Sunteți pe pagina 1din 10

1.1.

Research Design Definition and Application

Research design projects has thought me that the fundamental steps in a research design is to understand your projects purpose, needs and expectations. These steps would assist a researcher in aligning the intended strategy with the delivery plan and cost implications. Research design can be best described as a systematic plan for the collection and utilization of data so that information on what needs to be done, how it will be done and how the data will be analysed can be obtained. As described in Coldwell and Herbst (2004: 35), research design provides the glue that holds the research project together. This statement is true because a research design structure provides a framework for projects to operate effectively with direction. Research design plays a role similar to a support function which allows information and data collection to be carried out in a structured way that best suits the projects objectives, goals and scope. Without research approach and methodology, information validity, reliability and generalisability can not be confirmed (Shipham, 2009: 3) which would resulted in weak conclusions and not credible reaction to the research project. At ABC, the impact of globalisation on research design is apparent. The research designs plan is difficult to manage and carry out in regions and countries where the external and far environments are different, i.e. regions like Asia, Africa and Middle East. The likely factors that possess potential threats or obstacles to the researcher are: y y y y y 1.1.2 Sociological factors- gender role, pattern of work Technological factors- information and communication technology (lack or poor infrastructures) Economic factors- government trade restrictions Environmental factors- legislation Political factors- legislation, trading relationships

General Principles of Research Design and Application

CIOS (2009) mentions that research designs can be classified into the following three types based according to the amount of control a researcher maintains over the conduct of the research study: y Experimental research- the researcher controls the setting in which the research is conducted (the laboratory) and also manipulates the levels of the independent variable or variables, and follows this by observation of the corresponding changes in the dependent variable or variables. Field research- the researcher retains control over the independent variables, but conducts the research in a natural setting, without any control over environmental influences.

Observational research- the researcher is limited to measuring, rather than manipulating the independent variable. Like field research, observational research designs exert no control over the setting in which the hypothetical process occurs.

This classification will definitely improve identifying appropriate research designs for future research study with less confusion. Application of this classification would assist my methodology design in determining which research technique should be used, and how it should be applied to meet my research projects objectives. The selection of the research design type is a critical step in a research study because this step determines the methodologies and procedures for data and information collection and analysis. Therefore a researcher must be aware and clearly understand the following selection criteria described by Martz (2010) before making their choice: y Identify the objectives of the research y Clarify the exact question to be answered y Determine how much is known about the marketing problem y Determine what needs to be known to address the problem y Importance factors that influence research design  The degree of generalization desired  Constraints on time and resources  Social context Coldwell and Herbst (2004: 40) highlights that validity (both internal and external) is one of the great threats to the three types of research design. Unknown (no date) mentions that internal validity describes the ability of the research design to unambiguously test the research, whilst external validity refers to the generalisability of the research, that is, the ability of its conclusions to be validly extended from the specific environment in which the research study is conducted to similar actual situations hypothesis. Majority of the research study at the organisation predominately involves experimental research in product development. The common organisations objectives for research study in product development involves determining if there are distinct consumer liking segments with different product sensory needs, resulting in hypotheses for how to diversify ABC products & differentiate from competition.

The two main methods or approaches used for research designs are qualitative and quantitative and each approach possesses their methods. These methods have constraints and limitations and therefore careful consideration must be taken during selection of a method for a specific project. The fundamental step is to ensure that the method employed for data collection is aligned to the appropriate approach, i.e. qualitative or quantitative. 1.2.1. Quantitative Research Quantitative research is thought to be objective whereas qualitative research often involves a subjective element in which gaining, analysing and interpreting quantitative data, the researcher can remain detached and objective, i.e. quantitative designs of research tend to produce results that can be generalised (Neill, 2003). Coldwell and Herbst (2004: 38) and Neill (2003) identify three major types of quantitative research which include: y Experimental- The paradigm for scientific method in research is the true experiment or randomised control trial. Experimental designs are set up to allow the greatest amount of control possible so that causality may be examined closely. The three essential elements) of experimental design are manipulation, control and randomisation. Quasi-experimental- developed to provide alternate means for examining causality in situations which were not conducive to experimental control. Non-experimental- to gain more information about a particular characteristic within a particular field of study. A descriptive study may be used to, develop theory, identify problems with current practice, justify current practice, make judgements or identify what others in similar situations may be doing. There is no manipulation of variables and no attempt to establish causality.

