Unlike fields such as biology, chemistry and physics, psychology is not an absolute science. Psychology is often categorized under the “Social Sciences” or “Soft Sciences” because it deals with unpredictable elements such as human thoughts, feelings and behavior. Nevertheless, research and statistics are very crucial in psychology. These serve as the main tools in applying the concepts of psychology, as well as in gathering and interpreting related data.
Psychological research is the thorough study of a certain psychological phenomenon (AllPsych Online, 2004). It uses the scientific method, a systematic process of investigating observations, solving problems and testing hypotheses (Beltz, Monroe & Williford, n.d.). Psychological research is commonly used for the following:
a) Analysis of development and external factors and their effects on an individual’s mental health
b) Study of people with specific psychological disorders, symptoms or characteristics
c) Development of tests to measure specific psychological actualities
d) Development of treatment approaches that will improve mental health (AllPsych Online, 2004)
Conducting a Psychological Research
The first step in peforming a psychological research is to come up with a theory, or a general idea that explains the relationship between a given number of separate facts (AllPsych Online, 2004). A theory can be derived from primary and or secondary data. Primary data refers to information that was collected through first-hand means, such as conducting interviews, answering questionnaires or taking readings from instruments (Tiscali.Reference, 2008). Secondary data, meanwhile, is information that was gathered from other agents like books, newspaper articles and government reports (Tiscali.Reference, 2008).
Variables and Hypothesis.
However, a theory must be more specific for it to be considered testable. Therefore, a researcher must identify the variables in his or her theory. Every experiment has two types of variables: the independent variable (IV) and the dependent variable (DV). The independent variable is the “input variable” (AllPsych Online, 2004), or the variable that is manipulated by the experimenter. The dependent variable is the “outcome variable” (AllPsych Online, 2004), or the results of the experiment. The definition of variables in a particular experiment results in a testable form of a theory called a hypothesis (AllPsych Online, 2004).
The researcher must then validate his or hypothesis by subjecting it to an experiment. But the accuracy of an experiment can be affected by factors known as biases. The forms of biases that usually occur in any experiment are detailed below:
a) Selection Bias – Happens when there are already pre-existing differences between the subjects of an experiment.
b) Placebo Effect – The performance of an experiment is influenced by the subject’s belief about the results.
c) Experimenter Bias – Certain beliefs or attitudes of the experimenter towards the experiment that can affect the subject (AllPsych Online, 2004).
Selection bias is usually controlled through Random Assignment, wherein subjects are grouped according to chance instead of human decision. A blind study, in which the purpose of the experiment is not disclosed to the subjects, is used to control the placebo effect. Experimenter biases are regulated through the double-blind study – both the experimenter and the subjects have no idea regarding the purpose and the expected outcomes of the study (AllPsych Online, 2004).
Before the researcher can finally conduct the experiment, he or she must first standardize it by providing it with a specific set of examples. A standardized experiment ensures that all of its subjects are given the same instructions, handled in the same manner throughout the course of the experiment and are derived the same amount and quality of data. Furthermore, a standardized experiment entails that it be replicated by other experimenters with different subjects to prove the validity of its results (AllPsych Online, 2004).
Statistics and Psychology
The primary role of statistics in psychology is that the former concretizes the subtle effects of a given psychological phenomenon. Whenever psychologists use statistics-related concepts such as response times, scores on scales and tests, counts of errors or demographic data, they give a clearer picture of psychological relationships that are often hidden behind other reasons for variation. Moreover, statistics also serve an indicator of the feasibility of a certain psychological research. The study of human and animal behavior is both time-consuming and expensive. Hence, statistics will give psychologists an idea as to whether or not their study is worth accomplishing (CPA, n.d.).
Structural Equation Modeling (SEM)
Structural Equation Modeling is a group of different statistical techniques that uses path analysis and factor analysis (Statistics Solutions, 2008). SEM is commonly applied to various objective research problems in the social and behavioral sciences. These problems are often related to issues such as consumer behavior, genetic and cultural effects, racial discrimination in employment, housing and earnings, evaluation of social programs and intergenerational occupational mobility (Statistics Solutions, 2008).
Frequency distribution is a statistical technique that is commonly used in the study of a single population (IRI, n.d.). Some ways of illustrating frequency distributions include histograms and cumulative frequency distributions. Histograms present element frequency within a particular values range (IRI, n.d.). Cumulative frequency distributions, meanwhile, show element frequency below a given value (IRI, n.d.).
Even if psychology is a “Social Science” or a “Soft Science,” it is still necessary that it includes research and statistics. Research enables psychologists to understand why people feel, think and act in a certain manner. Statistics, meanwhile, gives psychologists the concrete and exact impact of a certain behavior on an individual and or society as a whole. Together, research and statistics allow psychology to predict the “unpredictable” human nature.