What three variables affect the outcome of a research design
After you learn about the scientific method used for psychological research, visit this page for an opposite point of view…….
As you have read, psychology addresses many topics. To study behavior, psychology must use many different research methods. Psychologists make educated guesses, called hypotheses, about human behavior. These hypotheses are based on theories. Theories are sets of facts or principles that explain why human behavior occurs. Once psychologists have a hypothesis, they test it in experimental situations. In scientific experiments in psychology, certain factors are manipulated to see their effect on what is being studied. These factors are called variables.
One method used in psychology is the case study. The case study looks at a few individuals in depth in an attempt to discover how behavior is affected by certain variables. The advantage of using the case study is that data, or information, is easy to collect. However, since only a few cases are studied, results may not apply to everyone. The survey is another technique which is used by psychologists. A survey is a questionnaire given to many people to gather data (information). The survey is a good research tool because it gathers data from a large number of people and can address a large number of issues at the same time. However, the survey has the disadvantage of having to rely on understandable responses and the possibility that people filling out the survey will try to guess what is expected and answer accordingly. This is called demand characteristics. (also called “set”) Naturalist observation involves studying people by watching them in their natural habitat. Sometimes this is done in person and sometimes video cameras are used. Naturalistic observation allows the researcher the chance to observe and record, rather than rely on memories or another’s data. Since the subjects do not know they are being observed, the chance for demand characteristics is minimized.
The psychological experiment is a research method in which the psychologist purposely manipulates variables. The variable that is studied is called the dependent variable. That which affects the dependent variable is the independent variable.
Look at it this way:
The dependent variable is affected by the independent variable
The independent variable affects the dependent variable.
For example, if a school psychologist wants to study the effect that an “upstairs only” sign has on how students move through the halls, the dependent variable is: student movement in hallways, and the independent variable (that which effects that movement) is the sign which reads “up only”.
Experiments in psychology have certain advantages. Cause and effect relationships can be established and can be verified by others wishing to repeat the experiment. Because the psychologist controls the variables, results can be easily analyzed. However, subjects may not respond naturally because they know they are being studied. The purpose of this reading is to help you design and carry out experiments. It will also help you analyze experiments performed by others.
Planning and carrying out an experiment. In conducting an experiment, you should first decide on the subject area in which you wish to perform the experiment. You might work in such areas as personality, emotions, learning, memory, or the senses. Second, you must draw up plans for carrying out the experiment. You would consider such factors as the amount of money, time, and equipment available. Third, you should submit the plans to someone else for assistance in working out details. Regardless of how careful you are, sometimes an outsider can find errors in design that the original designer, being very close to the experiment, might not notice. Forth, you then carry out the procedures and perform the experiment according to your plans. Fifth, you conclude the experiment with an analysis of the results. Finally, you write a report on the experimental procedures and results so that others can learn from your experience.
You need to develop a hypothesis, which you will test in the experiment A hypothesis is a statement that if a certain thing occurs, then it will cause a certain response. For example, a hypothesis might state, “If background music is played in a department store, then customers will buy more goods.” This statement implies that there is a relationship between playing background music in a department store and the purchases made by customers.
A hypothesis can be stated in a general way or in a very precise manner; or it can take any form between these two extremes. The hypothesis just given about music and purchasing in a department store is stated generally. It could be made more precise by specifying a particular department store, the hours during which the music will be played, and the amount of increase in the purchase of goods. It could be made even more precise by stating exactly what type of music will be played and which type of goods will increase in sales. Most hypotheses are about halfway between very general and very specific.
You now know how to state a hypothesis. But where do hypotheses originate? One source for developing a hypothesis is a previous experiment. Experiments cannot answer all possible questions about a given situation. Often, a well-designed experiment suggests new areas of investigation. A second source of hypotheses is observation of the environment. When observing the behavior of humans or animals, you may notice that certain events cause a particular response. You may then come to believe that if this same set of events occurred again, then the same behavior might result. A third source of a hypothesis is a theory. A theoryis a general principle, based on evidence, that certain phenomena are related. The theory can be tested to determine to what degree the phenomena are related.
The next step is organizing the experiment. The design of an experiment is a set of exactly and precisely planned procedures for testing a hypothesis. The experiment’s design depends on the procedures, materials, and personalities of the subjects to be used in the experiment. It also depends on the way in which the hypothesis is stated. For instance, compare the differences in procedures, materials, and subjects dictated by these two hypotheses:
(1) “If background music is played in a department store, then customers will buy more goods.”
(2) “If semi-classical music is played from 9:00 AM. to noon and from 1:00 P.M. to 5:00 P.M. on Saturdays in a clothing store, female customers will purchase 30 percent more goods.”
The design of the second experiment would differ from the first, since the second hypothesis dictates more specifically the particular stimuli and responses involved.
In an experiment, the stimulus acts on the organism, which then responds in some way. A stimulus is an action, object, event, condition, or situation that causes a behavioral response to occur. A response is the behavioral act that comes about as a result of a stimulus.
It is not possible to have a response without a stimulus, or a stimulus without a response.
Difficulty may arise, however, in correctly identifying the stimulus that is most closely related to a particular response. For example, in the experiment on music in a department store, is there any reason other than the playing of music that might increase the amount of goods purchased? Is there another response that might be more influenced by the music played than the purchasing of goods? Other stimuli could be investigated that might influence the selected response. Other responses could also be selected that might be influenced even more by other stimuli.
