
Disadvantages of sample size
Collecting and covering a larger fraction of the population will require high expenses to be
involved and incurred hence large samples are costly, this is according to (Goodman et al, 2013).
In this case, reaching the targeted number of bank workers of 15,000 by the research institutions
will make the Union of Belgian Banks to spend much in order for target to be achieved. On the
same, being that the fraction is relatively large and the participants to participate in the data
collection process are not in same geographical location since the banks are spread across the
country, achieving the targeted sample size will be time consuming.
Factors to consider when choosing a sample size
Prior information obtained about the topic under study will be important in determining the
sample size being that prior point estimates such as means and variances act as the reference to
deal with variation could arise in the groups (Button et al, 2013). Cost is another factor that can
be used to determine the size of the sample to be used in a survey. Depending on the risk value

involved in the values that need to be collected, if the risk needs to be high, then small sample
size can be used and if the risk involved is to be low, then that calls for large sample size.
Q2 Sampling methods
Sampling method is the process by which group characteristics are obtained from the population
under study. Stratified sampling method was used by the research institutions in the selection of
15,000 bank workers who were to participate in the process. According to (Ye et al, 2013),
stratified sampling method reduces sampling errors when used. The population is first divided
into small groups called strata that are distributed to ensure that every element of a population is
represented thereafter have characteristics from each stratum selected by simple random
sampling in order to reduce selection bias. This method ensures that the population is highly
represented in the sample. One of the problems with this method is the difficulty to identify
means to be applied in subdividing the population into subgroups, this makes this method
unpopular and rarely used by the researchers. Furthermore, a lot of time is involved in the
determination of strata which will later require the selection of the sample from the available
strata by simple random sampling method (Acharya et al, 2013). In our case, the research
institutions were first to find the banks within the country, categorize the workers according their
bank institutions which will form strata where workers will now be selected from to form a
sample. Simple random sampling is involved in the selection of individuals from the banks to
provide for equal chances to be in the sample that represent the population. I hereby recommend
for the increased number of strata in order to improve the effectiveness of stratified sampling
method that will increase the representativeness of the population as the marginal error is
minimized.
Q3 Research Design