This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
The research aims to identify management practices that can improve animal welfare and productivity, and to explore the economic and cost benefits of dairy cattle welfare in UK dairy farms.
This Chapter will be exploring the methods which were chosen for conducting the research, how it was conducted, data validity and reliability assessment, and finally identifying the limitations and ethical consideration while conducting the data collection for the entire research.
3.2 Research Philosophies:
According to Saunders, Lewis and Thornhill (2007), research philosophy adopted in the project contains important assumptions about the way in which the researchers view the world. These assumptions will be the base for the research strategy and the methods that the researchers choose as a part of that strategy. Adopted research philosophy will be influenced mainly by researchers' particular view of the relationship between knowledge and the process by which it is developed. According to Johnson and Clark (2006), the important issue for the researchers is to be able to reflect upon their philosophical choices and defend them in relation to the alternatives they could have adopted.
According to Saunders, Lewis and Thornhill (2007), the two major ways that can be used to understand research philosophy are ontology and epistemology. Each of these ways will influence the research process differently as each of them has important differences leads to different way of thinking in the research. Epistemology and ontology have many positions that will be explained and referenced for this piece of research:
It is a branch of philosophy that addresses the question of what make knowledge acceptable in a particular discipline. The central concern in epistemology context is weather to follow same principles, procedures and ethos of natural science when consider studying the social world. The natural science epistemology has three positions to study the knowledge which are: positivism, realism and interpretivism (Bryman and Bell, 2003).
It is an epistemological position that studies the social reality and beyond based on natural sciences methods. Credible data can be generated by observing a phenomenon and research strategy can be developed by collecting the credible data by using existing theory to develop hypothesis. A positivist approach has an important component that the research needs to be conducted, as far as possible, in a value free-way (Saunders, Lewis and Thornhill, 2007).
According to Bryman and Bell (2003), positivism approach combines both principles, inductive and deductive. In addition, the connection between theory and research is identified; the research is conducted to test theories to obtain material necessary for the development of law. The authors concluded that positivism should not be treated as synonymous with science and scientific.
According to Gill and Johnson (2002), researchers that follow positivism approach are more likely to use a highly structured research method to facilitate replication.
According to Saunders, Lewis and Thornhill (2007), realism is an epistemological position which relates to scientific enquiry and it is similar to positivism in assuming the scientific approach for the development of knowledge and in using social and natural science to same kind of data collection and explanation. The essence of realism is that objects exist in reality in independence of the human mind.
There are two major type of realism:
Empirical realism: this form indicates that realism can be understood by using appropriate methods. In addition, this form considered superficial because there is no recognised and enduring structure that can produce observable phenomena and events.
Critical realism: this form argues that people can experience the world as sensation of thing's image and not the thing itself directly. In addition, this form concludes that there are two steps to experience the world. Firstly, sensation of thing's image and the thing itself. Secondly, people can identify what they do not see by mental processing when the sensation conveyed by things meet their sense.
It is an epistemological contrasting term given as alternative to positivism. In this type of philosophy the researcher has to use empathetic stance and it is emphasis on understanding the different role of humans as social actors and it asserts that there is a difference of conducting research among people and objects. Interpretivism's origin comes from two intellectual traditions which are phenomenology and symbolic interactionism. Phenomenology refers to what way human use their feelings to make the world around us more sensible. In symbolic interactionism, there is an adjustment to our own meanings and actions by continual interpretation of the social world around us thus require from scientist to understand the subjective meaning of social action. It should be noticed that the two traditions, phenomenology and symbolic interactionism, are in the same direction in terms of antipathy for positivism and have in common an interpretative stance. Interpretivism approach is highly suitable in business and management research especially in organisational behaviour, marketing and human resources management fields (Bryman and Bell, 2003).
According to Bryman and Bell (2003), Ontology is a branch of philosophy that addresses the question concerned with the nature of social entities. The important consideration here to know whether social entities can be objective (known as objectivism) or they can be social construction comes from the perceptions and actions of social actors (known as constructionism). Objectivism is 'a philosophical system that implies that social phenomena confront us as external facts that are beyond our reach or influence'. From that, organization can be considered as tangible object and it has social order influencing people who are working within the organization to conform its requirements. This means that the organization has a reality affects its individual externally. An example of external influences are: organization's rules and regulation, work design and procedures needed for getting job done, mission statement and organization structure and hierarchy. Another important external factor that influences individual's ability to fulfil organization requirement is cultures and subcultures. Organization and its cultures have social entity as an external factor influencing the actor and as tangible reality of its own. Constructionism is a type of philosophy suggests that organization and culture can be learned and therefore confront social actors and maintain normal way of behaviour and relating. Organization built up by symbolic interactionism and research carried out in a psychiatric hospital and this considered best way to conceptualise as 'negotiated order'. In addition, this approach assumes that order within an organization is not pre-existing feature and it comes naturally from experiential learning. Same thing can be applied to the idea of culture. Culture is an emergent reality in a continually constructed and reconstructed process rather than external reality that act on and constraints people.
