Understanding Secondary Data in Research

This assignment delves into the different categories of secondary data employed in research studies. It examines government statistics, freely available quantitative data, and information published in professional journals as reliable sources for conducting research. The assignment also highlights the importance of utilizing secondary data to complement primary data collection and ensure representativeness in research findings.

Running head: STATISTICS AND BUSINESS RESEARCH
Statistics and Business Research
Name of the Student
Name of the University
Author note

1STATISTICS AND BUSINESS RESEARCH
Table of Contents
Answer 1....................................................................................................................................2
Sample size.............................................................................................................................2
Answer 2....................................................................................................................................3
Sampling method...................................................................................................................3
Answer 3....................................................................................................................................5
Research design......................................................................................................................5
Answer 4....................................................................................................................................6
Procedure of data collection...................................................................................................6
Answer 5....................................................................................................................................8
Secondary data.......................................................................................................................8
References................................................................................................................................10

2STATISTICS AND BUSINESS RESEARCH
Answer 1
Sample size
. Samples are drawn in order to make some precision about the population. It is not
possible to collect information from all the sixty nine thousand bank employees. Keeping the
time and budget constraint in mind only a small percentage (21%) of total population is taken
for consideration. The advantage of a small sample is the ease of calculation. Smaller the
sample is lesser will be the complexity in calculation. It is costly to arrange a primary survey
over a large number of samples. Setting of questionnaire and sending them to the respondents
is subject to high cost (Marshall et al., 2013). Apart from cost factor, small samples help the
researchers to complete the research within small time framework.
Besides the advantages, a sample of small size has some shortcomings. With a small
sample size the variability of the estimated statistics increases. The variability is reflected in
the measure of variance. Smaller the sample greater is the variance. A large variance also
increases the possibility of obtaining a biased estimate. One type of bias arises from small
sample is variable response bias (Malterud, Siersma & Guassora, 2016). When a considerably
small sample is selected then there is high chance of getting similar kind of responses. The
response may come from a group that strongly in support or oppose of something.
In times of selecting sample size, certain things need to be keep in mind. There should
clarity about the expectation of sample. The objective of sampling estimation helps to
determine sampling size. A probability statement is needed to connect the population
precision (Johnson & Wichern, 2014). with the sample size. Depending on that statement, the
sample size is selected. Estimation about the sampling cost is of greater importance in
determining sample size. In times of very crucial decision, making less variability is desirable

3STATISTICS AND BUSINESS RESEARCH
and hence a large sample is selected for valuable decision-making. The selected sample size
should pass the test of practical applicability. In order to obtain a close estimate to population
parameter a large sample size is always desirable. However, given the limited budget this is
not always possible to select a large sample. The way out for conducting a precise estimation
with a small sample is estimation and quantification of associated risk with the selected
sample.
Answer 2
Sampling method
The current method of sampling is called simple random sampling. In simple random
sampling, all observations in the population have an equal chance of being selected in the
sample. Here in order to find the relationship between job characteristics and job satisfaction
random samples are taken from each of the participating banks. Therefore, all the members
working in these banks have an equal chance of being included in the sample. Hence, the
method of sampling is similar to that with simple random sampling.
Advantages
 Simple random sampling requires minimum prior information about the intended
population. This works as the biggest advantage random sampling method.
 As the sample is selected from the whole population, randomly no classification is
needed at all. Therefore, there is no chance for occurrence of classification errors
(Koyuncu & Kadilar, 2016).
 It is the most suitable form of sampling method for drawing any inference from the
population. It gives bias free estimates of population parameters. Hence, random

4STATISTICS AND BUSINESS RESEARCH
sampling is considered as the best representative of population. In addition,
assessment of sampling error is also easy in this method.
 It is extremely easy to conduct simple random sampling. It does not make any
discrimination at times of selecting samples.
Disadvantages
 In times when there are widely varied subpopulation then there larger possibility of
contain sampling errors with simple random sampling. Here, stratified sampling is
more appropriate. In case population units are highly dispersed, it becomes extremely
difficult to collect sample using simple random sampling (Levy & Lemeshow, 2013).
 Simple random sampling cannot be used in case where population observations are
not homogenous.
 The sampling method has least scope of using prior knowledge of population
concerned.
Suggestion
Before going for random sampling technique, information about the population is
needed. If it is observed that there are specific subgroups in the population then stratified
sampling, techniques should be used. In the presence of clusters in population, clustering
samples is more appropriate (Lewis, 2015). In times of collecting samples, any types of
biasness should be avoided. Neither too large nor too big samples are desirable. A good
sampling method includes selection of samples of appropriate size.