Experimental and quasi-experimental quantitative approaches are designed to investigate the effect independent variable change has on dependent variables, i.e. cause and effect situations where else non-experimental research is more concerned with the examination of relationships between variables. The lack of universalism causes problems when critiquing research as many published studies do not identify the design used. Selecting an appropriate design for a study involves following a logical thought process. A calculating mind is required to explore all possible consequences of using a particular design in a study (Neill, 2003).

1.2.2 Qualitative Research Qualitative research will be used in my research study, and for the organisation when trying to seek a better understanding of the consumers needs, lifestyle and their perspective. This research can assist in development as a researcher to better understand the organisations brands values and personality and to identify any inconsistencies within a specific brand. Some specific examples when qualitative research was appropriate for an organisation research design: y To explore reactions to a new product in development, to gain insight into what consumers consider to be its unique benefits To explore reactions to ideas for advertising routes and executions, to help build understanding of strengths and opportunities for improving communication prior to quantitative testing To gain insight into how packaging (the actual pack and/or its graphics) communicates to consumers

ABC MRC (2000) states that qualitative research can be used for a variety of purposes and may be appropriate in any project such as fairly straightforward initial exploration in order to form hypotheses or to prepare a quantitative study, through to an intensive study using psychological methods to provide an understanding of behaviour, attitudes or opinions which a respondent could not articulate in response to direct questioning. The need for qualitative research is required when research objectives require deeper penetration into the consumers response to better understand their needs Because of the understanding it can provide, Qualitative Research is one of the most important tools of market research. It can provide profound insight into consumer motivation and can inspire a leap forward in the way a project is progressed. However, it can also be the most abused form of market research. This paper provides Guidelines on how to make the best use of Qualitative Research. The skills of the researcher become increasingly critical as the level of consciousness become deeper. It is entirely inappropriate to use anyone other than a highly skilled and experienced qualitative researcher to explore intuitive associations and unconscious factors (ABC MRC, 2000).

Question 1.3 Baker (2003: 157) describes primary data as those which are collected to fulfil the demand of the current research project and have to be gathered should secondary sources of data fail to provide the information necessary to meet the research objectives. Primary data is first hand collected information by a researcher; this source of data is considered to be expensive and time consuming. Open University (2003) highlights the following as sources of primary data: y y y y Interviews Surveys Questionnaires Focus groups

Secondary data is the information that already exists from other or previous research. Newson-Smith (2003) says that secondary data can provide a background to primary research if the research has already been conducted by someone else, and act as a substitute for field research since secondary data is relatively cheap. This may help save unnecessary expenditure in that published data may fully meet all the current research objectives. Open University (2003) highlights the following as sources for secondary data: y Journals y Books y Census Data y Newspaper articles In a FMCG industrial sector, real time and current data is crucial in the organisation survival and growth. I would consider primary data more reliable than secondary data because the information collected would be specific to the business marketplace and products. Information and data validity and authenticity collected before technological advancement in computer would be questionable because of the lack of analytic tools and slow process-ability. Boslaugh (no date) states that secondary data is inherent in its nature: because the data were not collected to answer a specific research questions, particular information that a researcher would like to have may not have been collected. This lack of or missing information could result in organisation failure to take advantage of opportunities.