3.4 Research approaches
According to Bryman and bell (2003), there are two broad research approaches used for reasoning which are deductive and inductive and this project will use both approaches. Deductive approach used to explain how research is related to theory. This can be done by researcher who deduces a hypothesis on the basis of what is known about a particular field and theoretical considerations in relation to that field and then it is subjected to empirical scrutiny. The hypothesis is made up of concepts which need to be translated into researchable entities. In addition to build up the hypothesis, social scientist will determine how to collect the data related to the concepts that make up the hypothesis. Another approach for reasoning is inductive which implies that theory is the outcome of research. This means that there is an observation about certain phenomena and from this observation the researcher will extract a conclusion.
Deduction approach uses part of induction elements and the induction process uses modicum of deduction. Researcher needs to collect a data which is necessary to carry out a theoretical reflection, and the researcher may want further data collection to create the situation in which a theory will and will not hold (Bryman and Bell, 2003).
3.5 Research purpose:
According to Saunders, Lewis and Thornhill (2007), conducting a good research means covering and answering the objectives and the way these objectives have been set would result in descriptive and explanatory, or explanatory answers. This means that a research may have more than one purpose because the research objectives can be descriptive and explanatory. An exploratory study is looking to find what is happening about particular phenomena in new insights and to ask new questions in new light. This study is quiet helpful for the researcher to get more understanding of the problem if they unsure of the precise nature of the problem. In addition exploratory study can be conducting by three ways which are:
a search of the literature;
interviewing 'experts' in the subject;
Focus group interviews.
In addition, a good advantage of exploratory approach is flexibility and adaptability to change. This means that the researcher can change the direction of the research when new data appear and when new insights that occur to them (Saunders, Lewis and Thornhill, 2007).
Descriptive study is another type of research purposes that is looking to describe an accurate profile of persons, events or situations. It is important for researchers to have a clear understanding of the phenomena on which they are looking to collect data prior to the collection of the data (Saunders, Lewis and Thornhill, 2007).
Explanatory studies are those which show causal relationships between variables. These studies require an understanding of a situation or a problem in order to explain the relationships between variables. An example for explanatory studies is researcher who analyzed a quantitative data on manufacturing scrap rates and the analysis shows that there is a relationship between scrap rates and the age of the machine being used. Alternatively qualitative data can be collected to explain the reasons why customers of a company rarely pay their bills according to the prescribed payment terms (Saunders, Lewis and Thornhill, 2007).
3.6 Research strategies:
According to Saunders, Lewis and Thornhill (2007), there are two strategies can be used in the business researches which are quantitative and qualitative strategy. The researcher uses these strategies to determine which technique to be used in both data collection and data analysis procedure. For example, quantitative strategy is continually used as a synonym for any technique that generates or uses numerical data such as questionnaire which is considered as data collection technique or graphs which is considered as data analysis procedures. On the other hand, qualitative strategy continually used as a synonym for any technique that generates or uses non-numerical data such as an interview which is data collection technique or categorising data which is data analysis procedure. In addition qualitative strategy refers to data such pictures and video clips rather than words. This research will use qualitative strategy.
3.7 Research design:
According to Bryman and Bell (2003), 'research design provides a framework for the collection and analysis of data'. There are five types of research design can be used in business and management research:
Cross sectional design
Case study design
3.7.1 Experimental design:
This design is rarely used in business and management research because of there is a problem in controlling different variable when dealing with organization behaviour. One advantage of experimental research that it is tends to be very strong in terms of internal validity. In this type of research there are independent and dependent variables and it is important to determine whether these variables influence each others. In addition, each experimental group in the research has different types or levels of independent variables and this can be useful to establish how far differences between the groups are responsible for the variation in the level of the dependent variables. The central feature for any experiment is to make a comparison between results obtained by an experimental and those obtained by a control group (Bryman and Bell, 2003).
3.7.2 Cross sectional design:
This type of design is often called social survey; however, it also employed in many other research methods including structured observation, content analysis, official statistics and diaries. Cross sectional design collect data on more than one case and the reason for this is because the researcher may encounter variation in all variables they chosen, and they need larger number of population in samples and finally the researcher can easily distinguishes between cases. In addition, data collection is at a single point in time to collect a base of quantitative or quantifiable data in connection with two or more variable which then examined to discover pattern of relationship between variables. Thus this design is used by researcher who is interested in variation of people, organizations, nation states or whatever. However, cross sectional design rarely has internal validity that is found in most of experimental design because when researcher find a relationship between two variable, they cannot determine whether this donates causal relationship which is the feature of experimental research (Bryman and Bell, 2003).
