Research Methods

Using good research methods is a necessity in scientific studies to ensure that the findings we conclude are accurate and useful. When research is performed incorrectly it can lead to many issues including possible harm to subjects. In order to ensure that research is conducted properly, scientists have certain safe guards in place which help keep errors to a minimum. Standardization and operational definitions are a good start when it comes to keeping data accurate and unbiased. When using standardization scientists ensure that any tests they are performing are done in a similar manner with the same questions and technique being applied to each subject. After the test is completed the same scoring method is then used to calculate the outcome. This prevents different subjects having different outcomes solely because they were tested differently for the same research. Operational definitions are when the variables in a study are clearly represented by how they are measured in the context of a study. 

Control procedures are another helpful tool when performing research so that results are not skewed by inaccurate information. Double-blind testing is performed when both the researcher and the subject are unaware of certain parts of a test. This ensures that neither party can accidentally alter the results of the test with their own predisposition or behavior. Another fine example of control procedures is the widely used placebo controls in which a group of research participants are given a fake test (placebo) such as a sugar pill in the case of drug studies which allows researches to observe whether or not the medication itself is causing the effects or if instead the idea that it is supposed to work is causing participants to believe it is working. Without such control groups it is possible that medicine such as pain killers would not have any real use but rather you would trick yourself into thinking they are working and thus cease the pain you feel. 

The reason we can sometimes have issues with our conclusions is due to two modes of thinking known as intuitive and analytical. When it comes to intuitive thinking we can talk about our reflexes and initial thoughts. For example when you first see someone and make a judgement you are using intuitive thinking which is fast but can sometimes be incorrect. Analytical thinking is usually slow and reflective which allows us to examine our subjects and dive further into what is going on. When using analytical thinking we are able to do complex functions such as mathematics. 

When it comes to actually administering a study there is much to think about but the first that will be considered here is how to go about the research. 

• Naturalistic observation occurs when a subject is observed in their natural environment. This would be for instance watching college students as they work in class or a library rather than bringing them to a laboratory for observation. The reason this method might be used is that it has high external validity meaning the findings can be applied to the real world. The main problem with naturalistic observation is that because the researcher is not controlling the variables the study is low in internal validity which keeps us from inferring causation. 
• Case Studies are when a researcher will examine one person or a small number of people, often over an extended period of time allowing them to study rare or unusual phenomena. These studies often allow of existence proofs which can show that something actually occurs and can offer insight for later systematic testing. The issue with case studies are that they are typically anecdotal (an issue with pseudoscience) and do not allow us to infer causation. 
• Correlational designs can help predict behavior by studying how certain variables cause other variables to change. Again though this study cannot infer causation. Correlational studies make use of positive and negative correlations to represent findings. A perfectly positive correlation is when one variable goes up so does the other and a perfectly negative correlation is when one variable goes up the other goes down. Correlations can be based from -1 to +1. 
• Experimental designs are used in a laboratory setting and allow us to infer causation thanks to the researcher being able to manipulate the variables. These studies are high in internal validity but are not high in external validity. Experimental designs use independent and dependent variables to test hypothesis. An independent variable is manipulated by the researcher and the dependent variable is influenced by the change in the independent variable. The dependent variable is the thing being measured. 
• Self-report measures are when surveys and questionnaires are used to do research. Using self-report studies allows for a large amount of the population to be studied over a relatively short period but is reliant on a few key features. In order to have a good study the participants must be selected at random  which also is a key for generalizing the results. We also must assume that the participants of such a study are honest in their reports and able to accurately study themselves.