Terminology
Experiment: A research method in which the investigator actively manipulates one or more
factors to observe their effect upon some behavior, while controlling other relevant factors.
EXAMPLE:
This spring and summer I am growing tomato plants in my backyard. I am using seed starter boxes
for the first time and wasn’t sure how compact to pack the dirt in the seed starter boxes. In one
box, I packed the dirt in firmly and in the other box I packed the dirt in loosely and I labeled the
boxes so that I would remember which was packed firmly or loosely. Throughout the spring I
have been monitoring the growth of the plants in centimeters to see which was the best way to
pack the dirt and in the summer, I measured the plants as well as counted the number of tomatoes
each plant produced to see if there was a difference between the way in which the dirt was packed.
The results will influence which way I plant my tomato seeds next year IV= A factor or condition
that is deliberately manipulated to determine if it causes change in behaviors or conditions. DV=
A factor or condition that is measured at the end of an experiment and is presumed to vary as a
result of the independent variable.
Hypothesis:
The first step in a scientific investigation is to translate a theory or an intuitive idea into a testable
hypothesis. Scientists make and test predictions called hypotheses. A hypothesis is a tentative
statement about the relationship between two or more variables. Variables are any measurable
conditions, events, characteristics, or behaviors that are controlled or observed in a study. If we
predicted that putting people under time pressure would lower the accuracy of their time
perception, the variables in our study would be time pressure and accuracy of time perception.
To be testable, scientific hypothesis must be formulated precisely, and the variables under study
must be clearly define. For the above example, I would need to make a hypothesis, which is a
testable statement made about what I expect to find regarding the relationship between the IV
and DV. For example, for the above study I may make the hypothesis that seeds planted in
firmly packed dirt will demonstrate better growth. At the end of the study I will be able to see
whether my hypothesis was correct or not. My hypothesis should be based on some prior
knowledge or theory and I need to have some sort of comparison group with which I am
comparing it.
Independent Variable (IV): The experimental factor or factors an investigator manipulates, i.e.,
the treatment itself. A factor or condition that is deliberately manipulated to determine if it causes
a change in behaviors or conditions. A factor may have different levels. For example, in the power
point when we reviewed a study of the best way to plant tomatoes the independent variable was
compactness of the dirt (firm or loose) and had two levels (Level 1= firm, level 2= loose). There
was one IV with 2 levels.
1
Dependent Variable (DV): The behavior you observe, or the outcome that might be affected
by the IV. The dependent variable will always involve some measurement. For example, the DV
when you measured the best way to plant tomatoes was (1) growth in centimeters and (2) number
of tomatoes produced. In this case there were two different ways the outcome was measured (2
different DVs).
Operational Definition: A definition that specifies the procedures used to measure some
variable or to produce some phenomenon.
Experimental Condition: The condition in which participants are exposed to the IV.
Control Condition: The condition in which participants are NOT exposed to the IV. This
condition provides a baseline for comparison with participants exposed to the independent
variable.
Confounding variables and Extraneous variables:
Experimenters concentrate on making sure that the experimental and control groups are alike on
a limited number of variables that could have a bearing on the results of the study. These
variables are called extraneous, secondary, or nuisance variables. Extraneous variables are any
variables other than the independent variable that seem likely to influence the dependent variable
in a specific study. If possible, it is good to identify what variables may be extraneous and
control for them.
“Does misery love company? This question intrigued social psychologist Stanley Schachter.
When people feel anxious, he wondered, do they want to be left alone, or do they prefer to have
others around? Schachter’s review of relevant theories suggested that in times of anxiety people
would want others around to help them sort out their feelings. Thus, his hypothesis was that
increases in anxiety would cause increases in the desire to be with others, which psychologists
call the need for affiliation. In Schachter’s study, one extraneous variable would have been the
participants’ tendency to be sociable. Why? Because subjects’ sociability could affect their
desire to be with others (the dependent variable).
If the participants in one group had happened to be more sociable (on the average) than those in
the other group, the variables of anxiety and sociability would have been con-founded. A
confounding of variables occurs when two variables are linked in a way that makes it difficult to
sort out their specific effects. When an extraneous variable is confounded with an independent
variable, a researcher cannot tell which is having what effect on the dependent variable.