3.7.3 Longitudinal design:
According to Bryman and Bell (2003), Researcher typically use this type of design to observe change in one item of business and management research over long period of time and it is usually exist as extension of social survey research such as questionnaire or structured interview research within cross sectional design. However, longitudinal design is better than cross sectional design in that it can deals with problem of ambiguity about the direction of casual influence. Sample within longitudinal design is surveyed on more than one occasion and according to that there are two type of longitudinal design: panel study and the cohort study. In panel study, the selected sample is national one and data collection is surveyed on at least two occasions and from different types of case such as individual and organization. In the cohort study data collection is focusing on an entire cohort of people or randomly selected sample of them who have common characteristic such same experience, and same education. However, cohort study is rarely used in business and management research. These two types of longitudinal design have common problem: firstly, researches such employee job changes, companies going out of business have sample attrition. Secondly, longitudinal research has poorly planned in terms of data collection. Longitudinal design has vertical and horizontal level of analysis and there is an interconnection between those levels through time.
3.7.4 Case study design:
The term case study is associated with a location such as country, person, workplace, organization and company. Case study design is characterised by detailed and intensive examination of one case and there is an investigation for the complexity and particular nature of the case in question. This means that there will be deep understanding of research context and processes being used (Bryman and Bell, 2003).
According to Saunders, Lewis and Thornhill (2007), case study design is mostly employed in explanatory and exploratory research, because it is able to obtain answers to the questions: why, what and how. Researcher who uses case study design is more likely need multiple data collection technique such as structured interview, observation, documentary analysis and questionnaire. Consequently, this type of research design is mostly uses triangulate multiple sources of data. Triangulation defined as the collection using different techniques within one study to ensure that the data provide essential knowledge that the researcher is looking for.
According to Yin (2003), there are four type of case study design and there is an opportunity to distinguish between them based upon two discrete dimensions:
Single case Vs multiple case;
Holistic case Vs embedded case.
Single case study research is often conducted for a critical case or unique case. This means that single case study can provide an opportunity to study a phenomena that few researchers have investigated before. On the other hand, multiple case studies are used to observe whether the results of the first case are obtained in other cases and whether there is need to generalize from these results (Yin, 2003).
For holistic and embedded case studies, the concern is on the unit of analysis. For example, when a researcher choose an organization as a case to be studied as whole then they treating the organization as a holistic case study. Conversely, if the concern is to examine subunits within an organization such as department or work group, then the way used to select these subunits called an embedded case study (Yin, 2003).
3.7.5 Comparative design:
This type of research design involves using two or more contrasting cases that are more or less identical methods. Using this type of research design will lead to better understanding of social phenomena because it is compared in relation to two or more meaningfully contrasting cases or situation. Comparative design can use either quantitative or qualitative data collection. In the quantitative data collection technique, there are two or more cases being used and cross-sectional design is the format for the data collected from each case. The logic of using comparative design is to be employed to a variety of cases to use different number of levels of analysis. Researcher in the qualitative data collection is using multiple-case study and in the case of business and management research, it is common to use this design that takes two or more organizations as cases for comparison. Multiple-cases are selected on the basis of types of organization such as successful and unsuccessful and the operation of the organization in certain commercial sector. Thus comparative design is used to find out the characteristics of each organization as first step for theoretical reflections about contrasting findings (Bryman and Bell, 2003).
3.8 Research limitation and generalization:
According to Jewell, Hardie (2009), there are four aspects to consider for a research to be reliable and valuable which are validity, reliability, generalization and transparent. These aspects are also used to evaluate business and management research.
According to Brymna and Bell (2003), this aspect is concerned about whether a concept is really measured by its specific measure. For example, from my research there is SCC that is used as indicator for udder health and this measure is helpful to improve udder health before problem occur by improving milking hygiene. Validity can be established by several ways: face validity, concurrent validity and vergent validity.
According to Bryman and Bell (2003), 'reliability refers to the consistency of a measure of a concept'. There are three essential factors used to determine whether a measure of a concept is reliable: Stability, internal reliability and inter-observer consistency. In the stability factor, researcher is concerned about whether a measure is stable over period of time and whether the results relating to the measure do not change over time. For example measuring mastitis prevalence in different season will be different because the environment surrounding cattle is not the same among seasons. Internal reliability means that there are multiple items need to be measured by applying reliability on it. This means that internal reliability measures whether the scores of one item is related to their score on the other item. Inter-observer consistency is concerned about number of observers involved in the recording activities of observations and translation of data and whether there is an in-consistency in the observers' decisions.
