Business research methods - Essay Example

? BUSINESS RESEARCH METHODS …………………. College/ ………….. …………. “The problem definition stage is perhaps more criticalin the research process than the problem solution stage". Discuss this statement in depth. Research Process, which is one of the most important aspect in business research, essentially involves several important stages such as preliminary steps, problem definition, management dilemma, primary or secondary data collection, literature reviewing, data analysis, presentation of findings and problem solution etc. Out of these research process stages, the problem definition stage is perhaps most important one as effectiveness and successful completion of all other stages are largely depending on problem definition stage. Some of the stages in the research process are more critical than others. For instance, problem definition is most important step because it determines how to set the direction for the remainder of the steps that follow (Neelankavil, 2007, p. 57). The problem to be solved in a research or the issue to be identified must be clearly defined so that more appropriate solution to the same problem can be developed and this illustrates why problem definition stage is more important in the research process. What is Problem Definition and Why it is Important? Sekaran (2003, p. 70) defined Problem Definition, or often termed as Problem Statement, as a clear, precise and succinct statement or description of the question or issue that the organization need to investigate with the goal of finding an answer or solution to the same problem. As far as an organization is concerned, problem definition may relate to an existing business problem as the manager is looking for its solution, or to an issue that is likely to arise in the future and the manager feels to develop a solution for it or areas where some theoretical and conceptual clarity is required. When it comes to a business organization, there can be a serious gap between actual and desired ideal states and therefore the management requires to improve the existing situation to solve the issue the organization faces. The research problem is to be defined because a research or more pacifically an investigation in to different functional areas in the organization such as sales, purchase, profit, inventory handling etc can be successfully done only if the problem is clearly defined. Malhotra (2008, p. 38) emphasized that problem definition in research process is the most important step because it involves stating the general problems and identifying various components of the research itself. If problem is clearly defined, significant components that can be made use in the research also can be identified. Therefore, it is very clear that the research can be well-designed and conducted properly only if the basic research question is properly defined. As Malhotra (2008, p. 38) noted, none in all the tasks involved in a research project is more vital to the ultimate fulfillment of a client’s needs and requirements than a proper problem definition. Time, efforts and money spent on any research process can be a mere waste if the problem is ill-defined. An example can illustrate how important is problem definition in a research process. An organization that experiences less- morale among its employees may need to investigate why its people are less morale and how it can be resolved. In order to investigate this issue, the problem should be clearly defined. If the researcher defined the problem as less efficiency among the employees, or less productivity among the workers or decreased commitment among managers, the problem is not clearly defined and the solution the research may bring would be of no use to the organization. Following are some of the main reasons why problem definition is more important than solution and any other steps in a research. A well defined problem helps the researcher plan all other research steps to be conducted properly, All other steps like data gathering, data analysis, and problem solution follow problem definition. The researcher collects primary data purely based on the problem to be investigated. There should be a positive and more direct relation between data to be gathered and the problem being defined. Therefore, if the problem is ill-defined, the researcher would probably collect data that won’t be useful to develop effective problem solution. The researcher reviews relevant literatures. Literatures become relevant only if they relate to the problem being defined. If the problem is not properly defined, the literatures to be reviewed will be irrelevant and the research itself becomes useless to the business organization. Problem solution is developed based on problem being defined. But, if the problem is not defined properly, the solution that the researcher developed wont be an effective solution to the exact problem the organization experiences. It is advantageous to develop a directional hypothesis whenever we are sure of the predicted direction. How will you justify this statement? ? Bring examples to support your answer. A hypothesis is logically surmised relation between two or more variables that are presented in statement forms as part of the research. If the two or more variables have been identified and the relationship among them has been established and clearly described, then the researcher is required to examine whether the relationship that has been theorized is true or not. Either qualitative analysis tools or quantitative analysis tools are used to test the relationship between variables in a research. The researcher is in a position to state the relationship between the variables and he may use different terms like positive, negative, more than, less than etc and these hypothesis is called