Introduction Collecting data Your research project may involve collecting new data, or it may use existing data, such as financial data or previously collected survey data from an organisation. Using existing data can be a practical and legitimate approach for master's-level research; you will, however, still need to determine which data you will use, from what source, from what time period and so forth. And you will need to determine how you will analyse the data you have identified. In this unit, you will explore methods for collecting or choosing primary or secondary data based on your chosen methodological approach, including methods for sampling and determining an appropriate sample size. Whichever methods you choose to collect your data, remember that your research design and methodology is like a blueprint – crucial in guiding your choice of methods and helping you to arrive at valid findings, comparisons and conclusions. Again, this unit only provides an overview of some of the key issues involved in sampling and collecting data – be prepared to learn more on your own with the help of your supervisor. Qualitative methods As the name implies, qualitative methods provide non-numerical, narrative or thematic data which derive from unique cases that occur in the world. Qualitative methods provide more flexible ways to perform data collection and subsequent analysis and interpretation of collected information. Qualitative data collection encourages a holistic view of the phenomena under investigation and allows the researcher to interact with the research subjects in their own language and on their own terms. Qualitative methods useful in management research include interviews, focus groups, open-ended questionnaires, direct observation and analysis of existing documents. Quantitative methods Quantitative methods provide numerical data that can be analysed using a variety of statistical techniques. Deciding upon a method of analysis depends upon the nature of the data collected, as well as the research design and methodology employed. Quantitative data can come in the following forms: Nominative: categorical data, such as male/female, or categories of employees with different job titles. Ordinal: ranked data, such as responses on a scale of 1 to 5. Interval: data that occur on a continuous scale but for which there is no true zero point, such as temperature. Ratio: continuous data for which there is a true zero point and for which averages and other descriptive statistics can be calculated, such as age, financial sums or test scores. Almost any data collection method that can be conducted qualitatively can also be conducted quantitatively. Interviews can be conducted in a highly structured manner, asking participants to respond on a scale or provide numerical responses. Observations can be carried out by coding and counting observable behaviors in particular predetermined categories. Questionnaires can produce quantitative data, though they can also include open-ended qualitative items. A great deal of pre-existing quantitative data are available in management research, including economic and financial data, regularly conducted employee surveys and data on staff, clients and customers. Some quantitative methods will allow you to test a hypothesis. Recall the example from the previous unit of a study on leadership and staff turnover; if your hypothesis is, 'Experienced leaders will have less staff turnover than inexperienced leaders', your research study should gather data that will enable you to determine if, in fact, experienced leaders do have less staff turnover in their companies. You would need to define what you mean by 'experienced' and 'inexperienced' and a way to measure turnover. You could then identify two sets of leaders, one experienced and one inexperienced, measure turnover for both groups and determine if there is any difference in the measure you chose. Sampling As it is impossible in most cases to include every member of a population under study, you will need to limit your data collection activities to a smaller sample of the population. As with your choice of data collection methods, your choice of a sampling strategy will depend upon your chosen methodology. Survey research, for example, often aims to produce conclusions that can be generalised from a sample to a larger population; these methodologies require a particular approach to sampling (Easterby-Smith et al., 2008). Research based on qualitative data may not have such generalisation as a goal and thus require different approaches to determining who participates in the research. There are two major decisions to be made regarding the sample: • Who or what to include in the sample • What the sample size should be Sampling strategies fall into two major categories, random and non-random. In random sampling, each member of the population has an equal chance of being selected to become part of the sample. There are several techniques for choosing a random sample, but the aim is always to be able to establish the relationship between the sample characteristics and the population characteristics (Easterby-Smith et al., 2008, p. 216). Non-random sampling strategies include purposive sampling, based on specified eligibility criteria; quota sampling, aimed at creating subsamples in proportion to the larger population; and convenience sampling, or sampling based purely on participants' accessibility. The goal with both random and non-random sampling strategies should always be to achieve a credible sample and avoid introducing bias into your study via inappropriate sampling strategies (Easterby-Smith et al. 2008). Sample size Deciding on the size of a sample is usually a compromise between the scope of the problem the researcher is working to address and practical issues such as cost, time and resources available. In the case of quantitative studies, statistical theory must be considered when trying to ensure a predefined degree of precision. If you plan to use quantitative methods, you should consult a statistics textbook or website and work with your supervisor to devise your sampling approach. Ethical considerations At the time you submit your proposal for your Management Research Project for formal approval, you will also need to submit a proposal for ethical approval of your project. All researchers must adhere to strict ethical requirements to ensure that their research does not inadvertently harm any individual or organisation by breaking confidentiality, sharing information that should not be shared or making it possible to identify a particular participant in your study. If your study involves collecting or using primary data from human participants, it is vitally important that you adhere to requirements for doing research with human subjects. If your research involves interviews, focus groups, surveys or observation, you must take appropriate steps to protect the identity of your participants. Your participants must know the purpose of the research in which they are participating, and you must document their understanding and willingness to participate. Additionally, you must not plan to interview anyone who is your subordinate in the organisation. Ethically, you should never leverage your authority for your own benefit, to include gathering identifiable data from people who report to you, as well as using existing data to which you have special access. Doing research in an ethical manner goes beyond the requirements of formal ethical approval. As you design your research study and select your methods and analytical approaches, consider how you can implement these in the most ethical and responsible manner possible.