The researcher in the quantitative research is normally aware whether their research finding can be used beyond the research context as general rules and guidelines and whether it can be applied it in another research. In this aspect the researcher is seeking to create a representative sample by using probability sampling because they need this sample for reaching generalization. This means that researcher cannot generalize beyond the sample population. Two points the researcher should be aware about generalization. Firstly, generalization of findings has a limit and the researcher need to be careful about applying these findings in another context. Secondly, generalization of finding is quite strong in quantitative researches which use cross-sectional and longitudinal designs (Bryman and Bell, 2003).
I will conduct this research and I will feel free to show it to anyone who is interesting in this topic. In addition, the achievements in this research in terms of data analysis and how it leads to the conclusion will show the reader the result of my research and will show them how the works has been carried out and the relationship between the data analysis and conclusions (Jewell, Hardie, 2009:94).
3.9 Ethical considerations:
Research ethics means what ways the researcher will use to formulate, clarify and design the research subject, and how the researcher can get access, collect, process, analyze the data and write the research outcomes taking into account more ethical consideration (Saunders, Lewis and Thornhill, 2007).
This research is conducted according to all the ethical rules and guidelines of Coventry University Business, Environment and Society and throughout this research I agree to follow these ethical guidelines and it is exist in the following website: http://www.coventry.ac.uk/researchnet/ethics/statement. As I mentioned before, this research is only using secondary data and these data can be accessed easily and it is available any time for me and I will only use it for the purpose of this research.
3.10 Data collection methods:
When researchers starting to answer their research questions or meet their research objectives, they mostly looking to collect new data which is primary data in order to achieve the purpose of the research. On the other hand, few researchers consider secondary data in which they reanalysing data that have already been collected for some other to know what already has been done in the research area. Secondary data include published journal, articles, books, raw data, and reports from quality daily newspapers, and survey and published official statistics from government departments which cover social, demographic and economic topics (Saunders, Lewis and Thornhill, 2007).
This research will only consider secondary data in order to answer the designed questions and meet the planned objectives. The reasons selecting only secondary data are: any research in animal farming takes long time to get the results, beside I do not have an access to any farms in the UK.
3.10.1 Secondary data:
According to Saunders, Lewis and Thornhill (2007), secondary data can be obtained in both forms: quantitative and qualitative which all can be used essentially in both descriptive and explanatory research. In business and management research, raw data and compiled data are used as part of a case study or survey research strategy. Secondary data are classified into three subgroups which are: documentary data, survey-based data, and those compiled from multiple sources. The following diagram show categories of secondary data with examples:
Documentary Multiple source Survey
Written materials Non-written area based time series Censuses
as personnel or
Types of secondary data (Saunders, Lewis and Thornhill, 2007)
3.10.2 Secondary data resources for this research:
The focus in secondary data collection of economics of dairy cattle welfare was on recently published articles and journals that can be accessed through the internet websites, university electronic database and books which included:
Literature review of most recent published journal that is related to 'diseases prevalence under organic (more welfare) management practices' from science direct, EBSCO, American Dairy Science Association, Elsevier, Journal of animal science and DEFRA.
Literature review of most recent published book about 'animal welfare economics' from Coventry University electronic book and Google scholar.
Literature review of most recent published journal and articles that are related to 'diseases cost' in farm animal from science direct, EBSCO, American Dairy Science Association, Elsevier, Journal of animal science and Journal of dairy science.
Literature review of most recent published journal and articles that are related to 'assessment of dairy cattle welfare' from science direct, EBSCO, American Dairy Science Association, Elsevier, Journal of animal science, DEFRA and Journal of dairy science.
This research is being conducted after having a good quality of secondary data and it depends largely on its sources. The quality of the secondary data will be high if its resources come from government database, academic database, and research organization and association. The reason for high quality of secondary data from these databases is because its articles and journals are reviewed by academic peers. For example, the Journal of Dairy Science (JDS) has good rank between dairy research journals in the world and in 2009 this journal ranked by ISI Journal Citation Reports as number one out of 45 journals in the Agriculture, Dairy and Animal Science category (Fass, 2010).
The journals are being used in this research are relevant because it answered the research question and meet the objectives. For example, Journal of dairy science, Journal of animal science, American Dairy Science Association, EBSCO and Science Direct are used to answer the planned objectives. In addition other websites are used to identify some concepts and theories such Google scholar, E-book from Coventry university databases.