Experimental versus Non-Experimental Research - Assignment Example

Experimental versus Non-Experimental Research Contents Contents 1 INTRODUCTION 2 QUANTITATIVE RESEARCH 2 Experimental Research 4 Non-Experimental Research 5 COMPARATIVE ANALYSIS OF APPROACHES 7 Sampling 8 Validity 9 Reliability 9 Bias 10 EMPLOYABILITY OF MILITARY VETERANS IN THE FEDERAL GOVERNMENT 10 Research Methodology 11 Sampling 12 Validity 12 Reliability 13 Bias 13 RECOMMENDATION 13 Bibliography 15 INTRODUCTION There are different methods and approaches for conducting research into a given social phenomenon. This includes a wide range of approaches and methodologies which can be applied to a given situation and context. Quantitative research is a unique and distinct approach and method for studying and analyzing a given phenomenon through the measurement of some aspects and units of a given situation. Two approaches and methods for conducting quantitative research are the experimental and non-experimental quantitative research. These two different approaches have distinct rules and conventions that guide the methods of collecting and analyzing data. They produce different research results and findings and this helps to draw conclusions and different responses to given studies. The purpose of this study is to compare and contrast experimental and non-experimental quantitative research approaches. In order to complete this research, the following objectives are going to be explored: 1. A critical review of the differences between experimental and non-experimental quantitative research; 2. Evaluation of the two approaches in relation to fundamental yardsticks for research; 3. Application of the two approaches to a real-life situation and context of research. QUANTITATIVE RESEARCH Quantitative research is a systematic empirical investigation of a given social phenomenon and it employs various statistical and mathematical techniques to define the variables and measure them in order to draw a conclusion (Nykiel, 2012). Quantitative research is an empirical research that seeks to use mathematical and other statistical methods to evaluate a given research topic. Quantitative research involves the development of an appropriate method and an appropriate approach to study a given phenomenon. Measurement is a distinct and unique aspect of quantitative research. This is because quantitative research involves finding ways and means to measure a given set of data objectively in order to draw a conclusion on the research variables (Carter & Thomas, 2010). This is done by formulating a hypothesis and evaluating the hypothesis through the observation of specific elements and aspects of interest (Macnee & McCabe, 2012). Therefore, there is the need for some kind of variables that can be numerically measured and analyzed in order to draw a conclusion on the subjects at hand and the matter being studied. Therefore, quantitative research is an empirical observation and a mathematical expression of the relationship between the variables (Lodico, Spaulding, & Voegtle, 2012). The main approach that is used in quantitative research is to define a hypothesis and operationalize the research. Operationalization of a research is done through the definition of variables and putting the variables in a way and manner in which they can be measured and matched against each other (Lodico, Spaulding, & Voegtle, 2012). The core finding of a quantitative research involves summarizing numerical results and uses it to draw conclusions on the subject at hand (Lodico, Spaulding, & Voegtle, 2012). Quantitative research methods also involves the use of various statistical techniques to summarize a volume of data from different sources in a given study in order to find a conclusion or an aggregation of different solutions to a given situation or problem (Muijs, 2011). Experimental and non-experimental research are two forms of quantitative research and it utilizes various techniques to gather data to study a given phenomenon. Experimental research is meant to establish the cause-and-effect of a given relationship involving variables (Cottrell & McKenzie, 2012). The hypothesis of a given study or research is tested through an experimental mode and this is a system whereby the variables are put in their natural state in order to observe and analyze them. Experimental Research Experimental research is a system where the researcher creates an atmosphere where the two variables are tested against each other in order to identify the main relationship that exists between the variables (Cottrell & McKenzie, 2012). Hence, experimental research involves the manipulation of one variable under its natural circumstances in a given control group which is a population that falls within the mainstream larger population (Gravetter & Lorzano, 2013). This implies that an experimental research is conducted through the selection of a given control group which involves a specific sample that is selected from the broader population in order to study and analyze (Creswell, 2009). The control group is subjected to various treatments in order to ascertain their responses to the situations and circumstances of the study. Therefore, the researcher is able to see the cause and effects on the control group directly during the research. Experimental research provides close association with the respondents and it gives room for creating and recreating circumstances and situations within which a research is conducted (Chandra & Sharma, 2012). Experimental research has three main variables and components as presented by Chandra and Sharma and they include: 1. Achieving Isolation: The respondents in the research are isolated from the larger population and society. And this helps the researcher to study the control group more closely and more critically; 2. Achieving Changes in Magnitude: There must be an opportunity for the researcher to make changes and manipulate the research. This will enable the researcher to observe the changes in the responses that will be exhibited by the members in the control group. 