5STATISTICS AND BUSINESS RESEARCH
Answer 3
Research design
In a cross, sectional research design analysis is over a specific context. A group of
observation or social phenomenon are studied here in terms of selecting a particular sample.
This is a widely used method of research design. It allows the researcher to conduct a
comprehensive analysis within a particular research framework. Longitudinal research design
involves a continuous analysis of the population. The research here continued for a much
longer period as compared to cross sectional research. In longitudinal research method, at
every phase of analysis same sample is used (Campbell & Stanley, 2015)
The most significant advantage of cross section research design over the longitudinal
research is, it is less time consuming in nature. It does not require overtime analysis of same
sample., Instead, the complete study is made only once. On the other hand, research method
under longitudinal design is subject to a long time. Cross sectional study is able to analyse
different variables at a same point of time. Therefore, it is particularly suitable for research
projects that have time bound. Here, the sample is selected only for once and decision has
been taken based on analysis of the samples. There are least amount of risk with cross
sectional arrangement. It is able to estimate the prevalence of interested outcome more
accurately as the sample is drawn considering the entire population (Palinkas et al., 2015).
Research design under cross sectional studies is able to generate hypotheses. Another
advantage of cross sectional research design is its relative cost advantage over longitudinal
research methodology. It is inexpensive to study the sample once as compared to studied
them overtime. Hence, there is no loss as such to follow up.
Cross sectional research unable to capture cause and affect relationship. Cross sectional
studies provides the picture of a single point of time. Scenario before or after the study is not

6STATISTICS AND BUSINESS RESEARCH
considered. Hence, it fails to give specific information regarding the relation among the
variables. This can be done by designing research with a longitudinal study (Lessler et al.,
2015). There are circumstances where overtime behaviour needs to be recorded. For this
purpose, use of longitudinal research is suitable as cross sectional research is not able state
overtime behaviour.
Answer 4
Procedure of data collection
There are several problems associated with the collection of data. In times of
gatheri8ng responses the following problems can be encountered.
Possibility of erroneous data collection
In times of primary data collection, there is high possibility that participants do not
give proper response to the designed questionnaire. This leads to misinterpretation of the
situation and make the research objectives futile (Miles, Huberman & Saldana, 2013). In the
above case if the respondent in the selected sample gives wrong information, then the true
level of job satisfaction can never be revealed.
Therefore, in future the researchers need to ensure that evaluation is made only on
correct responses. A cross verification of the responses can be done to judge the truthfulness
of the responses.
Lack of willingness among the respondent
The selected respondent may be unwilling to disclose information. In this situation,
even the questionnaires have been sent to the entire selected sample but only few answers.

7STATISTICS AND BUSINESS RESEARCH
Research cannot be conducted with too few responses. This is equally problematic as the case
for wrong data collection.
In order to address this issue assurance should be given about maintenance of privacy
the response. The respondents should be well aware about the research objectives.
Researchers should inform them in advance about the research questionnaires and ensure that
all of them revert as soon as they receive the questionnaires.
Obstacles in terms of language
Language of the prepared questionnaire is the foremost important aspects. There is no use
of making the questionnaire if the respondents unable to understand the language. In the
given survey, language appears as a serious problem in gathering data. There are two
dominants language in Belgium- French and Flemish. Thus, questioner needs to be prepared
in both the regional language. In times of sending questionnaire, any mismatch of location
and language creates serious problem.
Before sending questionnaire, researchers should conduct a detailed study on native
languages of chosen samples. This can be time consuming, but this will help to make a
successful research and fulfil the true objectives of research.
Insufficient time
In the surveying method where researchers send questionnaire there are considerable lag
between the time of questionnaire send and responses received. Sometimes the respondents
delay without any reasons. This is problematic for cross sectional research design because it
is conducted at a point of time. Too late responses are of no use for the researchers.

