BBRC4103 Research Methodology for Sampling Design

Introduction

In the given scenario, there exists a company named Triple A which is facing a host of issues including that of employee attitude. It is feared that if the issues are not sorted out then potentially the company may be liquidated in the near future. In the wake of this background, the problem statement for the given research is to improve the performance for the given company. In wake of the defined problem statement, the key objectives of the research would be as follows.

• To identify the key determinants leading to attitude problems related to employees
• To identify other internal issues which are hampering the performance of the company
• To identify ways through which the performance of the company can be improved

Sampling Design

One of the most crucial steps of any research is sampling design. This is primarily because in order to collect the requisite information, a sample is the rational choice in terms of both cost and time. As a result, it is crucial that the underlying sample must be reflective of the underlying population. This ensures that the data obtained from the sampling is reliable and leads to conclusions that are both valid and relevant. There is no ideal sampling design and essentially the best sampling design is driven by the underlying objectives of the research along with the resources available. Sampling design essentially highlights the sampling method which would be used to derive the sample based on the definition of the population and also the resources available along with research objectives (Koch, 2013).

Based on the problem at hand, the first objective would be to define the population of interest from which the sample is to be derived. For the given research problem, the population of interest would be essentially the employees of the company Triple A. Having defined the population, the next time is to decide the sampling method which would be used to draw the sample. With regards to sampling methods, there are broadly two types of techniques namely probabilistic sampling and non-probabilistic sampling. Considering the superiority of the former over the latter coupled with the readily available sample, the appropriate sampling technique should be a probabilistic sampling technique only (Eriksson & Kovalainen, 2015).

One of the simplest probability sampling methods is the simple random sampling whereby from all the employees of the company, a particular sample size would be selected on a random basis. However, the simple random sampling is not suitable in this case as it may happen that majority of the employees are working at a particular level in the company or belong to a particular department. Owing to selection of such a sample, it may happen that only selective issues get highlighted which would result in only limited issues being addressed. As a result, the performance of the organisation would continue to be sub-optimal. Hence, a superior alternative with regards to sampling method would be stratified random sampling. In this sampling method, the population of employees would be segregated in terms of organisational levels along with key departments (Flick, 2015).

Once this segregation is complete, then from each department and each organisation level, random sample would be selected which would constitute the overall sample for the research problem at hand. The individual contribution from each department and each level would be such that the sample would contain representation in the same proportion as in the population of interest. This would ensure that the manager would be able to understand the problems across the organisation. It is essential to note that since the employee count is not highlighted, hence no recommendation on the appropriate sample size has been provided. Further, since the attitude of employees is an issue, thus, the manager would be able to gain information at the organisation wide level. This would lead to identification of issues which otherwise may be ignored (Hair et. al., 2015).

Data Collection

Once the sample is selected, it is essential that data must be collected from the sample. With regards to data collection, in accordance with the broad classification, either a primary or secondary method of data collection may be used. Primary data collection is an umbrella term used to describe all the data collection methods where data is collected by the researcher or a designated person concerned. However, in case of secondary data collection, data is obtained from various databases or research studies which have been already conducted. In the given case, it is apparent that primary data collection technique would have to be used as the population of interest comprises of the employees of the given company i.e. Triple A. Further, there are a number of problems with the employee attitude being just one of the problems. As a result, the secondary data collection would not help as it is unlikely that any previous research study would have been done using Triple A company employees. Additionally, considering the varied nature of the problem, it makes sense to undergo primary research whereby data is collected from the selected sample (Hastie, Tibshirani & Friedman, 2011).