3. Quantitative Evaluation: It is important and vital for the changes observed in the manipulation of the study to be recorded and documented. This is important because the ability to measure the changes in the responses and the reaction of the control group is important and vital for the researcher to document the changes that will occur. Therefore, there is the need to find a measurement scale and process in order to come up with conclusions on how the study has progressed. This involves the identification of statistical methods and statistical processes. There must be a dependent variable and independent variable in order to undertake experimental research because these forms two classes of variables that will be studied in an experimental research process. The independent variable is the variable that cannot be changed but remains constant throughout the studies but the dependent variable is the variable that changes with the manipulation of the control group (Mangal & Mangal, 2012). There are several types of experimental research. This include quasi-experimental research, true research, post matched and pre-test random sampling and other approaches that can be used to conduct such studies (Creswell, 2009). A quasi-experimental research contains elements of experimental research as well as other approaches to research like non-experimental research (Carter & Thomas, 2010). True experimental research uses pure experimental research and nothing else (Banks, 1984). Matching is a statistical technique that is used by a researcher to evaluate the relationship between variables and this could be done before or after the experimental study (Nykiel, 2012). These methods and approaches work together to define the terrain of experimental research. Non-Experimental Research On a simple level, a non-experimental research is a form of research that does not utilize experimental methods. The fundamental difference lies in the fact that non-experimental research does not involve the level of control and manipulation that exists in experimental research (Burney & White, 2009). “Non-experimental research is generally present-oriented. It attempts to describe what exists. Variables are not deliberately manipulated nor the setting controlled. The analysis of the data often leads to the formulation of a hypothesis that can then be tested experimentally.” (Brockopp & Hasting-Tolsma, 2012, p. 232). This implies that non-experimental research is more related in studying the realities of what is happening at a given point in time. This is because it involves the study of general trends and processes in a given phenomenon and there is the general tendency to evaluate the factors as they are without trying to modify the setting. The main uniqueness of non-experimental research is the fact that it is not as robust and as critical as experimental research. This is because it does not zoom in to take a closer look at a given situation or sets of variables (Creswell, 2009). It seeks to look at general trends and as a result of this, non-experimental research culminates in conclusions that are less specific than those of experimental research. Non-experimental research is fundamentally involved in describing preexisting groups in order to determine the obvious linkages and relationships that exist between variables (Lodico, Spaulding, & Voegtle, 2012). Due to the nature of non-experimental research, it culminates in a number of possible conclusions that are presented and put forward as hypothesis which requires further studies and further analysis in order to come up with a more specific and a more direct conclusion that can be used to build theories in the future. Therefore, non-experimental research can be seen as a naturalistic observation of a given set of variables (Alves, Gopferich, & Mees, 2012). However, this belongs to the class of quantitative research because it utilizes different methods and approaches that are involve numerical and mathematical measurements which are steeped in statistics. And they are appropriate for exploring the attitudes and processes of a given situation in order to get the preliminary process and procedures related to a given concept or idea. This forms the foundation and basis for the operationalization of the research in order to conduct deeper and more insightful research into a given situation and phenomenon. Generally, non-experimental research is about studying the parent and universal population of the research (Shrader-Frachette, 2012). This is because non-experimental research is a major approach and method