1. In Janet’s final project, one of her ideas utilizes a correlational design. In a correlational design, there is research involved. These researches usually involve using observational, archival, or survey methods to test a hypothesis. There is at least two variables being looked at in a correlational design, and these variables are rarely manipulated. Instead, the variables are determined if there is a relationship between them. In saying a relationship, we are trying to determine the similarities or differences between the variables, and the way one variable can correlate or form a relationship with another (Brehm, Kassin, & Fein, p. 36).
This relationship can be measured based on the direction and strength. Let us call the variables X and Y. When X increases and Y doesn’t increase in values, we say there is no correlation or relationship between the two variables. If as X increases, Y increases also, we can fairly say there is a correlation or relationship between the two variables. The relationship is either negative or positive, depending on Y increasing or decreasing in value. If as X increases Y increases, there is positive correlation, and if as X increases Y decreases, there is negative correlation. The strength of the correlation indicates how strong the relationship is between the variables. The strength is calculated using a mathematical equation, but we won’t get into that in this paper. The measurement is between –1 and +1. The closer to –1, the stronger the relationship, and the closer to +1, the stronger the relationship. When the correlational coefficient, or the number value associated with strength, is closer to –1, we say a strong negative correlation. When closer to +1, we say a strong positive correlation. When it is 0, there is no relationship (Brehm, Kassin, & Fein, pg. 36,37).
It is always important to know there is no causation or cause and effect in a correlational design.
In Janet’s first design, it is clearly a correlational design being used. Janet utilizes two types of research methods, observation and self-report. She is observing the one variable, facial expression and scoring it on a scale of 1-5. Negative facial expression gets a low number and positive facial expression gets a high number. The questionnaire is the second variable. The person who Janet observed would be given a questionnaire and asked to rate their mood on a scale of 1-10. The lower the score, the more negative the mood, and the higher, the more positive of a mood.
The correlational study Janet is trying to determine is between the variable facial expression and the other variable mood. The design is trying to form a relationship between facial expression and mood. The criteria of this design is clearly correlational, because of the relationship trying to be determined between the two variables, and the fact that there is no manipulation being done.
Next, the second research design of Janet’s uses the idea of an experimental design. In an experiment, the idea is to establish cause and effect between variables. The experimenter does through control and random assignment. Control means that the experimenter has control over all procedures being done. He/she has control over the variables, the participants, and maintaining uniformity. Random assignment is that participants have an equal chance of being exposed to the independent variable or dependent variable. Also, participants aren’t chosen for the conditions of the experiment based on race, sex, behavioral characteristics, or personal characteristics (Brehm, Kassin, & Fein, p. 39).
In an experiment there are two main types of variables, the independent and dependent. The independent variable is the variable that is causation. The IV is manipulated to see the effect. The dependent variable is the effect of the IV (Lecture, 1/24/00).
Janet’s second idea is an experiment trying to prove cause and effect. The cause is the Intro to Psychology students’ mood and having to read passages about a funeral or birthday party. The effect is their facial expression being taped as they’re reading the passages. The facial expression is being scored on the same scale as in the correlational design.
The independent variable is the mood of the Intro to Psychology students being measured according to their facial expression as they’re reading the different passages. The dependent variable is their facial expression being taped and scored on a scale of 1-5.
There is uniformity, in the fact that the participants are given either of two conditions. The control is that the experimenter exposes the participants to the conditions, and the equal treatment of the participants. Whether or not there is random assignment, I don’t know, it wasn’t disclosed. If there isn’t, this could impact the validity of this experiment and others if not used.
2.Supposing Janet carries out her first study, which is a correlational design, Janet would find a positive correlation or relationship between facial expression and mood. First, Janet observes facial expression and rates it on a scale of 1-5, 1 being negative and 5 being a positive facial expression. Second, Janet gives a questionnaire to the participants rating their mood on a scale of 1-10, 1 being negative and 10 being positive. If X is facial expression, and Y is mood, there would be a positive correlation, because as X rises in value, Y rises in value also, which equals a positive correlation. In other words, facial expressions to be scored a 1, we would expect the mood to be low or 1 also. As facial expressions scores rise, indicating positive expression, mood scores would rise also, indicating a positive mood.