There are various methods which are available under the primary data collection methods. These include observation, scientific experiments, case studies, focus group interviews, questionnaires and interview. It is apparent for the problem at hand observation does not seem an appropriate data collection method since it would be time consuming in the given case and also because it makes more sense to collect the responses from the employees so that their issues could be understood. Further, scientific experiment and case studies are not feasible choices in the given case considering the nature of the problem. The focus group interview could be of relevance in the given case but it would essentially depend on the nature of the problem. If the problem regarding the attitude of the employees relates to the practices of the supervisors, then it makes sense to conduct interview of these employees so as to understand the key issues at hand. Hence, this data collection technique could be used at a departmental level also so that the issues pertaining to a particular department be brought to light. But it would be better if focus group interview pertains to only one group which comprises of representation from various levels and departments. This would enable prevention of blame game as representation is not limited to a particular department or level (Hillier, 2006).

However, another alternative to focus group interview could be individual interviews which could be structured or unstructured. In the given case, based on whether the manager is aware of the issues with the company or not, the appropriate choice would have to be made. For instance, if limited information about the issues plaguing the organisation is available with the manager and the objective is to uncover the issues, then unstructured interview could potentially be more suitable as the questions asked are more open ended and hence a lot of crucial information can be gained. On the other hand, if the manager is well aware of the issues facing the organisation and wants to merely test the existence of these issues, then a structured interview would be more suitable as same questions would be asked to each and every interviewee. Further, in case of structured interview, the training requirements for the interviewer are also comparatively lesser as the questions are essentially the same. A drawback of the interview method is the amount of time consumed along with the cost involved. As a result, interview could only be used to supplement the information collected through the use of questionnaires. Questionnaires essentially consist of certain questions which are presented to the sample and the output is recorded based on which analysis is performed to derive conclusions (Flick, 2015).

Survey Method

From the discussion carried out in the above section, it is apparent that survey would have to be conducted so as to collect the primary data in the form of responses of the sample employees selected. With regards to the survey, there are various choices as the survey could be a mail survey, telephonic survey or in the form of personal interview. It is apparent that collecting data in the form of a personal interview in the given case would be time consuming along with being expensive and therefore would be avoided. Instead, only limited people would be interviewed so as to substantiate the information collected through the survey and also for the formulation of the questionnaire. In the given case, in order to draw true responses from the sample employees, it would be imperative that they are able to keep their anonymity. This is especially true for employees at the lowest rung of the organisation who may fear wrath of their seniors and supervisors owing to which they may be diplomatic in their responses. As a result mail or telephonic survey would not be suitable in the given case (Hair et. al., 2015).

It would be advisable that for the given research problem printed questionnaire should be given to the various sample employees. These employees can then be briefed on the purpose of the questionnaire and the fact that if the issues were not addressed, then potentially the company could close down in the future. Also, explanation regarding the questions should be offered so as to avoid any misunderstanding and confusion. Further, a drop box should be set and employees should be given a deadline before which the filled questionnaire should be put in the box. The questionnaire should not ask for the identity of the concerned participant so that there is no fear amongst the employees and they can record their honest opinion. Moreover, this survey method would be quick and also cost effective. It is noteworthy that this method is also preferable to the administered survey method as it may result in responses being adversely impacted owing to peer pressure (Eriksson & Kovalainen, 2015).

Data Analysis

This is the step wherein the analysis of the data collected through survey is conducted for deriving meaningful conclusions so as to provide answers to the research problem and hence enable achievement of research objectives. However, it is essential that the data obtained from the data collection stage need to be cleansed and transformed before it can be analysed. This is essential as it is quite possible that some of the respondents may not have answered all the questions. Further, it is also a possibility that the answers to certain questions may constitute an invalid response. Considering the survey method that has been used for the given research, it is apparent that such mistakes cannot be corrected once the respondent has dropped the completed form in the drop box as the identity of the respondent cannot be established. As a result, before analysing the data suitable adjustments need to be made along with potential transformation so that statistical analysis becomes convenient (Hillier, 2006).