of analyzing and evaluating a large and general population. This provides an insight into the general trends and the general processes in the population and this helps the researcher to conduct detailed and further study into the phenomenon in order to draw logical conclusions into the topic and also find trends in specific populations. COMPARATIVE ANALYSIS OF APPROACHES So far, this study has identified that non-experimental research is more of a generalist approach and method to study a given phenomenon. In the quantitative research context, non-experimental research can be conducted by studying and analyzing the broad terms and processes of the universal process. This will be an impetus for the identification of specific trends and processes in the larger and broader population. On the other hand, there is an inherent limitation in the utilization of non-experimental research in studying specific aspects of the population. This is because it is not robust or critical enough to examine and evaluate specific aspects and processes within a given set of people in the wider population (Creswell, 2009). In these situations, there is the need for the researcher to employ an experimental research in order to find the true changes and modifications and responses to specific trends in a given research. This is because experimental research enables the researcher to create situations and contexts within which a given set of variables can be manipulated and evaluated in order to find solutions and answers to important trends and processes in the research or study. However, it is imperative to examine and evaluate the two approaches on the basis of specific yardsticks. This will involve the examining and identification of specific elements and aspects of these approaches to research in order to find their individual strengths and weaknesses in order to find out their appropriateness for specific research processes. Sampling In most cases, a researcher is not capable of studying 100% of the entire population of the research (Creswell, 2009). Therefore, it is appropriate for the researcher to find a section of the research that is somewhat representative to the wider population because it enables the researcher to make findings about a section of the population that can be appropriately generalized to explain trends and processes for the whole population. Non-experimental research involves the use of probability and non-probability techniques because it is used to examine and evaluate the wider elements and aspects of the total population (Daniel, 2011). This is because there could be different processes and different approaches that will be used to evaluate and analyze a larger population and a larger process. Thus, according to the appropriateness of the study, the researcher might either use probability or non-probabilities to sample the entire population. On the other hand, experimental research involves the use of statistical methods and statistical approaches in selecting the samples (Daniel, 2011). This is because the experimental research involves a more specific evaluation and analysis of specific section and process of dealing with a given population. And since it is about the studying of cause and effect, there is the need for the researcher to use statistical approaches to ensure that the results of the findings will have to be more representative of different sections and processes of the research. Validity Validity is an attempt to establish logical soundness in a given research. It is about attaining the right kind of response and the proper trend and logical views of things in order to find the proper conclusions of a given research. Validity is about the ability to meet the criterion or standards of expectations in the research (Carmines, 2012; Imai, Tingley, & Yamamoto, 2013). Seymour Banks on the other hand uses a realistic view on explaining how validity can be maintained in a given research. He argues that the design process of a given research, whether experimental or non-experimental will define the kind of research approach that can maintain validity (Banks, 1984). Intervention tests are important in examining and evaluating the processes and the methodologies that will be used in seeking the practical interest of the groups in the study (Calder, Phillips, & Tybout, 1983). Validity is about seeking to develop appropriate frameworks and conceptual processes for measuring and providing the right conclusions (MacKenzie, Podsakoff, & Podsakoff, 2011). This shows that the nature of the two approaches – experimental and non-experimental will demand different approaches and different processes to deal with issues and matters. Reliability Reliability is about the results that are employed by the two approaches and in the general sense, experimental research identifies more specific results and hence more reliable findings and theories (Duvendack & Palmer-Jones, 2013). Reliability refers to the kind of approach that is used in either types of research. To ensure that a research or study is realistic and positive to the researcher, there is the need for the measurement instruments and theory on which it is based to be proper and representative of the study. Experimental studies have a much more appropriate approach to ensuring reliability. However, in cases where