directional hypothesis. It is because, the direction of the relationship between the variables is clearly mentioned. If the researcher has hypothesized the relationship between the variables, but hasn’t indicated the direction of the relationship between the variables, then it is called non-directional hypotheses (Sekaran, 2003, p. 104). If the researcher is sure of the predicted direction, then directional hypothesis is more advantageous because it not only establishes a relation between the research-variables, but also indicates how the relation is, in relation to whether it is more than or less than, positive or negative, higher or greater than etc. In a research, some time the researcher will be sure of the predicted direction. It is very similar to the concept of forming assumption in the beginning of the research. More often, the researcher will have a specific assumption and he will target his research process to conclude his research findings to that specific direction. If he is sure about the direction, then directional hypothesis is more appropriate as well as advantageous to him. For instance, when the researcher assumes that group A is more effective than group B, and he progresses his research accordingly, a directional hypothesis will be more appropriate for him. Directional hypothesis states the nature of the interaction between two or more variables and these types of hypotheses are developed from theoretical statements, or findings of previous studies or any previous experiences etc (Gravetter and Wallnau, 2008, p. 248). A non-directional hypothesis states that there is relationship between the variables or there is no relationship between the variables, but does not mention how the relationship is. Whereas directional hypothesis gives an amount of predicted relationship and therefore when the researcher is sure of the predicted direction, directional hypothesis more advantageous to him (Quinton and Smallbone 2006, p. 51). As Sekaran (2003, p. 105) noted, non-directional hypothesis is formulated may be because of the relationship or because of that the differences have never been previously explored and therefore there is no basis for indicating the direction. Non-directional hypothesis is also appropriate if there has been conflicting findings in previous research on the variables. More specifically, while reviewing the literatures, the researcher has found some researches having concluded that there is a positive relationship and some others having found a negative relationship. In this case, the researcher may not indicate a specific direction and he may therefore use non-directional hypothesis. Examples for Directional hypothesis, and its appropriateness when the researcher is sure of the predicted direction For example, “When employees experience more stress on job, they more likely to search for another job” (Hair, Celsi and Money, 2011, p. 137). This is purely a directional hypothesis. While conducting the research, the researcher has found that there is close and rather positive relationship between job-stress and employee turn over. The researcher has concluded the above finding may be from one of the followings: While reviewing the literatures, the researcher has found that most researchers have already concluded the same, and thus the researcher was sure of his direction. While conducting the research, he found two different opinions. Some researchers argued there is no such positive relation between job stress and employee turn over. Some other researchers argued its opposite. But, when he analyzed the data he gathered and made a qualitative or quantitative analysis through logical reasoning etc, he found that job stress causes employees to search for another job and it thus causes employee turnover. If the hypothesis is “There is a relationship between the stress experienced on the job and the likelihood that an employee will search another job”, it is non-directional hypothesis, because this statement does not give specific direction of the relation between the two variables. This type of hypothesis is not appropriate if the researcher is sure of the direction. a- Situation where a manager would use pretest and posttest experimental design only Pretest is not only an established practice for discovering errors in questionnaire and other research programs, but also is very useful for training then research team. Pretesting helps the researcher discover participants’ reactions to the questions. It also helps his find where there is repetitiveness or redundancy etc (Cooper and Schindler, 2003, p. 389). In order to test a treatment effectiveness, a pretest is first given with a focus of measuring the dependant variable and a second test is also given which is termed as post-test, so that if there is any difference between theses two tests, the difference and the related score can be attributed to the treatment. Situation No- 1= A manager will be more likely to use pretest and post-test experimental design in a situation where the manager expects that a better workplace environment causes increase in job satisfaction. When the manager administers a pretest asking employees’ job satisfaction with the current workplace, people would be some time more sensitized to the issue of job satisfaction. When employees are working in a good workplace environment and a post test is given to them asking the level of job satisfaction, employee may react quite differently. So, the manager should use pre-test and post-test experimental design when he would like to know how people react to two both pre-test and post-test and how these two reactions or answers differ from each other. Situation= 2- The Manager needs to find out how sales and advertising are interrelated. In order to find out the relationship between increased advertising