Abstract
The field of research sees many methods and systems being used. Important among many various research methodologies are the concepts of co-relational and experimental types of research. Each type of research uses different concepts and practices. Co-relational research is used to determine the extent to which multiple variables forms relations with each other among a single group of variable, there is no attempt to manipulate one variable to see the effect on others.
Experimental research on the other hand takes a set of variables and attempts to simulate and manipulate one variable to understand how the effect on other variables. The paper provides a discussion of these two methods and develops an understanding of their differences and where each can be used.
Co-relational and Experimental Research
Bordens (et all, 2005) have provided an insight into various aspects of conducting different types of research and the authors speak of two important types of research: co-relational and experimental types of research. Both types have certain differences and need to be used in specific instances and cases.
According to Bordens, co-relational research is used in instances where a general interrelation between different variables is studied, ‘in situ, meaning that there is no attempt to find out how changing one variable would affect the other. Certain assumptions and criteria have to be observed during the sample preparation phase and the sample for the study should not have units that are influenced by non-study parameters.
This type of research is more static in nature and reflects the co-relations that exist at a certain point in time and with certain conditions having been assumed. The study would have the objective of finding co-relation between variables and have a conclusion that would prove or disprove a hypothesis. An example would be the Intelligent Quotient among children of different demographic groups and different ethnic origins. The result may reveal that there would be a relation in the IQ among children from minority groups who may study in financially less funded schools and children who go to better-funded schools.
The results would not attempt to find out how a child from a minority group would fare if placed for a few years in the better-funded schools. The results would attempt to show that variation in IQ is not only related to the ethnic minority but also to the type of school the child attends. Another example would be to find the correlation between height, weight, and physical fitness between people of different demographics. The study would attempt to show that white and African American ethnic origin people are physically more robustly built than Asian people.
Bordens (et all, 2005) suggest that Experimental research would first create a scenario where the relationship between different variables is formed and then an experiment would be run to find how changes in one variable would affect the other. It involves observing phenomena, developing hypotheses, empirically testing the hypotheses, and refining and revising hypotheses. An example of experimental research would be to find the relation between conductivity and temperature of two different samples of an alloy conductor. The conductivity at room temperature for both samples would be measured initially and then the temperature would be raised at different predetermined control points to find the conductivity of the two samples. The results can be tabulated and a graph plotted to provide the behavior characteristic of the two samples.
The main difference between the two research methods is that while co-relation research would not change any variable but only study the relation between the variables, experimental research requires that one variable be manipulated to find the relation with other variables. In some cases, co-relational research is the initial phase for performing the experimental research. A correlation has direction and can be either positive or negative and it can also display different strengths in the relation. An example of the strength is the Liker 5 scale test where responders to a survey are expected to select a value ranging from 1 to 5 with 1 meaning strongly disagree and 5 meaning strongly agrees.
Co-relation methods are used in areas of research such as customer or employee satisfaction, brand perception, demographic studies, and so on. Experimental research is used in areas such as sciences, engineering, biology, psychology, and others.
Reliability, Validity, and Accuracy of a Measure
Bordens (et all, 2005) have written about the importance of reliability and accuracy of a measure. According to the authors “Reliability is the consistency of the measurement or the degree to which an instrument measures the same way each time it is used under the same condition with the same subjects”. Reliability denotes the repeatability of the measurement and a measure is considered reliable if a person’s score on the same test given twice is similar. It should be noted that reliability is not measured but is estimated. The authors have defined validity “as the strength of the conclusions, inferences or propositions”.
Stinchfield (2003) defines validity as the “best available approximation to the truth or falsity of a given inference, proposition or conclusion.” Again Bordens suggests that Accuracy is the “agreement of measurement with a known standard” and the definition suggests that accuracy is the common factor for both reliability and validity. For research to produce conclusive evidence, measures should be with reference to a known standard whose value is accepted by all. For results to be tenable the accuracy must be acceptable as per the tolerance level of acceptable errors required for the study.
For research to be acceptable, all three terms must go hand in hand. An example of the three terms is an experiment involving the reaction among subjects to a drug for the Stroop test (Cox, et all, 2006). the Stroop test is a psychological test of mental or attention flexibility and vitality. The task takes advantage of the ability to read words more quickly and automatically than one can name colors. Cox had used two groups of students for the test and had informed them that they would be administered medication to increase the concentration level.