the approach is wrongly formulated, experimental studies could provide divergent results that might be unrealistic. On the other hand, non-experimental research could provide less reliable results since it is less specific. However, the possibility of getting an accurate finding is generally higher. Bias Biases are often found to be part of the sentiments of the researcher and this can affect the design and conduct of the research (Shrader-Frachette, 2012). This is because the researcher has a high degree of autonomy in defining the kind of research design and the lack of objectivity and fairness can lead to major problems and issues with bias (Schonlau, 2009). In cases where a research suffers a setback linked to the prescriptive nature of the study, experimental studies are more likely to be misleading than non-experimental data because the design and processes of experimental data depends more on the researcher’s judgment. This is less of an issue with non-experimental research. EMPLOYABILITY OF MILITARY VETERANS IN THE FEDERAL GOVERNMENT Empirical studies indicate that there is little difference between the experimental and non-experimental approaches in studying elements and aspects of the public sector (Greenurg, Michalopoulos, & Robins, 2006). However, in the case of hiring military veterans, there is the need for the design of an appropriate research method and an approach to study a number of sections of the veteran process and an identification of what their challenges in hiring are. First of all, to conduct a more thorough and more reasonable research, there is the need to define the actual research question, which should be the enquiry into the factors influencing the employability of military veterans in the federal government. First of all there is the need to identify an appropriate method and an appropriate approach for the design of research in order to study the issue in a more specific way and manner. Theoretically, it must be established that in conducting a research into hiring a competent and well functioning staff, there is the need to be familiar with the personnel rules and hiring policies of the organization in question (Stouthamer-Loeber & Van Kammen, 1995). This implies that there is the need to draw a framework into the federal process and the federal network within which employment is granted and guaranteed. Research Methodology There are two possible approaches to the study of this phenomenon. One approach will be to use a generalized method to examine and evaluate the main factors and situations that affect the hiring practices of the federal entities. This will involve evaluating the main limits and the processes that stand in the way of hiring people. This approach can be done by evaluating the general trends of the hiring environment and why military veterans are excluded. This approach can use non-experimental methods and it can be done by examining the main factors that exists in the work place that keep military veterans out. On the other hand, a more specific study can be conducted by examining a number of veterans who have faced challenges in acquiring jobs in the federal agencies. This will mean examining the main factors and matters that went against these people and this can be done by studying what barriers and factors prevented these veterans from gaining a job of their choice. The findings can be subjected to various forms of manipulations to examine what actually happens in the hiring process and why veterans are actually kept out. Sampling For the non-experimental method which will involve an evaluation of the hiring process of the federal government, there will be a probability approach which will examine the situations and settings within the hiring process that can affect a military veteran in the quest to get a job with the federal authorities. This will involve zooming in on circumstances and matters in which veterans may be affected by the general process of the hiring process. On the other hand, statistical methods can be used to examine a number of cases involving veterans and how these veterans represent the larger community of veterans. This can be skewed skillfully in order to find the right selection of veterans which will represent the entire population of veterans in the whole United States or a section of the United States. This will provide a more critical review of findings in order to deal with the situation and find specific explanations that can be varied to provide more thorough and more critical conclusions on the matter. Validity The validity of the research that will be conducted through non-experimental study is quite questionable. This is because examining the position of veterans as a part of the larger population of applicants who apply to federal position might be difficult to ascertain. This is because it is difficult to find out how veterans will react to this and there might be other external factors that could not be found. However, this approach ensures validity in a way and manner that involves the fact that it helps to define what exists in the federal employment agencies that are for or against veterans. Validity seems to be higher if we use an experimental approach of studying veterans and how they fare with the federal employment agencies. This is because there is a more