How Do Psychologists Conduct Research?

Experimental Studies

In experimental studies, investigators carefully control conditions to discover causal relationships among variables. They are able to do this by conducting the experiment in a laboratory. However, there are situations when it is impossible to bring the participants in the laboratory, in which case, the investigators opt for the natural setting.

The variables are what the investigators try to control and measure. The independent variable is what the investigators control and variate to induce changes in the participants. It is the presumed cause. The dependent variable is the change that occurs in the participants in response to the independent variable. It is the presumed effect, and what the investigators measure. During the first step in research, when investigators state the hypothesis, they talk about the dependent variable as "a function of" the independent variable. For this reason, after they collect the data from the experiment, during data analysis, the independent variable is graphed on the x axis, and the dependent variable on the y axis.

In order to ensure that there are no other variables present in the study besides the independent and dependent variables, the investigators attempt to dilute all other confounding variables that may exist inherently among certain types of participants. Investigators achieve this by randomly assigning the participants in either the control group or the experimental group. There is no independent variable in the control group. It only serves as a benchmark to compare it with the experimental group. The experimental group, on the other hand, is where the independent variable is introduced. Nowadays, most investigators conduct multivariate experiments, wherein several independent variables are manipulated at once. Consequently, in this kind of experiment, investigators randomly assign the participants in several experimental groups, each group pertaining to one of the independent variables the investigators wish to introduce.

Investigators use statistics to analyze the data derived from experiments. They often use the mean to consolidate the dependent variable. The mean is the statistical term for the arithmetic average, or the sum of a set of scores divided by the number of scores in the set. The investigators then compare the mean of the experimental groups to that of the control group, and then to each other, to see if there is a statistically significant difference between them, that is, their difference is not due to chance, or because of a few extreme cases. If a statistically significant difference is established between the dependent variables from the experimental groups, especially in comparison to that of the control group, then investigators can conclude that the independent variable has a causal relationship with the dependent variable.

Correlational Studies

Correlational Studies are similar to experimental studies, except the presumed cause cannot be transformed into an independent variable because it is pre-existing, and cannot be manipulated in different variations or degrees. For example, studying the effect of poverty on educational attainment is a correlational study because poverty, as a presumed cause, cannot be controlled by the investigators, that is, they cannot randomly assign participants in experimental groups by varying household incomes. Instead, as a correlational study, investigators set ranges of household incomes as the presumed cause, and then look for participants who meet the set criteria. Another example of a correlational study is when investigators conduct psychological tests that measure a specific variable to a group of people who vary in a particular trait, such as extent of brain damage and mathematical ability. Because the presumed cause is not under the complete control of the investigators, they cannot establish a causal relationship between the variables, only a correlational relationship.

The correlational relationship between the presumed cause and the presumed effect is established during data analysis, using a statistical tool called the correlation coefficient (represented by the letter r. It indicates the strength and direction of the relationship between the variables. If the correlation coefficient is equal to 1 (r=1.00), there is a perfect relationship between the variables. This rarely happens. Any correlation coefficient greater than or equal to 0.60 (r>=0.60) provides experimental value to the research, that is, if conducted as an experiment, the variables may show a causal relationship. Correlation coefficients between 0.20 and 0.60 (0.20

Observational Studies

Some investigators do not wish to find a relationship among the variables. Instead, they uncover, identify, explore, and describe variables in different situations. Oftentimes, the variables identified in observational studies are then utilized and tested in correlational and experimental studies. Observational studies are very useful in psychological topics that lack literature and theoretical background.

There are 3 research methods commonly used in observational research — direct observation, surveys, and case histories.

Direct Observation is simply observing a phenomenon as it occurs naturally. It may appear easy, but observers are required to undergo training to reduce bias. Famous examples of observational studies using direct observation are that of Strum's research on Kenyan baboons, and Masters and Johnson's research on human sexuality. Shirley Strum observed the same troop of baboons in Kenya for decades. She identified and named each baboon, and she continue to make daily record of their behaviors and social interactions. Her data is rich with information on the mental abilities of baboons and their social dynamics. William Masters and Virginia Johnson, on the other hand, developed, in 1966, techniques for observing human sexuality in the laboratory, which include behavioral observation, recordings of physiological changes, and questionnaires about sensations before, during, and after sexual stimulation.

Survey is the use of interviews and questionnaires to ask people about the topic of interest. It is done by administering a set of pretested questions to a sample (or representative) of the target population. The only limitation and confounding variable in this method is the Social Desirability Effect, which occurs when responders lie to present themselves in a favorable light. Such is the reason why questions must be pretested before actual administration — to identify and mitigate item questions that trigger this effect. A popular survey on people's political opinions, product preferences, and healthcare needs is the Gallup poll. Masters and Johnson's study on human sexuality is based on thousands of interviews conducted by Alfred Kinsey and his associates, which they published in "Sexual Behavior in the Human Male" (1948) and "Sexual Behavior in the Human Female" (1953).

Case Histories are partial biographies of particular individuals. In a case study, investigators ask the participant to recall relevant experiences from their past. The main purpose is to diagnose and treat mental problems, although sometimes they are used to generate variables for further research. The only limitation with this method is the trustworthiness and reliability of the participants' memory, that is, if statements should be accepted at face value, or if they should be investigated and corroborated with factual evidence.

Literature Review

A literature review is a scholarly summary of the existing body of research on a given topic. It has 2 forms. In a narrative review, the investigators (a.k.a. authors) describe studies previously conducted, and then discuss the strength of their data-based conclusions. In meta-analysis, authors use statistical techniques to combine data from previously conducted studies, and then draw conclusions from it.