Major Types Of Researches
This essay is introduced by briefly explaining the major types of researches. Terms used in research are also defined and described. The author discusses about the same problems that face researches and how to alleviate them. The author also gives the pros and cons of differentiating descriptive research and compares it with other research designs. The author proceeds to give the similarity of subject design to case-studies and time-series. The threats facing single subjects and their advantages are also discussed.
The purpose of the research is the determiner of the research design that will be employed by a researcher. Researchers often carry out two types of researches. They include descriptive research and explanatory research. Descriptive research tends to answer the questions of what is happening. Explanatory research tries to find out why things are happening the way they are at the time of research. The research designs and the questions are greatly affected by these two factors. They determine the research questions and even control direct the type of questions that are going to be asked. The following are answers to questions related to the research designs.
Description and definition of the terms linear relationship, Pearson correlation, Spearman correlation, regression, and multiple regression
A linear relationship is whereby increasing the amount of one variable by x times will also increase the number of the other variable by the same x times (Kalla, 2011). The two variables are said to be linearly related. An example of such relationship is when making purchases. If one glass costs $5 then five glasses will cost $25.
This is a type of coefficient that shows the relationship between two variables. These two variables must be measured in the same interval or under the same ratio (Investopedia, 2013). Although it is hard to measure the degree of relationship of two variables, Pearson correlation uses a coefficient that eases the research.
This coefficient is often referred to as Spearman’s rho. It is a coefficient that describes the relationship between two variables using a monotonic function (McDonald 2009)
This is a statistical measure that determines the relationship of two dependable and a list of other changing variables (Investopedia, 2013). There are two types of regression; linear and multiple regression.
This is a method of data analysis that is most appropriate when a dependent variable is to be evaluated in relationship to independent variables (Berger, 2003)
Third-variable problem and the directionality problem and how can these be mitigated
These are variables that a researcher is unable to control or eradicate during the research. They put the validity of the research into question. This variable is also known as a mediator variable. They greatly affect the relationship between the dependent and independent variables. This can lead to the analyst analyzing the results wrongly (Shuttleworth, 2008). This problem can be solved by planning the design well in advance before the actual research. Constantly check the progress of the research can also filter out most of the variables.
This is a problem that arises when two variables are dependent and it is not clear to evaluate which factor brings rise to the other factor. The presence of correlation of one variable to another makes it difficult to know the cause of an action. These variables pose a challenge during the experimentation. During research, such a problem can be solved by taking a position. When a researcher decides the causative, the experiment gets a direction and a start point.
General features that differentiate descriptive research from other types.
As discussed earlier, descriptive researches tends to answer the question of what is happening. It is also known as statistical research. The main objective of this kind of research is usually to explain the collected data and study what is happening. Studying frequencies, averages, and other statistical figures is among the ideas behind this kind of research. This type of research also has some unique features that make it stand out from the rest. The fact that it does not research on the reasons behind the occurrence of something makes it effective.
Comparison and distinction of the three types of descriptive research designs and their advantages and disadvantages
There are several types of research design. These descriptive designs all try to answer the question of what is happening. They include observational, case study and survey method. These types have their shortcomings and also merits. They are discussed below.
Observational is the simplest method of data collection. It is easy and convenient since the researcher needs only to set up the objective and the point of focus. He or she observes it and records the observations. This method has a shortcoming in that it is subject to the researcher. The results may be biased from time to time because of the attitude and the moods of the researcher. The researcher might have some prejudice and may be get results that are not true.
Doing a case study is also another descriptive research design. The common advantage of this type of redesign is that it uses live examples. It uses objects r people that are tangible or can be seen. This makes it effective and easier to validate the results. Shortcoming in using this type of design is that there is higher chance of using data that is not genuine or may have been altered.
The last type of descriptive research is by doing surveys. Unlike the other two types, surveys are more thorough and deals with real people on the ground. An advantage of this method is that the research questions are structured to suit all the requirement of research. It is possible to get immediate feedback and understand the causative agents of the research questions unlike the other methods where decisions are made on prepared data. The worst disadvantage of this method is that there is a likelihood of people giving wrong information during the survey. This will lead to recording of results that are fictions. There is also a tendency of people shying away from surveys. This can be a major drawback in some cases.
Why might single subject designs be preferred over other designs
Single subject design is a method that uses one individual on one group as a single entity during the research. This type of design therapeutically is in observation of the behavior of an individual with the aim of some treatment of the subject. These designs are selected since they bring out the macro-level environment of the neighborhood, communities and even the larger society. The major difference of single subject designs is that they use one subject to do the study. Single subject valuable involves the measurement of an independent valuable after the changing of several valuables. In the psychology field, the independent valuable is the patient while the independent valuables are the various therapeutic interventions.
Similarity of single subject design to case-studies and time-series designs
The above techniques are used in psychology and in the medical field in general in order to offer support to weak patient. A common similarity of the three is that they offer personalized attention to the patient in question. The patient’s behavioral changes and mannerism are taken as the main diagnosis feature. The way they relate with others and everything around them is noted down and compared to the normal. Actions are thus taken from the diagnosis. All of the above techniques are time consuming and require the patient’s will for it to work.
Threats to validity of single subjects design
One of the main purposes of research is validity. It is crucial to validate the results and minimize the chances of faulty results. There are a number of reasons that poses threats to validity of single subjects design. They include degree and speed of changes that occur, the length of the intervention and the intervention conditions.
Bodies are always opposed to change, it is important to note the time and the degree to which a body is subjected to a lot of change over a short period of time. This can lead to damage of the body or even acquiring of false results. The intervention conditions should be favorable for the subject. This can increase the chances of better results.
The type of results wanted should always be the key determiner in evaluation of the type of research to use. A pre experiment assessment should always be done in assessing the validity of the experiment. During the undertaking of a research, regular assessing should be done to reduce chances of errors. The errors can also be reduced by adequate planning before the execution of the research.
BIBLIOGRAPHY l 1033 Berger D.E 2003 introduction to multiple regression retrieved Nov 09, 2013 from
http://www.google.co.ke/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&ved=0CGYQFjAG&url=http%3A%2F%2Fwise.cgu.edu%2Fdownloads%2FRegression.doc&ei=4l1-UsrXNovB7Abv24DoBg&usg=AFQjCNFYiY4Z5E7ZWEoySAOczgvYTV6stQ&sig2=kjX3FXLKTuIB3DzOXbxupw&bvm=bv.56146854,d.ZGUInvestopedia 2013.Peason coefficient retrieved Nov 09, 2013 from http://www.investopedia.com/terms/p/pearsoncoefficient.asp
Investopedia 2013.Regression retrieved Nov 09, 2013 from http://www.investopedia.com/terms/r/regression.asp
McDonald, J.H. 2009. p.221-223 Handbook of Biological Statistics (2nd ed.). Sparky House Publishing, Baltimore, Maryland
Siddharth Kalla (Jan 10, 2011). Linear Relationship. Retrieved Nov 09, 2013 from Explorable.com: http://explorable.com/linear-relationship.
Shuttleworth M(Aug 16, 2008). Confounding Variable / Third Variable. Retrieved Nov 09, 2013 from Explorable.com: http://explorable.com/confounding-variables.