Question 1.4 Triangulation provides researchers the ability to explore a particular area of research with different methods and kinds of data to enhance their understanding and to look at a situation from a holistic point of view. Easterby-Smith, Thorpe and Lowe (2004: 146) identify four forms of triangulations, these include: y Theoretical triangulation- involves borrowing models from one discipline and using them to explain situations in another discipline. Date triangulation- refers to research where data is collected over different time frames or from different sources. Triangulation by investigators- is where different people collect data on the same situation and data, and the results are then compared. Methodological triangulation- uses both quantitative and qualitative methods for data collection.

Ramchander (2004: 109) mentions that whilst quantitative methods have been regarded as deductive (associated with the formulation and testing of hypotheses), quantitative methods are associated with an inductive approach. The use of deductive and inductive approaches to a research design will enhance objective thinking and reduce preconception thoughts which hinder the flow-ability of creative ideas. In my research design, the benefits below unravelled by Ramchander (2004: 110), Massey (2004: 110) and Blaikie (2004: 110) by combining quantitative and qualitative methods will yield improved validity and reliability in my project finding. These benefits include: y Quantitative data is collected under controlled conditions in order to rule the possibility that variables other than the one under study may account for the relationship identified while qualitative data is collected within the context of its natural occurrence Both quantitative and qualitative research seeks reliable and valid results. Data that is consistent or stable as indicated by the researchers ability to replicate the findings, is of major concern in the quantitative arena, while the validity of the qualitative findings is paramount so that data is representative of a true and full picture of the constructs under investigation.

Question 2 2.1. Representative sample A representative sample is sample taken from a particular sample plan that best reflects the characteristics and properties of interest of the population. In order to get a representative sample, the researchers must first identify the population being sampled. Researchers must think of a multitude of methods for collecting data to ensure that evenly samples all aspects of the population being studied (Smith, 2010). When a researcher read a study which has been conducted with the use of representative sampling, it is a good idea to find out which methods the researchers used. Sampling error can yield incorrect results, and therefore a researcher needs to know how the data was collected, who it was collected from, and what sort of controls were in place to ensure that the sampling was representative. By using critical thinking to look at statistics and representative sampling, a researcher will be able to determine whether or not they are truly useful and applicable. Some clues that a study might not be valid include the use of self-response surveys, which have a high rate of non-response which would skew the sample, and indications that the sample was taken from a smaller sub-community of a larger group (Smith, 2010). 2.2. Sample Size Open University (2003: 61) mentions that the larger the sample, the greater the likelihood of obtaining results that represent an average cross sectional of the population under study with any chance deviation tending to cancel each other. Coldwell and Herbst (2004: 82) states that the sample size depends on the following: y y y y y Whether the population is finite or infinite The precision of the estimates one wishes to achieve The confidence one needs to have in the finding being accurate/correct The number of variables that have to be examined simultaneously How heterogeneous the population sample is.

Sample size chosen by the organisation for a particular research design is determined in most cases by available funding limitation. Allocation of samples and management of the size is costly exercise for the organisation and therefore a size restriction is normally introduced to reduce cost. Authenticity is questionable when the sample size comprises of various samples of similar characteristics but different behavioural reactions to external stimulus. 2.3. Sampling bias and sampling error A sample should be expected to best represent its population from with it was drawn from but this is not true as there is no guarantee that any sample will taken from a population will precisely represent that population. Chances may dictate that a disproportionate number of the untypical observations will be made (Coldwell and Herbst, 2004:76).

The standard error or sampling error gives a researcher some idea of the precision of the statistical estimate. A low sampling error means that a relatively less variability or range in the sampling distribution made. The standard error has an inversely proportional relationship with the sample size (Coldwell and Herbst, 2004:76). Coldwell and Herbst (2004: 77) highlights two basic causes of sampling error, these are: y Chance- error that occurs just because of bad luck, i.e. result of untypical choices. y Sampling bias- is a tendency to favour the selection of units that have particular characteristics, i.e. result of poor sampling plan.