There is no way Janet can conclude that mood causes facial expression. In correlational designs, there is no causation, because we don’t know if X causes Y, Y causes X, or if a third variable causes both X and Y.
An alternative explanation for the relationship with facial expression and mood is that a third variable that wasn’t measured formed the relationship. The third variable could have been that people just received a good deal at the mall or school cafeteria making their mood and facial expression positive. The people that had a negative mood and facial expression didn’t receive a good deal, at the school or mall.
3. An independent variable, in my own words, is the character of interest in an experiment that is going to be manipulated to show an effect on the dependent variable. The independent variable is the cause in cause and effect of an experiment (Lecture, 1/24/00).
The independent variable in Janet’s second study is the Intro to Psychology students. Janet plans on manipulating the student’s mood by giving half the students one passage to read and the other half, another passage. Half of the students will get a passage about a funeral, and the other half will get a passage about a birthday party to read.
A different way Janet could manipulate the independent variable would be to create a control group that got no passages to read. This would give a basis to measure everything on. This would also help in eliminating extraneous variables.
4. Janet’s dependent variable in the study is the facial expressions caused by the independent variable. Dependent variable is the effect of the independent variable. Janet plans on measuring the dependent variable with a scale of 1-5. A negative facial expression, probably one caused by the reading of the funeral passage, will get a low number of 1. A positive facial expression, probably one caused by the reading of the birthday party passage, would be scored a high number of 5. Janet is doing this by videotaping the facial expressions as they’re reading the different passages and scoring the expressions.
A different way Janet could measure facial expressions, or the dependent variable, would be to allow someone else to rate their facial expressions, besides Janet. This would create less bias, so Janet’s experiment could have more validity.
5. Random assignment, not random sampling, is a technique used by the experimenter to assign participants, who are already in the experiment, to the conditions. The participants will have an equal chance of being exposed to any of the conditions, dependent or independent variables, as any other participant in the experiment (Brehm, Kassin, & Fein, p. 39).
Random assignment is important in an experiment to have an equalization of the participants chance to be exposed to the variables and conditions. By doing this, the experimenter has greater internal validity, or the independent variable causes the changes in the dependent variable, (cause and effect) (Brehm, Kassin, & Fein, p.40, Lecture, 1/26/00).
6. In Janet’s first study, she uses the research methods of observation and survey. There are many advantages and disadvantages that go along with these methods. The advantage of observation is that the observer gets a “direct measurement of behavior,” (Lecture, 1/19/00). The disadvantage of observation is that observers may not know what they’re doing or observing correctly. Observing also has a disadvantage of low interrater reliability, where different observers can’t agree on their observations (Brehm, p. 31). The advantage of the questionnaire is that you can get information on behavior that isn’t directly observable (Lecture, 1/19/00). The disadvantage of a questionnaire is that self-esteem and self-presentations can get in the way. This can impact the way questions are answered and many other issues.
In Janet’s second study, which is an experiment, have many advantages and disadvantages, but here is one of each. The advantage of to an experiment is the high internal validity, or that the indepemdent variable caused effects of the dependent variable. The disadvantage is the low external validity, where since the experiment took place in a controlled environment, it’s hard to generalize pass that controlled environment or general public.
If I had to choose one of Janet’s studies, I would choose the experiment. I would choose the experiment because I like cause and effect, rather than a relationship. Experiments, in my words, can do away with many of the confounding variables, if done right. In a correlational design, it is harder to do away with the confounding variables, because you’re trying to form a relationship of two variables.
Brehm, S.S., Kassin, S. M., & Fein, S. (1999). Social Psychology (4th ed.). Boston: Houghton Mifflin.
Brehm, S.S., Kassin, S. M., & Fein, S. (1999). Social Psychology (4th ed.). Boston: Houghton Mifflin.