Four Types of Experimental Designs - Essay Example

 The current investigation looks at four different research designs: survey, case study, ethnography, and mixed design. A survey can be qualitative or quantitative, and looks at research objectives through the attitudes and opinions of a population. A case study is more narrowly defined, and looks at how individual cases represent general trends or issues. Ethnography is a more cultural form of research, while a mixed design may include elements of fixed and flexible research. Introduction There are many ways that experimenters go about doing research. It all depends on the variables of how they want to define their results, what processes they want to pursue, and what methods match their research questions. This report examines four fairly common research methodologies: survey research, case study research, ethnography, and mixed design. The pros and cons of each design are assessed, and examples are given. Survey design A survey or questionnaire is considered to be a relatively accessible, flexible research option. Generally, flexible designs offer benefits to research that is willing to change as new variables come up throughout the process, whereas fixed or quantitative designs such as will be mentioned later, offer benefits to research that is about stabilizing and controlling a certain state of variables or predictions. The quantitative or qualitative survey method is a popular choice among researchers. This methodology differs from explicit ethnography in that it is more rationalized and quantitative in nature, and primarily seeks to answer research questions through the proposition that scientific research cannot be proposed without substantive or grounded data to back it up. This makes the use of statistical information very popular with other type theorists, but getting statistics on the future is difficult, and therefore this is another appropriate reason for the research subject to be covered in a quantitative survey manner, rather than grounded theory or ethnographic methods. When researchers use a survey, they often wind up finding out new research questions and directions, while measuring the opinions and attitudes of subjects. At the same time, they can present their data in a scientific way, through the use of ANOVA, SPSS software, and other methods. Survey research has qualitative and quantitative elements, but despite this, it is not considered to be a mixed methodology, in and of itself. The average survey will still have facets of scientific inquiry, however, which relies on hypothesis testing, verification techniques, and quantitative forms of analysis. For example, in a survey format, researchers of a cross section of interests and organizations can be asked what they think about the interstices between a field of given variables, and therefore from the data that is culled from this research, the degree of perceived attitude and belief can be assessed. Surveys are especially popular in psychological research. The main rationale for this choice is that surveys can give a fairly easy-to-get sample of relevant information from a fairly large sample size, while still being very cost-effective to the researcher and giving quality results. Although surveys have their drawbacks, in terms of false reports, self-report bias, and other issues, such as participants putting down wrong answers on purpose, surveys can shed a lot of light on issues, particularly in a study that seeks to measure attitudes and beliefs in a given population or populations. “Surveys can be classified by their method of data collection. Mail, telephone interview, and in-person interview surveys are the most common. Extracting data from samples of medical and other records is also frequently done. In newer methods of data collection, information is entered directly into computers” (Research, 2007). Generally the advantages of a self-administered survey are “economy, speed, lack of interviewer bias, and the possibility of anonymity and privacy to encourage more candid responses on sensitive issues” (Research, 2007). Surveys can also collect a relatively large amount of information for a relatively low price, as mentioned above, and they can help researchers to narrow down their focus through constructive feedback. However, if a survey only has closed questions, there is no opportunity for this feedback. In other words, there are good surveys and bad surveys done experimentally. Qualitative and quantitative survey research designs and sampling plans can exist together, and there does not really have to be one that is better than the other generally. Sometimes these designs are even used within the same study, but this will most likely not be the case for the assessment measures within a pilot program, which should most likely stick to a quantitative survey with sampling that is as unbiased as possible. “Biases may exist in the selection of sample elements within a given cell even though its proportion of the population is accurately estimated… attempts have been made to combine probability and quota sampling techniques, but the effectiveness of this effort remains to be seen” (Babbie, 1995). It is possible to create something that may include scales of attitudes and belief as well, which can then be presented in the form of charts and statistics. For example, political polls taken during election years are a type of survey research. They measure people’s attitudes and beliefs, and put them in the form of ratios and percentages. A more quantitative method may measure output more effectively. “If the members of a population have unequal probabilities of selection into the sample, it is necessary to assign weights to the different observations made in order to provide a representative picture of the total population” (Babbie, 1995). To make the survey a qualitative measure, the output quality and outcomes would have to be based on a measurement of the attitudes and beliefs of a population, such as clients of a healthcare facility. Clients could rate their level of satisfaction with various aspects of the system, and in this way they would be espousing their attitudes and beliefs, which can lead to interesting conclusions, and constructive criticism that may lead to the formation of new programmatic directions. In this manner, survey research can open up new questions about the challenge of the future. A quantitative measure examining output, quality and outcomes may be more based on empirical data than attitudes and beliefs. It may be better to do another type of assessment measure in addition to a survey, but it also may behoove people to use quantitative methods, because these types of designs are more rigid. However, they also offer researchers results that are often more solid and considered to be more scientifically