2.4. Reliability It is extremely important for reliable sampling that any procedure allows every individual item in a lot an equal chance of being chosen, i.e. random sampling. Sampling procedures which prohibit or reduce the chances of any sample from the sample size from being chosen, e.g. sampling instrumentation that does not allow larger particles to be captured, results in bias. If a researcher is able to test a particular hypothesis to see whether it sustains the test, the researcher needs to know that he/she is subjecting the same hypothesis to exactly the same test to which it was subjected before. There is one technique that can determine reliability, this is test-retest reliability. The same population is tested twice and the results are compared. The comparison usually consists of a correlation relation, and the greater the magnitude of the correlation of the scores of the test given to the same population on two separation occasions, the greater the reliability of the test (Coldwell and Herbst, 2004:17). 2.5. Validity including threats to validity/improving validity Although weve identified some general ways that research designs may fail to achieve internal and external validity, we need to talk in more detail about some of the specific problems in the validity of research design. For both internal and external validity, well discuss threats that occur when measurement takes place over a time span, and threats that occur at single time points (CIOS, 2009). This will not be a completely exhaustive list of the threats to validity. Specific research designs, subject populations, or research procedures may be vulnerable to other threats. What follows is a brief discussion of some of the most common threats. Any research design should be critically reviewed by the researcher, looking not only for the following threats to validity, but for any other way by which the action of the independent variable on the dependent variable might be confused with other factors, or by which the nature of the research may fail to generalize to the population being studied (CIOS, 2009).

Question 3 References Baker, M. J. 2003. The marketing book. 5th ed. Great Britain: Butterworth-Heinemann Blaike, N. 1991. Research design and methodology [online]. Pretoria: University of Pretoria. Available from:< http://upetd.up.ac.za/thesis/available/etd-08262004130507/unrestricted/05chapter5.pdf > [Accessed 29 April 2010] Boslaugh, S. No date. Secondary data sources for public health, A Practical Guide. Cambridge University Press. CIOS. 2009. Principles of research design [online]. Communication Institute for Online Scholarship. Available from:< www.cios.org/readbook/rmcs/ch13.pdf > [Accessed 15 April 2010] Coldwell, D. and Herbst, F. 2004. Business Research. Cape Town: Juta and Co Ltd Easterby-Smith, M., Thorpe, R. & Lowe, A. 2004. 2nd ed. Management research: an introduction. London: Sage Publication Ltd. Martz, W. 2010. Research design. [PowerPoint Presentation] 2 May 2010. MKTG 3710: Springs Massey, A. 2003. Research design and methodology [online]. Pretoria: University of Pretoria. Available from:< http://upetd.up.ac.za/thesis/available/etd-08262004130507/unrestricted/05chapter5.pdf > [Accessed 29 April 2010] Neill, J. 2003. Quantitative research designs: experimental, quasi-experimental and non-experimental. Available from: < http://wilderdom.com/OEcourses/PROFLIT/Class4 QuantitativeResearchDesigns.htm > [Accessed 29 April 2010] Newson-Smith, J. G. B. 2003. The marketing book. 5th ed. Great Britain: ButterworthHeinemann. Open University. OU Business School. 2003. Toolkit 1. Milton-Keynes: Open University. Palmer, A. 1998. Principles of services marketing. 2nd ed. New York: McGraw-Hill. Ramchander, P. 2004. Research design and methodology [online]. Pretoria: University of Pretoria. Available from:< http://upetd.up.ac.za/thesis/available/etd-08262004130507/unrestricted/05chapter5.pdf > [Accessed 29 April 2010]

Smith, S. E. 2010. What is a representative sampling? [online]. USA: Sparks Available from:< http://www.wisegeek.com/what-is-a-representative-sampling.htm> [Accessed 3 May 2010] ABC MRC, 2000. Commissioning and using qualitative research. United Kingdom: ABC MRC.

S-ar putea să vă placă și