As a result of studying this chapter, students will:


• the main features of the experimental method, its fundamental differences from other empirical methods in psychology;

• principles for planning an experimental study;

• The main threats to the validity of the experiment;

be able to

• evaluate the internal, external and co-structural validity of the experiment;

• put forward competing explanations for the results obtained in the experiment;


• the basic methodical apparatus of the theory of planning a psychological experiment.

General characteristic of the experimental method

The theory of the experiment is devoted to an extensive literature, and it is impossible to present this material in one chapter. In addition, such a statement is not included in the goals and objectives of the course of the general psychological workshop. Therefore, further we introduce only the basic, most important concepts and representations from this theory. In the main, we will focus on the applied and operational aspects of experiment planning. In other words, here and below we will define a set of actions to be taken during the planning of the experiment and explain them at the level of common sense without going into deeper details of the theoretical justification.

An experiment is the only method of psychological research that allows you to test the cause-and-effect hypotheses. Thus, the experiment allows us to fully realize both the main goals of scientific research in psychology, which we examined in detail in the very first chapter of our textbook - to predict and explain the behavior of people. This experiment in the exact sense of this word differs from the methods of obtaining new scientific facts, including correlative and quasi-experimental studies, in which it is possible to directly reflect any connections, except causally - investigative.

Identifying the connections between certain phenomena allows us to predict the state of one phenomenon according to the state of another. For example, if we know of the existence of reliable links between alopecia and aggressiveness, we can predict by the amount of hair on the head of a person how he will behave in an unpleasant situation for him: rather aggressively or peacefully. However, such a study does not give grounds to believe that a decrease in aggressiveness will help a balding person to restore hair or that hair restoration will reduce its aggressiveness. Identifying the connections between phenomena, we can predict certain phenomena on the basis of others, without knowing anything about their causes.

In order to identify the cause-effect relationship, it is necessary to prove three things.

First, it is necessary to prove that some phenomenon that we consider to be the cause precedes a change in those events that we consider a consequence. And if the investigation is really a consequence, then no matter how we influence the consequences, this will not change the cause. Only in this case it can be argued that the event that we consider to be the cause is indeed the source that influences the investigation.

For example, if we want to test whether people in a stuffy room are worse at solving problems than in well-ventilated ones, then we should ask the subjects to solve problems in a stuffy room, and then ventilate the room and again ask the subjects to solve problems in it. If, as a result, in a stuffy room, subjects find fewer correct answers than after airing, we can assume that the reason for this was stuffiness. At the same time, if the researcher somehow influences the investigation in this experiment - the number of correct answers of the subjects (for example, with the help of training), then the improvement of the solution of problems can not in any way affect the supposed cause - the air quality in the room. Therefore, in this case we are dealing with a causal relationship.

For contrast, we can give another example. The researcher wants to check whether the dominant position of the mother really leads to the fact that the child grows bezynitsiativnym. To do this, he can recruit two groups of mothers - dominant and easily obeying, and compare the initiative of their children. But if he discovers the difference and concludes that the dominance of the mother leads to a decrease in the initiative of the child, then it will be wrong. It is no less likely that an unemployed child, by his passive behavior, forces the mother to take the initiative on herself and become dominant. In other words, not only the alleged cause - the dominance of the mother - causes the child to be childless, but, on the contrary, the child's lack of initiative determines the behavior of the mother, causing her to behave more dominantly.

In this example, we are talking about the existence of a link between the dominance of the mother and the child's lack of initiative, but we can not call this causal relationship until we understand what facts are discovered earlier and which later. In causality, cause and effect can never be reversed. In other words, it is required that changing one variable , which is the cause , always preceded change of another variable , which is considered as a consequence. If both variables act on an equal footing, as in the example just discussed, about the cause-and-effect relationships can only be with great caution.

This condition is met by special methods of organizing the experiment, which were called experimental planning. It presupposes the control of experimental conditions on the part of the experimenter, which varies various effects on the subjects. These effects are called independent variables, since they are set by the researcher himself. The researcher's task is to track and measure the consequences of changing independent variables. Such consequences are called dependent variables, because they do not directly depend on the researcher, but only register them.

Second , we must be sure that there is indeed a connection between the variables being examined, and if we observe a change in one of the two variables, that the second variable has changed. So, in the example given with the influence of stuffiness in the room on the efficiency of mental activity, we must show that the change in the success of the solution of problems must, at least in part, be reflected in the facts of the change in the stuffiness of the room.

Imagine that we have two groups of subjects, one of which solves problems in a stuffy room, and the other - in a well-ventilated room. It is clear that within each group we will observe this or that variability in the success of mental work. However, if the stuffiness really affects the success of the task, the overall differences between the groups should exceed the differences within the groups.

Typically, to assess the relationship between the alleged cause and the intended consequence, the researcher uses special mathematical procedures, called the methods of statistical data processing. As such, in experimental psychology, various types of dispersion and correlation/regression analysis are most often used. These methods with a certain degree of probability make it possible to establish the presence or absence of a relationship between the independent and dependent variables.

Thirdly , in order to prove that the measured relationship between the two variables reflects the influence of one of them on another, it is necessary to eliminate the influence of any other events that could lead to a change in the dependent. For example, in the previous example, it could turn out that while people were solving tasks in a stuffy room, they were tired, they became bored, they switched to something more important and after the ventilation they began to solve tasks even worse than in stuffiness. Or, conversely, solving problems in a stuffy room, they tuned in to work and after the ventilation began to solve tasks better not because of the air in the room, but because they were so well tuned, trained to solve problems, had time to discuss them with each other, argue , who will solve more problems, etc.

Therefore, if in this study we take different test subjects, so that they simultaneously solve the same problems in a stuffy and ventilated room, then we need to make sure that they do not initially differ in abilities and skills in solving problems, motivation, etc. As a rule , most of the experiments are carried out in laboratory, as strictly controlled conditions, precisely in order to control all extraneous, side effects that along with the supposed cause can influence the event-consequence.

The implementation of the third condition is facilitated by various methods of experimental control. They are standard schemes of the organization of the experiment, which allow to stabilize or exclude at all the most probable side effects associated with the influence of time factors, the sequence of events and the individual differences of the subjects. Such methods of organizing the experiment are intended to approximate the actual experiment to the ideal, in which all conditions are stabilized, except for the possibility of the experimenter changing an independent variable. If under such conditions there is a change in the dependent variable, we can conclude that they are really determined by changes in the independent variable.

All the variety of experimental plans and procedures that implement them, ultimately, is aimed at checking whether these three conditions of cause and effect are observed. The purpose of this check is the special structure of any experimental research, which necessarily involves a preliminary measurement of the researcher of interest, the subsequent experimental impact on it and the final measurement of the same process. If the re-measurement is significantly different from the first one and the researcher can prove that all other influences on the subjects, except for the intended cause, have been eliminated, he can safely conclude that it is his impact that is the cause of the detected changes.

And if the evaluation of the difference between the two dimensions - the preliminary and final ones - is a matter of technical rather than statistical ownership, then the proof of the absence of other side effects is a serious problem that the researcher must solve every time in the planning of the experiment procedure. The theory of experiment gives only some very approximate guidelines on this path.

Also We Can Offer!

Other services that we offer

If you don’t see the necessary subject, paper type, or topic in our list of available services and examples, don’t worry! We have a number of other academic disciplines to suit the needs of anyone who visits this website looking for help.

How to ...

We made your life easier with putting together a big number of articles and guidelines on how to plan and write different types of assignments (Essay, Research Paper, Dissertation etc)