With regards to the analysis of the data obtained after cleaning and transformation, both descriptive as well as inferential statistical techniques would be used. The purpose of these two techniques would be different. The descriptive statistical techniques would tend to summarise the responses of the sample respondents through the use of suitable charts and tables. As a result, these would not convey any direct information about the population as a whole. On the other hand, through the use of inferential statistics, hypothesis testing would be performed through the use of sample data collected. Thus, these techniques would focus on deriving conclusions about the conclusion while using the sample data as the basis. In order to perform the above analysis, MS-Excel would be used as an enabling tool so that time can be saved and higher accuracy could be achieved (Harmon, 2011).

Further, in order to understand the relationships between various variables of interest, both correlation analysis along with regression analysis would be performed. This would enable not only better understanding of the issues at hand but also enable insight into potential solutions of the concerns to be addressed. This could include the correlation between salary levels and job satisfaction of employees at various analysis levels. Further, in order to interpret the results, emphasis can be paid on the existing literature in this regards which could provide a suitable reference. A low correlation would imply that various factors besides money are responsible for the poor attitude of the employees and hence requisite steps would be required to address these concerns. As a result, relational analysis would be extremely critical in data analysis. Moreover, while selecting suitable tools and techniques for analysis, consideration would need to be paid on the research objectives coupled with the potential weaknesses and shortcomings of the methods deployed (Lieberman et. al., 2013). This would enable the researcher to carry out a comprehensive analysis of data and draw conclusions which are both relevant and valid. The results could then be summarised in the form of a report which can be shared with the top management so that a course of action may be outlined.

Summary

On the basis of the above discussion, it can be summarised that the objective of the given research is to find out the key issues with special emphasis on the poor attitude and possible dissatisfaction of employees. Further, a key objective of the study is also to focus on the potential solutions which would address the business problem which is to improve the performance of the company. The requisite population of interest comprises of the employees of the company i.e. Triple A. In order to obtain the sample for the given research, the appropriate sampling method would be stratified random sampling. This is because it would ensure that sample obtained would be quite representative of the underlying population. With regards to data collection, primary methods would be given priority over secondary data since there would be absence of much research involving employees of the company. Additional for the purposes of this research, historical data would not be relevant. In regards to primary data collection, the suitable methods would be interview and questionnaires. However, it is noteworthy that questionnaires would be directed or aimed at all the sample employees unlike interview which would be used in a restricted manner owing to the high requirements related to time and money.

With regards to implementation of the survey, keeping the anonymity of the sample respondents is pivotal so that the responses collected are unbiased. As a result, a drop box would be used to collect the filled questionnaire from sample respondents. Further, to avoid any confusion and to communicate the purpose of the research, the sample employees would be briefed beforehand. After the collection of the data, cleansing and transformation of data would be performed before the same can be analysed. With regards to carrying out analysis, a combination of both descriptive and inferential statistical techniques would be used. Descriptive statistics would be aimed at summarising the responses and highlighting the relationships between various variables. However, inferential statistics would aim to reach conclusions about the population based on the sample data through the use of hypothesis testing.

References

Eriksson, P. & Kovalainen, A. (2015). Quantitative methods in business research (3^rd ed.). London: Sage Publications.

Flick, U. (2015). Introducing research methodology: A beginner's guide to doing a research project (4^th ed.). New York: Sage Publications.

Hair, J. F., Wolfinbarger, M., Money, A. H., Samouel, P., & Page, M. J. (2015). Essentials of business research methods (2^nd ed.). New York: Routledge.

Harmon, M. (2011). Hypothesis Testing in Excel - The Excel Statistical Master (7^th ed.). Florida: Mark Harmon.

Hastie, T., Tibshirani, R. & Friedman, J. (2011). The Elements of Statistical Learning (4^th ed.). New York: Springer Publications.

Hillier, F. (2006). Introduction to Operations Research. (6^th ed.). New York: McGraw Hill Publications.

Koch, K.R. (2013). Parameter Estimation and Hypothesis Testing in Linear Models (2^nd ed.). London: Springer Science & Business Media.

Lieberman, F. J., Nag, B., Hiller, F.S. & Basu, P. (2013). Introduction To Operations Research (5^th ed.). New Delhi: Tata McGraw Hill Publishers.