Introduction
When conducting research, it is important to have an understanding of the appropriate methods that can help in gathering relevant data to directly respond to the research questions. The case study of Swire Oil and Gas Company is typical research where it will be very important to collect specific data that respond to specific research questions. In this study, the researcher was interested in investigating whether the development program has instilled the necessary skills in the employees of this firm. It also focused on investigating whether it enabled it to retain its best workforce. According to Tharenou, Donohue, and Cooper (2007), when researching such a specific issue, it becomes very important to formulate a clear path and instruments that will be needed to respond to the specific questions. In this critical literature synthesis, the researcher will look at several factors that should be considered when conducting such a specific study.
Literature review
According to Creswell (2013), when conducting research under the context explained in the section above, formulating a workable researchable question is very important. A research question should target specific issues which are relevant to the study. The workable research question should, therefore, focus on an issue that can be explained by the respondents in clear terms. This means that the question should neither be ambiguous nor too complex to comprehend. It should be self-explanatory. Even when the researcher is not around to expound on what is needed from such a question, the respondent should be in a position to give clear answers without having to think about the meaning of the question. As Flick (2009) notes, such a question should also be specific. It should focus on a specific issue, letting other questions focus on other issues. This will eliminate cases where responses are crowded around a single question making it difficult to compile data for subsequent analysis.
Choosing a data collection method and developing specific measurements or instruments to address the question developed above are critical activities in this study. The data collection method can be defined based on the nature of the questions and the responses expected from the participants. Sometimes questions may be structured, while in other cases, questions may be unstructured. When a researcher is collecting data using structured questions, then using instruments such as questionnaires which are e-mailed to the respondents may be appropriate. This is so because, in most cases, such data only requires the respondents to tick as appropriate among the choices provided. The researcher will be playing minor roles in such contexts unless the respondent is illiterate, which is not the case in the scenario provided. On the other hand, when questions are unstructured, then the data required will be qualitative. In such contexts, it will force the respondent to be physically present because of the need to get a detailed explanation of a phenomenon. Face-to-face interactions also make it possible for the researcher to collect non-verbal cues. The non-verbal cues sometimes help in explaining some issues when collecting qualitative data. A researcher can determine when the respondent is holding back information by looking at their facial expressions and other bodily languages.
With a limited budget and the fact that respondents are from different cohorts, locations, and backgrounds, defining an appropriate sampling method is of critical importance. According to Alvesson and Skoldberg (2000) quality data can be determined by the type of sampling method used. In many cases, researchers would allow their emotions and personal opinions to guide them in selecting the participants in a study. When this happens, it becomes almost impossible to have unbiased data because the process of selecting sample participants was in itself biased. For this reason, the researcher will use stratified sampling. This means that the researcher will first classify the participants into different strata based on their managerial positions in the firm, social backgrounds, locations, and any other relevant demographics. The strata will then be assigned values based on their significance to know the number of the respondents to be taken from each of them. The researcher will then use a simple random sampling method to select the individual participants in each of the selected stratum. Simple random sampling helps in eliminating any form of biases in data collection (Greene, Speizer & Wiitala, 2008). This sampling technique is unbiased and very comprehensive. It will help in ensuring that the collected information looks at all the perspectives of the issues under investigation.
When conducting the study, the researcher will put into consideration the fact that participants are from different cohorts, locations, and backgrounds. This means that the study must capture the diversity that is characterized in the study to come up with a conclusion that pays attention to all the variables. To do this, the researcher will find a way of getting the right data from all the regions where this firm operates. As Creswell (2013) says, a given strategy may work well in the United Kingdom but fail to work in China because of the differences in the socio-economic and political background in these two countries. For that reason, the researcher proposes that it may be necessary to look at each of the major regions where this firm operates independently before coming up with an overall conclusion about the issue. The success rates of the human resource development program will be looked at independently in England, China, and the United States before compiling a single report explaining how effective the program is in retaining the best skills and talents within a firm.
Reformulating the research questions
Easterby-Smith, Thorpe, and Jackson (2012) advise that it is important to understand the orientation of the research when collecting primary data. For instance, it may be necessary to reformulate the research question to fit into a quantitative study. To do this, the researcher may need to adjust the question from an unstructured form to a structured format. This means that the question will be quantified to ensure that the response can be computed for statistical analysis. To have a research question that can be addressed using quantitative-oriented research methodologies, Lewis, Watson, and White (2009) say that one must structure the question to ensure the responses can be given a numerical value, say 1-5.