Only one group was administered the real drugs while the second was given Placebo, but was not informed that they had been given a placebo. The Liker 5 point scale was used to assess the performance. Cox used some methods to ensure the accuracy, validity, and reliability of the result. The researchers ensured that reliability was maintained by subjecting multiple tests of a similar nature to each subject and the variance levels were filtered to remove the errors. He ensured that validity was maintained by regression analysis of the observations to remove any inconsistencies among subject backgrounds. Cox ensured that accuracy was maintained by using known parameters as defined for the Stroop test.
The results pointed out that there was actually not much difference between people who had been given a Placebo and those that had been actually given the drugs. The validity and reliability of the experiment would have been undermined if at any point in time if the second group had been told that they were given a placebo. So in conclusion, for an experiment to be acceptable, one cannot have reliability without validity and the converse also holds true.
Bordens also speaks of two types of errors, Systematic errors, and Random errors. The systematic error occurs to the same extent in each one of a series of measurements and an example is a wrongly calibrated instrument. Random error occurs in any measurement as a result of variations in the measurement technique and natural drift in an experiment.
ShyQ Shyness Measure
Henderson (et all, 2002) has studied the phenomenon of shyness and have attempted to study methods for bringing change behavior and reduce negative cognitions and emotions. The authors have defined a measure for estimating the shyness measure called the Shyness Quotient or Shy Q and the authors suggest that Shy Q is correlated with being cold or distant, socially inhibited, non-assertive, overly accommodating, self-sacrificing and vindictive or self-centered The authors suggest that shyness and social phobia do not have to interfere with achieving professional and interpersonal goals.
They suggest that challenging automatic thoughts and beliefs and learning new behaviors can relieve the pain of shyness. By conducting surveys using structured questionnaires, the authors have attempted to set define means values of scores using CSIV scales, IIP scales, Clinic Scales, Revised Cheek, and Buss Shyness Scale that measure answers on a 1 to five-point or 1 to seven-point scale.
The case presented in the assignment is discussed with reference to the suggestions for treatment of shyness, as suggested by the American Psychological Association (APA, 2004). The case as presented in the assignment refers to shyness therapy to be provided to two groups of patients and while one group would receive face-to-face therapy, the other would receive online therapy.
As per the recommendations given APA (APA, 2004), the problem of shyness can be only resolved when patients are able to interact with other group members. Subjects undergoing therapy need to be able to learn how to mix with other people who themselves may be shy and have an inability to speak out in front of others. The best method of providing treatment is through peer counseling, group discussion, role-playing, and other cooperative task completion activities and this includes making a speech on any topic. The therapy cannot be undergone by the patient in isolation by listening to audiotapes, seeing broadcast shows, or taking online sessions.
APA suggests that by avoiding face-to-face interactions with others and hiding behind the anonymity of the web, patients may undergo a further relapse and the unaddressed fears of self-ridicule. Such therapy sessions and experiments are more harmful to patients.
In light of these observations, the case as suggested in the assignment for providing therapy sessions online is not ethical. They may not only cause harm to patients but also the results of the therapy cannot be gauged correctly. Moreover, the conditions under which the therapy sessions are taken by the subjects when they are online cannot be assumed as constant. Face-to-face therapy sessions will ensure an environment that is reasonably secure from external variations and is designed to be stress-free.
Participants in the face-to-face therapy would all be seated in the same surroundings and hence the prescriptions of the therapy can be made consistent and with a common base. The online sessions on the other hand would be subject to the very environment that has made the subject shy and the very purpose of the session is defeated. Hence it is suggested that online therapy sessions are not ethical.
Conclusions
The paper has examined factors like co-relational research, experimental research, the relation between validity, reliability, and accuracy in the research, and also the concept of Shyness Quotient. Appropriate examples about all these phenomena have been discussed in detail and the paper also suggests when each type of research can be used.
References
APA, (2004). Painful Shyness: A Psychologist Can Help. Web.
Bordens, Kenneth S., & Abbott, B. (2005). Research design and methods: a process approach. (6th ed.). New York; McGraw-Hill.
Cox W. Miles, Fadardi Javad Salehi, Pothos Emmanuel M. (2006). The addiction Stroop test: Theoretical considerations and procedural recommendations. Journal of Psychological Bulletin. Volume 132. No. 3. pp: 443-476.
Henderson. Lynne., Zimbardo. Philip., (2002), Dimensions of Shyness: The ShyQ. Web.
Stinchfield, Randy., (2003). Reliability, Validity, and Classification Accuracy of a Measure of DSM-IV Diagnostic Criteria for Pathological Gambling. American Journal of Psychiatry. Volume 160. Issue 1. pp: 180-182.