8STATISTICS AND BUSINESS RESEARCH
It is not likely to be case in times of direct or one to one response collection. In times of
mailing or sending questionnaire, a certain time can be mentioned. Collection of proper
responses within time increases the accuracy of the research.
Answer 5
Secondary data
Secondary data are those collected from any reliable source rather that gathering the
data by direct investigation. For examining representativeness of selected sample,
information collected with primary survey is compared with already recorded information.
After checking compatibility of the selected sample decisions are taken on regarding
weighting of selected data set, usage of it and in the extreme case, the data are discarded.
In order to check representativeness of the sample, three types o0f secondary data can be
used.
i) Secondary data, that are free standing
ii) Data collected as for the purpose of government measure or official statistics.
iii) Data collected for research purposes like professional data and others.
Data classified as freestanding qualitative data are not collected either by government
official or by professional researchers. There are different organizations that provide
information related to financial industry, food industry, leisure industry or for other general
purposes (Sekaran & Bougie, 2016). The data here is available at free of cost. These data
often plays an important role in research methodology by integrating the information with
qualitative aspect. Freestanding data depicts a complete picture of features for social
organisation structure and thus comple3ments the qualitative research capturing individual
behaviour. However, the cr5edibilioty aspects of these kinds of data are always questioned

9STATISTICS AND BUSINESS RESEARCH
because free access of the data. Therefore, for crucial research purpose use of these data is
least.
Government data or official statistics are one vital form of secondary data. Government
collects data on important social trends. In the official site, data are available on labour
market, transport, crime rate, education, household and others. Statistics are also available in
the form of census data, registration of birth rate, death rate, marriage and other important
events (Clark, G. (2013). This information is more reliable than those of free quantitative
data. Vital information is often of limited access and thus is free from any manipulation.
The third category of secondary data is that published on professional journals. The data
are usually kept on data archive. These data along with government data and free quantitative
data are considered as secondary data. These can be used to link with primary data for
checking representativeness. The professional research data is appropriate for projects
undertaken at small scale.
For research conducted in large scale, despite being reliable cannot be used. In order to
check representativeness and obtain significant result, secondary data collected from large-
scale surveys are used. The use of government official statistics is mostly used for this
purpose.

10STATISTICS AND BUSINESS RESEARCH
References
Campbell, D. T., & Stanley, J. C. (2015). Experimental and quasi-experimental designs for
research. Ravenio Books.
Clark, G. (2013). 5 Secondary data. Methods in Human Geography, 57.
Johnson, R. A., & Wichern, D. W. (2014). Applied multivariate statistical analysis (Vol. 4).
New Jersey: Prentice-Hall.
Koyuncu, N., & Kadilar, C. (2016). Calibration Weighting in Stratified Random
Sampling. Communications in Statistics-Simulation and Computation, 45(7), 2267-
2275.
Lessler, J., Edmunds, W. J., Halloran, M. E., Hollingsworth, T. D., & Lloyd, A. L. (2015).
Seven challenges for model-driven data collection in experimental and observational
studies. Epidemics, 10, 78-82.
Levy, P. S., & Lemeshow, S. (2013). Sampling of populations: methods and applications.
John Wiley & Sons.
Lewis, S. (2015). Qualitative inquiry and research design: Choosing among five
approaches. Health promotion practice, 16(4), 473-475.
Malterud, K., Siersma, V. D., & Guassora, A. D. (2016). Sample size in qualitative interview
studies: guided by information power. Qualitative health research, 26(13), 1753-
1760.
Marshall, B., Cardon, P., Poddar, A., & Fontenot, R. (2013). Does sample size matter in
qualitative research?: A review of qualitative interviews in IS research. Journal of
Computer Information Systems, 54(1), 11-22.

11STATISTICS AND BUSINESS RESEARCH
Miles, M. B., Huberman, A. M., & Saldana, J. (2013). Qualitative data analysis. Sage.
Palinkas, L. A., Horwitz, S. M., Green, C. A., Wisdom, J. P., Duan, N., & Hoagwood, K.
(2015). Purposeful sampling for qualitative data collection and analysis in mixed
method implementation research. Administration and Policy in Mental Health and
Mental Health Services Research, 42(5), 533-544.
Sekaran, U., & Bougie, R. (2016). Research methods for business: A skill building approach.
John Wiley & Sons.