THE EXPERIMENTAL METHOD
SPECIFICATION:
Students should demonstrate knowledge and understanding of the following research methods,
The Experimental method.
Types of experiments: laboratory
Types of experiments: field experiments
Types of experiments: Natural experiments
Types of experiments: Quasi-experiments.
Experimental designs: repeated measures, independent groups, matched pairs.
WHY DOES PSYCHOLOGY NEED RESEARCH METHODS?
Research serves as the foundational pillar for both the social sciences and sciences. It is through research that theories and hypotheses can be tested and verified. In the study of psychology, relying solely on personal arguments, beliefs, and observations is insufficient, as these are inherently biased, subjective, and often based on unrepresentative samples. Additionally, personal opinions are frequently influenced by individual desires to affirm pre-existing political, religious, and moral agendas. Psychologists, therefore, are expected to approach their work without preconceived expectations or preferences, embracing the insights and conclusions drawn from empirical research. This objective stance enables the field to advance based on evidence and scientific inquiry rather than personal bias
THE EXPERIMENT IS THE BEST BEST RESEARCH METHOD
Research is split into two camps: Experimental research and non-experimental research, but experimentation is by far the most robust method.
.
EXPERIMENTS
The experiment is pivotal in the psychologist's arsenal, distinguished as the sole scientific method capable of uncovering causal relationships—the underlying causes of behaviours. To achieve this, an experiment necessitates the manipulation of the independent variable while meticulously controlling for any extraneous variables within a regulated setting. This methodological rigour ensures that any observed changes in the dependent variable can be directly attributed to manipulating the independent variable, thus providing concrete insights into cause-and-effect dynamics in psychological research.
WHAT IS A TRUE EXPERIMENT?
In the realm of experimental research, there are various types of experiments: Laboratory, Field, Natural and Quasi.However, only laboratory and Field experiments are commonly recognized as "true experiments" due to their rigorous design and ability to establish cause-and-effect relationships. The following rules characterize these true experiments:
RULE ONE:
Conducting a “true experiment” stipulates the necessity of having at least two conditions, meaning that participants must engage in distinctly different activities. A scenario where all participants are subjected to the same condition, such as being requested to lend money on a high street, fails to qualify as an experiment because it lacks variability in conditions (only one condition is present). To transform this scenario into a genuine experiment, one could introduce varied conditions by asking one group of participants to lend money during rainy weather and another group during sunny weather. This variation creates two distinct conditions: soliciting during sunny and rainy circumstances. The introduction of this second condition is crucial for the foundation of experimental research, as it allows for a comparison essential for drawing cause-and-effect conclusions. Without this comparative basis, the experimental design is incomplete and cannot adequately explore the causal relationships between variables.
In more technical terms, social scientists refer to different conditions in an experiment as independent variables (IVs). Therefore, it's crucial to articulate the foundational rule of experimental design with the appropriate terminology for clarity and precision. An experiment must feature at least one independent variable (IV) alongside a control condition, ensuring participants engage in at least two distinct activities.
This requirement can manifest in several forms, including:
There are at least two independent variables (IVs); for example, participants might listen to music by Mozart or Bach while completing an IQ test (the dependent variable or DV).
A scenario with one IV and a control condition, such as participants listening to Mozart or experiencing silence while completing an IQ test (the DV).
In a design incorporating multiple IVs and a control condition, participants might listen to Mozart, Bach, Beethoven, or no music while completing an IQ test (the DV).
Crucially, every experiment must also identify a dependent variable (DV), the outcome or effect measured to assess the impact of the independent variable(s). This structure allows for a precise examination of how variations in the IV(s) influence changes in the DV, facilitating a robust analysis of cause-and-effect relationships
RULE TWO:
In experimental research, the dependent variable (DV) is essentially the result that is measured to assess the impact of the independent variable(s) (IVs). It is the outcome or effect that researchers are interested in examining, specifically how it changes in response to manipulations of the IV. This relationship allows for a precise investigation of cause-and-effect dynamics within the study.
For example, consider a hypothesis stating that consuming cheese before bedtime leads to nightmares. In this scenario, the IV would be the consumption of cheese, and the DV—the measured result—would be the occurrence of nightmares. Researchers would manipulate the IV by having some participants consume cheese before sleep. In contrast, others do not and then measure the DV by recording the frequency or intensity of nightmares experienced by participants. This structure highlights how variations in the IV (cheese consumption) directly influence DV (nightmares) changes, facilitating a robust analysis of the hypothesized cause-and-effect relationship.
RULE THREE: To accurately infer a cause-and-effect relationship in experimental research, controlling or eliminating all variables other than the independent variable (IV) is crucial. Failure to do so can lead to results being influenced by extraneous variables (EVs) or confounding variables, compromising the study's validity.
Take, for example, a scenario where a research student conducts a study asking participants to rate the attractiveness of photographs of the opposite sex. If the student overlooks the need for uniformity in the photographs, resulting in some images featuring smiling individuals while others depict bored or unfriendly expressions, the study's outcomes could be skewed. Instead of measuring facial attractiveness, the ratings may inadvertently reflect perceptions of friendliness or approachability due to the variance in expressions. This oversight demonstrates how uncontrolled extraneous variables, such as the presence or absence of a smile, can invalidate the research findings, as the effect observed might not be due to the intended IV but instead to these unaccounted-for factors.
RULE FOUR
For an investigation to qualify as a “true experiment”, the researcher must actively manipulate the independent variable (IV)—that is, they must introduce or modify the IV to observe its effect on the dependent variable (DV). In contrast, if the IV occurs naturally without the researcher's intervention, the study is classified as a 'natural experiment.' Natural experiments leverage events in the real world, such as natural disasters, to study their impact on individuals or communities.
A prime example of a natural experiment is the examination of panic attack prevalence among New Yorkers before and after the 9/11 terrorist attacks. The 9/11 event, serving as the IV, was not orchestrated by researchers; it was an unforeseen tragedy that occurred independently of any scientific study. Due to the impossibility of reproducing such an event for research purposes and because researchers cannot control extraneous variables or the assignment of participants in the context of natural disasters, natural experiments offer less certainty in establishing cause-and-effect relationships compared to true experiments. The inherent lack of control over the IV and other influencing factors means that, while natural experiments can provide valuable insights, they are not classified as true experiments due to these limitations.
RULE FIVE
In a “true experiment”, participants (PPs) must be randomly allocated to different conditions. This random allocation ensures that each participant has an equal chance of being assigned to any of the study's conditions, thereby safeguarding against selection bias and ensuring the comparability of groups. Without random allocation, the study might be categorized as a quasi-experiment, which, while still valuable, lacks the random assignment feature that helps establish stronger causal relationships.
The Stanford prison experiment conducted by Philip Zimbardo serves as a prominent example to illustrate the importance of random allocation in experimental design. In this study, participants were randomly assigned to either the role of a prisoner or a prison guard. This methodological choice was crucial for the experiment's integrity, as it ensured that any participant could end up in either role, independent of their physical size, personality, or other characteristics. This randomisation process eliminates potential selection bias, allowing for more credible conclusions. The absence of bias in the allocation process means that the observed differences between the prisoner and guard groups can be more confidently attributed to the effects of the experimental conditions rather than pre-existing differences between the groups. Hence, the random allocation in Zimbardo's study is a key factor contributing to the believability and validity of the results.