specific study of how the two variables work together and its impacts and implications. Reliability Experimental studies are about the cause and effect of military veteran applicants and the federal employment agencies. This shows a more specific and a more reliable finding that will be able to bring to light the facts and proper information about how veterans fare. This will be done in an absolute manner as opposed to a relativist situation. Non-Experimental research will provide a much more generalist idea and insight into the position of veterans in the employment cycle and system. This is opposed to the more specific results that is expected from an experimental study. This is because the experimental study provides more direct information. However, the non-experimental study will provide a relativist answer and set of solutions to the situation and this might be less explanative of the situation at hand. Bias Bias in sampling could affect the experimental research process significantly. This is because the findings are expected to be more absolute and more explanative. Thus, the method of searching for samples and designing the research can cover a given area in-depth. And this could be to the neglect of another approach and another method. On the other hand, a non-experimental research could provide a general insight into the universal problems and challenges faced by most veterans. This could prove to be vague but the risk of bias could be significantly offset by the fact that the scope of the study will be wide and focus on many general trends and processes. RECOMMENDATION It is recommended that an experimental approach should be used to study a sample of veterans who have applied to be hired by the federal agencies over a given period of time. This is more specific and it studies the main framework of what veterans go through in an absolute sense and this can lead to some kind of solutions to the issue. Thus, this approach is more constructive and result oriented. A non-experimental study would be more appropriate if the researchers want a general or relativist approach to the subject. This will give a broad scope and broad framework on what veterans go through. However, this approach is not recommended for this study. In an ideal situation though, a researcher can use a non-experimental method to deduce hypothesis. This will guide the researcher so that s/he designs a more specific experimental research that will address issues more critically. Bibliography Alves, F., Gopferich, S., & Mees, I. M. (2012). New Approaches in Translation Process Research. Frederiksberg C: Samfundslitteratur. Banks, S. (1984). Designing Market Research to Increase Validity. Journal of Marketing , 32 - 40. Brockopp, D. Y., & Hasting-Tolsma, M. T. (2012). Fundamentals of Nursing Research. New York: Jones and Bartlett Publishing. Burney, D., & White, D. (2009). Research Methods . Mason, OH: Cengage. Calder, B. J., Phillips, L. W., & Tybout, A. M. (1983). Beyond External Validity. Journal of Consumer Research , 112 - 114. Carmines, E. G. (2012). Reliability and Validity Assessment. London: SAGE Publications. Carter, Y., & Thomas, C. (2010). Research Methods in Primary Care. London: Radcliffe Publishing. Chandra, S. S., & Sharma, R. K. (2012). Research in Education. New York: Atlantic Publishers. Cottrell, R., & McKenzie, J. F. (2012). Health Promotion and Education Research Methods. New York: Jones and Bartlett. Daniel, J. (2011). Sampling Essentials: Practical Guidelines for Making Sampling Choices. London: SAGE Publications. Duvendack, M., & Palmer-Jones, R. (2013). Replication of Quantitative Work in Development Studies: Experiences and Suggestions. Progress in Development Studies , 307 - 322. Gravetter, F., & Lorzano, L. A. (2013). Research Methods for the Behavioural Sciences. Mason, OH: Cengage. Greenurg, D. H., Michalopoulos, C., & Robins, P. K. (2006). Do Experimental and Nonexperimental Evaluations Give Different Answers about the Effectiveness of Government Funding Training Programs. Journal of Policy Analysis and Management , 523 - 552. Imai, K., Tingley, D., & Yamamoto, T. (2013). Experimental Designs for Defining Casual Mechanisms. Journal of The Royal Statistical Society , 5-51. Lodico, M. G., Spaulding, D. T., & Voegtle, K. H. (2012). Methods in Educational Research. Hoboken, NJ: John Wiley and Sons. MacKenzie, S. B., Podsakoff, P. M., & Podsakoff, N. P. (2011). Construct Measurement and Validation Procedures in MIS Behavioural Research: Integrating New and Existing Techniques. MIS Quarterly , 293 - 297. Macnee, C. L., & McCabe, S. (2012). Understanding Nursing Research. New York: Wolters Kluwer. Mangal, S. K., & Mangal, S. (2012). Research Methodology in Behavioural Sciences. Delhi: PHI Publishing. Muijs, D. (2011). Doing Quantitative Research. London: SAGE. Nykiel, R. A. (2012). Handbook of Marketing Research Methodologies. London: Haworth Press. Schonlau, M. (2009). Selection Bias in Web Surveys and the use of Propensity Scores. Sociological Methods in Research , 291 - 318. Shrader-Frachette, K. (2012). What Will Work: Fighting Climate Change with Renewable Energy, Not Nuclear Power. New York: Oxford University Press. Stouthamer-Loeber, M., & Van Kammen, W. B. (1995). Data Collection and Analysis: A Practical Guide - Hiring. SAGE Research Methods , 3 - 19. Read More