Developing specific instruments for the research
It has been stated that this study will have to be carried out in different regions where the current firm operates to determine the effectiveness of the employee management development program used in this organization. However, it may be realistically impossible for the researcher to visit China, Hong Kong, England, and the United States to collect data given the financial and time constraints. This makes it necessary that an instrument be developed that can help the researcher to collect relevant data from the respondents in these regions without necessarily having to make expensive and time-consuming travels. Using a questionnaire can help in this complex situation. It has been stated that the study will use quantitative methods. This means that the researcher can use a properly designed questionnaire and collect the needed data from the respondents in these locations. The only caution that the researcher will have to take is that the questions should be self-explanatory. In cases where the respondent is expected to give a tick, this should be clearly stated, and when it becomes necessary that the respondents give explanations to support their answers, then this should also be clearly stated in the questionnaire. This is so because any form of confusion would bring about the collection of wrong data that does not make any sense to the study.
The questions used in the study will be given in Likert scales, especially those meant for qualitative analysis. The values will be running from 1-5 where 1 would be strongly disagreeing and 5 will be strongly agreeing based on the nature of the question asked. The terms ‘agree’ and ‘disagree’ may change depending on the question posed to the respondents. The researcher will not be physically available at the sites where the primary data will be collected. However, there are some explanations needed to help support the statistical values given in the questionnaires. To have such explanations, it will be appropriate to have a few open-ended questions where the respondents are allowed to give their personal opinions about the employee management development program. Their views will help in validating the responses obtained from the statistical values.
Sampling
Sampling will be required because of the need to collect data from the employees of different cadres. There will be the top managers, mid-managers, and junior employees. The three strata have numerous employees that cannot be surveyed as a whole. This makes it necessary to have a sample (Hutchison, 2009). The sample will be defined in the three categories mentioned above. The major element of the sampling frame will be the managerial position of the respondents. The reason behind this is that sometimes the views of the managers may differ from that of the employees. The managers may want to hide the truth about some factors as a way of protecting the image of the firm. If necessary, demographical factors such as the age of the respondents, gender, race, religion, and level of education may form other important elements, especially in cases where it is believed that they may influence the response of the participants.
Response rate
The issue of response is of critical importance in this study. According to Giacalone and Rosenfeld (2013), sometimes the number of those who respond and send back the questionnaires e-mailed to them may be very low. A researcher is not guaranteed that the respondent will fill the questionnaire and e-mail it back on time. To address this issue, the researcher will maintain close interaction with the respondents, especially through social media such as Facebook to constantly remind them of the need to fill the questionnaires and e-mail them back. In other cases, the researcher will make a direct phone call to the respondents to follow up on the progress (Hoonakker & Carayon, 2009). When developing an approach to this research, it will be very important to ensure that the data collected takes into consideration issues such as the diversity of the workforce and the unique environmental factors in the four countries identified above.
Conclusion
When conducting research, some factors should be put into consideration to ensure that the relevant data is gathered and analyzed to give the most accurate information about the state of affairs of the issue being investigated. Developing a research question is the first step after the objectives of the study have been defined. The research questions help in knowing the type of data to be collected. When the size of the population to be studied is large as was the case in our study, then it may be necessary to develop a sample that will be a representation of the entire population. The sampling method should be chosen based on the diversity of the respondents. If the issue is affecting a population with almost similar characteristics such as the impact of exams on students, then simple random sampling may be appropriate. However, if the factor under investigation may affect different segments of the population differently, then a stratified sampling technique should be used. This study should be designed in a way that will facilitate comprehensive data collection from the diversified respondents.
References
Alvesson, K., & Skoldberg, M. (2000). Reflexive methodology: New vistas for qualitative research. London, United Kingdom: Sage.
Creswell, J. (2013). Qualitative inquiry and research design: choosing among five approaches. London: Sage.
Easterby-Smith, M., Thorpe, R. & Jackson, P. (2012). Management research. London: Sage.
Flick, U. (2009). An introduction to qualitative research. Los Angeles: Sage Publications.
Giacalone, R. A., & Rosenfeld, P. (2013). Impression management in the organisation. New York: Psychology Press.
Greene, J., Speizer, H. & Wiitala, W. (2008). Telephone and Web: mixed-mode challenge. Health Services Research, 43(1), 230-248.
Hoonakker, P. & Carayon, P. (2009). Questionnaire survey nonresponsive: a comparison of postal mail and Internet surveys. International Journal of Human-Computer Interaction, 25(5), 348-373.
Hutchison, D. (2009). Designing your sample anciently: clustering effects in education survey. Educational Research, 51(1), 109-126.
Lewis, I., Watson, B. & White, K. (2009). Internet versus paper-and-pencil survey methods in psychological experiments: equivalence testing of participant responses to health-related messages. Australian Journal of Psychology, 61(2), 107-116.
Merriam, S. B., & Merriam, S. B. (2009). Qualitative research: A guide to design and implementation. San Francisco: Jossey-Bass.
Tharenou, P., Donohue, R., & Cooper, B. (2007). Management research methods. New York: Cambridge University Press.