In experiments, the stimulus is called the independent variable. The behavioral response that occurs is called the dependent variable. In this experiment, music is the independent variable. The increase in buying goods is the dependent variable. Actually, an experiment is an attempt to establish what relationship, if any, exists between an independent and a dependent variable.
Controlling variables. Suppose you want to state that the independent variable causes the dependent variable to occur or change. Then it is necessary to eliminate or control everything other than the independent variable that might influence the dependent variable. This is sometimes very difficult to do. For instance, imagine controlling variables in different home environments to study the emotional development of children. If all variables except one are not eliminated or controlled, there is no way to know which of the variables has caused a response. Variables that might influence the dependent variable in an experiment may be controlled in several ways. Three of the more common ways are (1) to set up a control group, (2) to group individuals by random selection, or (3) to have one group serve as both the control and the experimental group.
A control group is a group of individuals who are similar or identical to the experimental group in such areas as sex, age, intelligence and socioeconomic status. However, the independent variable is not introduced into the group. A control group is set up so that the results of the independent variable operating on the experimental group can be compared with a group that is similar but does not have the independent variable operating on it.
For example, suppose you were planning to experiment on how distributed studying (spaced sessions separated by rest periods) versus massed studying (one long session) affects grades. You would have to eliminate or control the influence of all major stimuli on grades except studying. One way of designing such an experiment would be to set up a control (comparison) group, which would study in massed periods. You would also have an experimental group, which would use distributed studying. It is important to match both groups as closely as possible for such influential variables as intelligence, sex, grades, interest, motivation, and grade level. Otherwise, if the average IQ for the experimental group were 20 points higher than for the control group, and if the experimental group received higher grades, how would you know if the higher grades were caused by IQ differences or by distributed studying?
A second method of controlling variables is to form two groups by random selection. This procedure consists of distributing individuals into two groups in such a way that the placement of any individual in one group has no bearing on the placement of any individual in the other group. This same principle of selection is followed when you wish to select a sample of people to represent a larger group. For instance, you might want to select a small percentage of individuals to represent the total television-watching population. Every individual in the population has an equal chance of being selected. For accurate results, however, the sample must accurately represent the total group. There must be the same percentage of males and females, of education levels, and so forth in the sample as there is in the total group.
A third way to control variables is to use the same group of individuals for both the experimental and the control groups. This method requires setting up a standard of behavior for the subjects in the group. Then you introduce the independent variable and measure the amount of change, if any, that takes place. For example, consider the experiment on the effect of study habits on grades. After a group of subjects have set up their massed study habits as a standard, they would try distributed studying. Their grades under distributed learning would then be compared with grades given under massed study conditions. It is best if the time lapse between the two situations is short, so that time alone is not responsible for any major changes in the subjects.
One of the hardest variables to control is the set, or attitude
of expectancy, of the subjects.
If the subjects know or suspect that they are to react in a certain way, they are likely to react that way because of their attitude and not because of the independent variable (also called “demand characteristics”). Therefore, it is important in an experiment to maintain some control over the subjects’ set. You would not, for example, broadcast to customers in the store that you believe the playing of music will make them buy more goods.
The conclusion. After the experiment is finished, you should review all procedures and try to find ways to improve them. You might also raise questions about how the results could apply to similar situations.
To summarize the steps involved in conducting an experiment: (1) select an area of behavior to investigate, (2) develop a hypothesis, (3) design an experiment to test the hypothesis, (4) carry out the experiment, (5) report the results, (6) explain them, and (7) review the procedures to see how they could be improved.
Lesson 2 Review
Directions: Follow the instructions in each section below.
Part A: Explain the meaning of these terms and how each relates to the scientific method.
2. Demand characteristic
7. Case study
8. Dependent variable
13. Independent variable
14. Control group
Part B: Read each statement below. Explain how each statement describes part of the design, theory, variable or a hypothesis of an experiment.
17. An educated guess in a psychological experiment that states if a certain thing happens, then it will cause a certain response:
18. A general principle, based on evidence, that certain phenomena are related:
19. The behavioral act that results from a stimulus is a:
20. The stimulus, or that which is effecting what you are studying:
21. That which is being studied in a psychological experiment is the:
22. Selecting individuals from a larger group in such a way that their selection will have no bearing on the experiment:
23. A group identical to or similar to that group which is being studied, used to compare results:
24. A set of exactly planned procedures for testing a hypothesis:
25. Factors that are manipulated in an experiment:
26. Research method that looks at a few individuals from which to draw conclusions:
27. When psychologists send out questionnaires to many people to gather information:
28. When a group or individual answers questions or behaves in the way they think the experimenter wants them to, this is called:
29. Watching people in their natural habitat is:
30. In a psychological experiment, information is called:
31. A research method in which the experimenter purposely manipulates variables is:
Part C: Answer the following questions in your own words.
32. What is the correct order in a psychological experiment? Explain why this order is important in conducting good research.
33. Why is the survey method of research a good research tool?Provide an example.
34. How can using a survey create research problems? Provide an example.
35. What three variables affect the outcome of a research design? How does each variable affect the outcome of a research design?
Answer preview to what three variables affect the outcome of a research design