credible. In some cases, researchers using more qualitative designs do not even want to be considered part of the scientific process. In other cases, surveys can be treated as the progenitors of hard data. Survey research is a very common methodology, especially in the social sciences. Case study design Case study research methodology usually signifies that the case being examined is in some way unique or rare, or represents traits or trends that can be expanded to a larger population or research setting. This method focuses specifically on one concern or case study, which could be spread out over a few individuals or organizations, but remains fairly unique and distinctive, thus limiting the sample size severely in this method. The case study research methodology is useful in examining conditions that are rare or infrequent, and often draw interest in this manner. For example, a case study research methodology could focus on individuals who believe they are being controlled by aliens from Mars. Unlike survey methods, however, case studies are mostly qualitative: they do not lead to a lot of quantifiable data. However, a case study may be part of a larger mixed methodology, which shows the conclusions of a more expansive sample size, and makes correlations. Case studies may also rely on existing literature to cull a set of cases from extant research. As Robson implies through the categorical division of validity issues in research, the categorization or coding process is just as important as the research itself in many cases: “Literature discussing similar findings is important as well because it ties together underlying similarities in phenomena normally not associated with each other. The result is often a theory with stronger internal validity, wider generalizability, and higher conceptual level” (Robson, 2003). The stronger the validity of the question, the stronger the process will be in eventuality in terms of producing definite results through a case study method. As with surveys, researchers using case study methods may uncover new research questions. A case study may also use multiple cases to illustrate its points; however, due to the constrictions of time, space, and detail, the sample size of cases often remains relatively small, especially when compared with a different type of research, which could cull the answers of thousands of respondents. Case studies are more detailed and intimate. There are many ways that researchers can find case reflections in terms of the working parts of the society, but there are also costs to be considered. Case studies are predominantly used in business studies, sometimes called “business cases.” “There are dangers in transferring the case methods in law and business to business ethics. Not much is drearier than a tedious and unrewarding exposure to the moral opinions of those ignorant of the kinds of material outlined in the… history of ethics and types of ethical theory. Theory and history, however, also should not remain isolated from modification by case study” (Research, 2007). The case study is just one way that researchers can signify data in terms of the measurement of variables in any given case example, which then exists to highlight a general population. In other words, an isolated case example can often shed light on a larger issue. One case example format’s benefit is that it is often more informative and raises more new research questions than a more quantitative methodology tends to, in terms of exploring new variables throughout the case example as a sort of meta-analytical field research. “Case study research differs from some other models of observation in that it is not only a data-collecting activity. Frequently, perhaps typically, it is a theory-generating activity as well. As a case researcher you will seldom approach your task with precisely defined hypothesis to be tested” (Babbie, 1995). Case studies, on the downside, tend to focus specifically on one community in terms of how it is affected by certain variables, rather than presenting the problem and solutions holistically in terms of what works on a broader scale. Case studies remain a valid methodological consideration, but they are also limited in terms of scope and applicability to a wider set of respondents or communities. If a case study is not applicable to wider populations, then it loses some of its research validity. Despite these possible limitations, case study methodology does add detail and specificity to the problems presented in many scientific studies. Qualitative research can be achieved through many techniques, including interview, survey, meta-analysis, and case study. “Stake argued for another approach centered on a more intuitive, empirically-grounded generalization. He termed it naturalistic generalization. His argument was based on the harmonious relationship between the reader’s experiences and the case study itself. He expected that the data generated by case studies would often resonate experientially with a broad cross section of readers, thereby facilitating a greater understanding of the phenomenon” (Tellis, 1997). This research is appropriate in many different fields and populations: psychologists use case studies, and so do business researchers in organizational studies. Case studies do not present their readers with a list of complicated numbers and statistics; the form is narrative, making this type of research more accessible. The fact that case studies can be used in many disciplines and for many populations, shows their usefulness. These populations that can be said to be truly representative of the issue at hand, based on solid statistical research regarding the evolution and continuance of the case study method, from its beginnings to the present. This may concentrate on meta-analysis or it also may use case example, both of which are considered to be qualitative methods of inquiry. “Case studies are multi-perspectival analyses. This means that the researcher considers not just the voice and perspective of the actors, but also of the relevant groups of actors and the interaction between them. This is one aspect that is a salient point in the characteristic that case studies possess. They give a voice to the powerless and voiceless. When sociological investigations present many studies of the homeless and powerless, they do so from the voice of the elite” (Tellis, 1997). Case studies, as mentioned, are particularly common in psychological, as well as sociological, texts, which hope to illustrate examples. This is because the individual case often works to shed light on a common mental problem or social issue. As noted above, narrative analysis in the average case study tends to read more like a story than like facts and statistics. This makes the use of statistical information very popular