and more sales, the manager can design pre-test to the customers asking some specific questions and later on he can administer a post-test as well to measure the difference between two answering. Situation- 3- Similarly, the manager can administer both pre-test and post test to check how shareholders differently react for questions given to them regarding return on investment. b- Situation where the design would consist of an experimental group and a control group, both of which would be given a pretest and a posttest Often, both pretest and posttest will be exposed to experimental group and control group. Experimental group is exposed to a treatment whereas control group is not so. The net effects of the treatment can be found by comparing the results or differences between control group and experimental group. Both these groups are given pretest and posttest on a random basis. When a particular subject is pretested and exposed to the experimental condition, and then tested again, the difference between the pretest and posttest will be measured and considered as experimental stimulus (Monette, Sullivan and DeJong, 2010, p. 268). The pretest and posttest control group design is very strong design because both groups namely experimental and control groups are randomly assigned (R) and experience a pretest (O1) and a posttest (O2). This design is a control system for all threats to internal validity. If there are pre-intervention differences between control and experimental groups, this difference also can be controlled (Cottrell and McKenzie, 2011, p. 186). When both pretest and posttest are exposed to control as well as experimental group, there is possibly a weakness that pretest-intervention interaction can limit external validity. A manager may use both pretest and posttest being exposed to both control group as well as experiential group in cases like measuring the impact of training on employees, assessing skill development and measuring the impact of technology advancement etc (Sekaran, 2003, p. 161). Situation- 1- Measuring the impact of training on employees. The manager will be able to expose pretest and posttest to both control group and experimental group in order to find out the differences between the pretest and posttest and thus to find how effective were the trainings given to them or how do employees react about the training and its effectiveness. Situation= 2- For measuring skill development among the employees. The manager may use both pretest and posttest to be exposed to both control and experimental group to find the differences between these two tests and then to assess the skill development among the employees. Situation- 3- Measuring the impact of technology advancements. The manager may also like to identify what specific impacts have had technology advancement on employees. To find out this as well, the manager can expose both pretest and posttest to control and experimental groups and then to find out the differences between these two tests. C- Situations where a full-fledged Solomon Four Group Design would be contemplated. The Solomon Four-Group Design is one of the most robust and sophisticated search designs. It is helpful to control factors that are associated with both internal and external validity. By using Solomon Four-Group Design, the researcher will be able to determine the effects of treatment in relation to a control group and to recognize the pretest on dependant variables. This is also helpful to determine and explain any interactions between the pretest and the treatment (Cottrell and McKenzie, 2011, p. 187). Solomon Four-Group Design sets up two experimental groups and two control groups for an experiment and this gives more confidence in internal validity in experimental designs. One experimental group and one control group can be exposed to both pretest and posttest and the other two groups can be given only the posttest. As the Solomon Four-Group design is most comprehensive design in experimental design tools, it is more appropriate when there are complicated situations that a manager needs to investigate or where he requires gaining very accurate results. Following are three situations where Solomon Four-Group Design can be best used. Situation- 1- An organization requires to check effectiveness of employee development after a development program has been arranged to them. The manager may create two control and two experimental groups and they both can be exposed with pretest and posttest in order to find the answer. Situation-2- The organization may use Solomon Four-group Design for assessing proclivity and manufacturing after installing an advanced technology or after an infrastructure has been developed. Situation-3- For assessing the effectiveness of employees’ performance, Solomon four-group design can be used. References Cottrell, RR and McKenzie, JF 2011, Health promotion and education research methods: using the five-chapter thesis/dissertation model, Revised second edition, Jones & Bartlett Learning Gravetter, FJ and Wallnau, LB 2008, Essentials of statistics for the behavioral sciences, Sixth Edition, Cengage Learning Hair, JF, Celsi, MW and Money, AH 2011, Essentials of Business Research Methods, Second revised edition, M.E Sharpe Malhotra, NK 2008, Marketing Research: An Applied Orientation, Fifth edition, Pearson Education India Monette, DR, Sullivan, TJ and DeJong, CR 2010, Applied Social Research: A Tool for the Human Services, Eighth edition, Cengage Learning Neelankavil, JP, 2007, International business research, Illustrated edition, M.E. Sharpe Quinton, S and Smallbone, T 2006, Postgraduate research in business: a critical guide, Illustrated edition, SAGE Sekaran, U 2003, Research Methods for Business, A Skill building Approach, Fourth edition, John Wiley and Sons Read More