Unanticipated confoundings of variables have wrecked innumerable experiments. That is why
so much care, planning, and forethought must go into designing an experiment. A key quality
that separates a talented experimenter from a mediocre one is the ability to foresee troublesome
extraneous variables and control them to avoid confoundings.” (Weiten, 2016)
Weiten (2016). Psychology: Themes and Variations, 10th Edition. Belmont, CA: Thomas Wadsworth.
2
Ethics in research
Figure 2.12 Ethics in research
- For further review of ethical requirements see textbook pages 53-57 (pdf edition 52-55), and
slides 38-39 (ebook section 2-5).
- For further information regarding evaluating research see textbook pages 49-53 (pdf edition
47-52) and slides 20 & 37 (ebook section 2-4).
3
Causation: The cause-and-effect relation in which a given event, the cause,
produces an observed result, the effect. For example, in the experiment in which
a neural impulse was measured, the longer distance from ankle to brain than from
shoulder to brain caused people to react more slowly. Experimental design and
methodology determines whether or not cause-and-effect statements can be made.
•
If I observe that people who drive sports cars get more tickets than people
who drive family cars, can I say that driving a sports car causes you to get
more speeding tickets?
•
Suppose I randomly choose people and randomly give them a sports car or
a family car to drive for 6 months and track the number of tickets they get.
At the end of 6 months, the people driving sports cars received more
tickets. Can I conclude that driving sports cars results in more tickets?
•
In which example can I make this causal statement with more confidence?
It’s how you design your research study that allows you to be more or less able to
make causal statements with any kind of certainty.
Observational Studies: Observational studies involve the observation of
phenomena of interest as they present themselves to the scientist. Unlike active
manipulation in an experiment, observational studies involve examination of the
relation between two variables as they naturally occur in the environment.
The advantage of observational studies is that the relation between variables that
cannot be manipulated experimentally can be studied (e.g., poverty and nutrition,
social isolation and language development). The disadvantage is that cause-and
effect relations between the variables are much more difficult to establish.
For example, let's assume that self-esteem of attractive people is higher than the
self-esteem of average-looking people. It is impossible to say that attractiveness
causes people to feel good about themselves (have high self-esteem). There is
only a relation between the two. The first factor could cause the second factor;
the second factor could cause the first factor; or the relation may be caused by
some third factor. For example, attractive people may be given more
opportunities (a third factor) to experience or learn new things by the people
around them and derive their high self esteem from good performance in these
situations.
4
Correlation: A correlation is a statistical measure of the relation between two
factors that varies from -1 through 0 to +1. As a correlation approaches +1 or -1
the relation becomes stronger. Negative and positive correlations of the same
value (e.g., +0.75 and -0.75) are equally strong. Just because a correlation is
negative does not mean it reflects a weaker relation than a positive correlation.
Correlations of 0 or near 0 indicate that there is no relation between the two
factors.
Positive Correlation: Both factors increase together (e.g., age
and the amount of hair losses in men, amount of time studying and
grades). Positive correlations are above 0.
Years of education
Positive
20
16
12
8
4
0
0
5 10 15 20 25
Annual Income in Thousands
Negative Correlation: As one factor increases, the other
decreases (e.g., as temperature decreases the number of babies
conceived increases, as self-esteem increases anxiety decreases).
Negative correlations are below 0.
Negative
Dental problems
requiring care
20
16
12
8
4
0
0
5 10 15
20 25
Annual income in thousands
5
Name
Course and section number
Assignment 1:
Designing an Experiment
Your professor has designed a research study to examine the effects of alcohol
on reaction time. She has 24 subjects. Eight receive 2 ounces of vodka, eight
students receive 4 ounces of alcohol, and eight students receive 4 ounces of
alcohol flavored and scented non-alcoholic beverage. Following consumption
of the beverage, they play a video driving game in which their responses are
timed. They do not know what they are drinking nor what the purpose of the
study is, but the person administering the game does know whether and how
much alcohol they received. Using the above study, answer the following questions.
•
What is the hypothesis? ( points)
•
What is/are the Independent variable(s). How is it/are they operationally
defined?(this is where you discuss the levels) (1 points)
•
What is/are the Dependent variable(s). How is it/are they operationally
defined? (1 points)
•
What is the Experimental group(s)? ( point)
•
What is the Control group? ( point)
•
Describe some problems with this study. How well does it meet the
ethical requirements? (
points)
•
What biases (more than one) in research may be a factor in this study
and how have these issues been addressed or could researchers address
these issues? (
s
points)
6
Purchase answer to see full
attachment