with other type theorists, but as noted above, often getting statistics is difficult, and therefore this is another appropriate reason for the subject of the case study research to be covered. This will still have facets of scientific inquiry however, which relies on hypothesis testing, verification techniques, and quantitative forms of analysis. As Creswell implies through the categorical division of validity issues in research, the categorization or coding process is just as important as the research itself in many cases: “Narrative researchers collect stories from individuals and retell or restory the participants’ stories into a framework such as a chronology of the characters, the setting, the problem, the actions, and a resolution of those actions” (Creswell, 2008). Case studies are also very accessible to a lay public, because they are much easier for the average person to read than many other designs, especially in sections on data collection and results. Ethnographic design Many research authors cite a strong reliance on ethnography, particularly in the field of the social sciences. “Ethnography in general is concerned with the study of social groups and social systems within a culture (Bogdan & Bicklen, 1992; Denzin & Lincoln, 2000). It uses an ‘emic’ approach (McMillan & Schemacher, 2001) that focuses on discovering and understanding patterns, actions, language, and explanations that are unique to the culture of particular groups of people” (Knotek, 2003). The research also has elements of grounded theory in terms of its categorization, and considers culture to be a significant variable. Ethnography can be compared somewhat with the study of cultural anthropology. Experimental ethnographic research methodology may deal with, literally, making experiments to prove or disprove a hypothesis or central concern. Sampling and assignment methods may be used for research participants in this form of research. Different experimental conditions could be achieved to contrast different results. Usually a control group is established: for example, two different groups may studied within a cultural setting, such as a shopping mall, to determine the relationships of their behavior to their surroundings, with the control group in place. Ethnographic research methodology is particularly useful if the researcher wants to establish cause and effect relationships and make inferences from them, regarding the society or culture which surrounds the individual. Ethnographic research may also be correlational. This research methodology focuses more on the relationship between ethnographic variables. For example, if researchers think that watching television causes people in the USA to be more passive, they will try to demonstrate links between these two variables, frequency of television viewing and passivity, through correlational ethnographic research, based on cultural considerations. This research methodology is sometimes vulnerable to unexpected or unanticipated causal relationships outside of the variables being compared; however, this can be said of most methods. Extraneous variables are not the sole domain of any one research approach. Ethnography is very popular in social work and sociological research. Ethnography and ethnographic research methods help these types of studies to meet and even exceeds criteria for qualitative social and psychological work research methodologies. Note how one set of authors define ethnographic research and also give examples of how different teams of ethnographers will work in their proposed experiment, as well as showing clearly how these methods will be translated from planning and theory to the reality of the research experiment: “Ethnographic research is a naturalistic, observational method that elicits rich descriptions of context and culture. It alternates between a narrow and broad focus. Ethnographers study a wide range of topical subjects, including individual behavior, material conditions, and shared, taken-for-granted patterns of thought” (Burton and Wilson, 2000). These authors spend a lot of time talking about ethnography, which adequately clarifies their specification of the chosen study method. Although ethnographic research is a qualitative method, it is not always really appropriate for such an already-generalized research issue. First of all, attention must be paid to the facets of ethnographic research which in general is concerned with the study of social groups and social systems within a culture, and finding uniqueness in this culture. The research also has elements of grounded theory in terms of its categorization, but primarily uses an ethnographic viewpoint. In this research methodology, the subjects or respondents involved in the research are seen in the light of the social context, circumstances, and surroundings which are seen by ethnographers to play a vital role in their expressions, behaviors, and other variables that may come under scrutiny. It is argued that this method of research therefore is applicable naturally to a wide range of concerns within this theoretical construct of externalizing the focus and looking at how societal variables affect people on an individual basis. The focus of ethnographic research is naturally flexible so as to provide the researcher with a wide range of options during the process of investigation. Also, “There is a strong anti-philosophical strand in ethnographic thinking that places value on the practice and products of research and has little patience with or interest in discussions about research. Some have a degree of sympathy with this. Philosophical discussion and debate can easily become a distraction” (Huberman and Miles, 2002). This makes the ethnographic method somewhat inappropriate in terms of its actual or perceived relevance to the future of some research questions, in terms of its applicability to a proposed research method of literature review and general survey. Some researchers found the advantages of the ethnographic design approach to be primarily centered on its directness and interactivity of observation. This also involves the abovementioned concept of ethnographic research. “In ethnography, researchers engage systematically with those they are studying: participating in their lives, and methodically asking questions about the information they are learning… This approach allows us to focus on the research questions and topics central to the overall projects, while building in sufficient flexibility to capture emergent themes and unanticipated information” (Huberman and Miles, 2002). These authors did not mention any disadvantages to the method, but it is arguable that such direct involvement with research participants may increase the likelihood of research bias and the possible provision of leading questions by researchers who become too involved in the respondent set. Ethnographic research is a qualitative method, and it tends to be very specific and look at differences in demographics instead of policies. Ethnography may concentrate on looking at people in terms of their society and/or their community, and seeing them on these demographic levels as a primary perspective. Ethnography involves focus and looking at how societal variables affect people on an individual basis. Some also state that the focus of ethnographic research is naturally flexible so as to provide the researcher with a wide range of options during the process of investigation (Babbie, 1995). This is proposed as useful because in many experiments, researchers begin by asking questions, and then, through the process of research, discover new questions as well as answers. Therefore, in this line of argument, the research design needs to be malleable enough to account for these new questions without approaching the subjects from too rigid a set of expectations. However, generally, the more malleable a research design is, the more margin there is for error. Mixed design Mixing qualitative and quantitative research means looking actively for results by using all means at one’s disposal. A study may also have elements of qualitative as well as quantitative designs. Mixed designs defy type because they are both qualitative and quantitative. They might in a mixed method even combine qualitative and qualitative methods in the same study. For example, there could be a case study that is followed by a statistical analysis of a social problem shown in the case study, all in the same scientific paper or article. Therefore, there is no single type: it is two types, and sometimes more. “There are several reasons for using a mixed methods design to conduct a study. In general, you conduct a mixed methods study when you have both quantitative and qualitative data and both types of data together provide a better understanding of the research problem” (Creswell, 2008). There are many different types of research that can be done, and many ways of doing research. For example, in an optimal design, flexibility is maximized, and output is also maximized. Mixed designs are often applicable for a variety of applications, when researchers want to make sure that they cover all their bases, or test variables in multiple fashions. Mixing flexible and fixed designs in the same study is a way of including holism as a goal of the research. As noted above, generally, flexible designs offer benefits to research that is willing to change as new variables come up throughout the process, whereas fixed or quantitative designs offer benefits to research that is about stabilizing and controlling a certain state of variables or predictions. Qualitative research methods were originally developed not to record scientific phenomena, but rather to record social phenomena. This makes qualitative research again very clearly differentiated from quantiative research, which has a distinctively scientific approach that is based on natural science rather than social science. This is another reason that qualitative research methodology could perhaps be seen as more appropriate, for example, for educational research purposes, because it focuses on social relations between individuals, rather than scientific explanations for natural phenomena. At the same time, quantitative research may be more appropriate to research that has fewer gray areas, such as the results of a medical experiment or poll. If the research setting itself is not basically social, in terms of the relationships and interrelationships between researchers and participants, quantitative research may be called-for. It is a general mix of pros and cons for both qualitative and quantitative approaches, so the researcher using both in one study can, hopefully, optimize the pros. Mixed research often involves the researcher taking an active role in solving problems, and looking at strategic measures and decisions that are made. This type of research is reflective, and it focuses on problem solving. “Qualitative research can also be undertaken by larger organizations or institutions, assisted or guided by professional researchers, with the aim of improving their strategies, practices, and knowledge of the environments within which they practice. As designers and stakeholders, researchers work with others to propose a new course of action to help their community improve its work practices” (Babbie, 1995). In an example of a mixed method, an experiment could include an informal survey or literature review (flexible), along with an experimental post-test design (fixed). Mixed-design research is often researcher focused. This type of research is dynamic and can be optimized for different environments requiring different circumstances. “Overall, this process emphasizes practical action research centered around studying a local problem, engaging in inquiry by an individual teacher or team, and focusing on development” (Creswell, 2008). In the first stage of this research, there is an action plan and process. In the second stage, data collection and development becomes optimized within the processes. The third stage involves the production of mixed research, and, “In the final stage, the researcher finally completes an action plan. This includes a summary of findings, recommend actions, and be the identification of individuals responsible for action and individuals who need to be consulted and informed” (Creswell, 2008). It is important to summarize objectively when using mixed methods. Conclusion This report has examined several common research methodologies. Each has its advantages and disadvantages, and can be tailored to the research situation. For example, a researcher who wants to know how teachers feel about their students might be better off using a survey methodology, while a researcher who wants to know about how people act differently in various stores in a shopping mall, might want to use ethnographic designs. There are many research designs that can be used by experimenters, and they all share basic common goals. REFERENCE Babbie, Earl (1995). The Practice of Social Research. New York: Wadsworth- Thompson Publishing Burton, L. and W. Wilson (2000). Brief Description of Ethnographic Component. Welfare, Children, and Families. Cresswell, J (2008). Conducting Research. Upper Saddle River, NJ: Pearson- Prentice Hall. Knotek, S. (2003, Spring). Bias in problem solving and the social process of student study teams: a qualitative investigation. Journal of Special Education. Robson, C. (2002). Real World Research. New York: Blackwell. Tellis, W. (1997). Introduction to Case Study. The Qualitative Report 3(2). Research: Experimental methods (2007). http://psy1.clarion.edu/mm/General/Methods/Methods.html Read More