Chapter 2: Research Methods - Suffolk County Community College

Report
Chapter 2: Research
Methods
Dr. Mary Streit
Suffolk Community College
Chapter 2 – Research Methods
Goal of Research: to find the truth! What is really
going on??
How can we conduct research to get at the truth?
1. Adhere to the scientific method
2. Make sure your study is falsifiable or subject to
replication. Include operational definitions.
3. Propose a theory that is parsimonious or not
overly complicated. Keep it simple whenever
possible.
Chapter 2 – Research Methods
Two major categories or types of research:
Basic
- laboratory setting
- Theory building
Applied
- field or real-world setting
- solve a problem
Chapter 2 – Research methods
Two types of research methods:
1. Non-experimental.
- hands-off: no control
- real-world setting
2. Experimental.
- hands-on: control
- laboratory setting
Which category seems most similar to applied
research? Basic research?
Chapter 2 – Research Methods
Non-Experimental Research Methods.
1. Naturalistic observation.
- AKA observational research
- meets first goal of psychology – Description
- answers the question “What?”
- qualitative
- operational definitions & hypotheses emerge
-Jane Goodal and Chimpanzees
-Dian Fossey and Gorillas
Chapter 2 – Research Methods
Limitations:
- lengthy and time-consuming
- observer bias: AKA experimenter bias.
When the researcher is biased in his/her
observations – usually in the direction that
supports his/her hypothesis.
- observer effect / reactivity: When people
behave differently when they know they
are being watched or observed.
Chapter 2 – Research Methods
Non-Experimental Research Methods [cont’d]
2. Case study.
- A detailed description of an individual with a rare
or unusual disorder
- In-depth analysis
- Can be used to describe and explain behavior
- Limitations: results can not be applied to the
general population
- Memory problems
- Famous cases: patient H.M., Phineas Gage,
Genie “the wild child”
- For more information on Genie’s case see:
http://www.psychology.sbc.edu/cesarz.htm
Phineas Gage
for more info: http://www.deakin.edu.au/hmnbs/psychology/gagepage/
Patient H.M.
For more information on H.M., see the link below:
http://thebrain.mcgill.ca/flash/capsules/pdf_articles/patient_hm.pdf
Chapter 2 – Research Methods
Non-Experimental Research Methods.
Surveys.
- questionnaires that allow psychologists to
measure covert behaviors: attitudes, opinions,
values, beliefs, etc.
- can be easily used to measure large groups
of people
- more quantitative
- can be used to describe, explain, and even
predict behavior
Chapter 2 – Research Methods
Chapter 2 – Research Methods
Non-experimental Research Methods.
Surveys.
- population: everyone (the entire pizza pie)
- sample: not everyone or a subset of the
population( a slice of the pie)
You want your sample to be representative of
the population. Why?
So you can generalize your results to the
population. (e.g. - You don’t have to eat the
entire pizza pie to see how it tastes. You can
just eat one slice!)
Chapter 2 – Research Methods
Non-Experimental Research Methods [continued]
3. Survey limitations
- Sample bias: when your sample is different from the
population
- Example: When you happen to eat the only slice in the
entire pie that is pepperoni. This is problematic if you
erroneously assume that the rest of the pie is also
pepperoni.
- Sample results can NOT be generalized to the population
- 1934- Literary Digest. Al Landon vs. FDR: famous case of
sample bias. Sampled all US citizens with phones.
- Random samples eliminate sample bias
- Random sample: when every element in the population
has an equal chance of being selected into the sample.
- Lotto number drawings
Chapter 2 – Research Methods
Chapter 2 – Research methods
If you are interested in finding out how many
hours per week the average full-time
college student in the US studies, what
would be your population?
Your sample?
Which study would you prefer to do?
Why?
Chapter 2 – Research methods
Non-experimental Research Methods [cont’d]
Survey limitations
- social desirability: When people change their
actual attitudes, values, beliefs, etc. in order to
please the experimenter and/or to appear
politically correct.
In other words, when people lie on a survey to
make themselves look or feel good.
- anonymous responses – helps to eliminate
this problem somewhat
- teen survey of sexual activity by the National
Institute of Health in 1997 vs. 2000. Huge
increase in abstinence reported. Why??
Chapter 2 – Research Methods
Non-Experimental Research Methods.
4. Correlational research.
- examines the relationship between two
variables as they occur naturally
- hands-off
- no experimenter manipulation or control
- real world data
- heavily quantitative
- correlation coefficient (r2)
- frequently used to predict behavior
Chapter 2 – Research methods
Non-Experimental Research Methods.
Correlation coefficient.
- A number with a range of -1.0 and +1.0 that
expresses numerically the relationship between
two variables.
- The sign of the number indicates the direction the
two variables are moving in relation to each other
- A positive correlation means both variables are
moving in the same direction (e.g. – temperature
outside and # of ice cream cones consumed)
- A negative correlation means both variables are
moving in opposite directions (e.g. – temperature
outside and the number of layers of clothing)
Chapter 2 – research methods
Non-experimental research Methods.
Correlation coefficient (cont’d).
- The number itself represents the strength of the
relationship. The closer the number is to 1.0,
regardless of the sign, the stronger the
relationship.
- For example, which number represents the
strongest correlation coefficient?
a.
b.
c.
d.
-.72
+.63
-.56
+.49
Chapter 2 – Research Methods
Non-experimental Research Methods.
Correlation coefficients.
What do the numbers mean?
Say you have 2 variables: height and IQ,
and you want to find out if there is a
relationship between these two variables.
Because you can NOT control an
individuals height or IQ, you decide to do
correlational research.
Chapter 2 – Research Methods
- If you have a perfect +1.0 correlation, it means that for
every 1” increase in height, you will see a 1 point
increase in IQ. In other words, there is a perfect 1 to 1
correspondence.
- If you have a +.90 correlation, it means that for every 1”
increase in height, you will see a +.90 increase in IQ.
- If you have a +.80 correlation, it means that for every 1”
increase in height, you will see a +.80 increase in IQ.
- What about a negative correlation coefficient? See if
you can figure out what would happen in the above
examples if the numbers were negative instead of
positive?
Chapter 2 – Research Methods
Non-Experimental Research Methods [cont’d]
4. Correlational research Limitations.
- correlation does NOT mean causation!!!
- Why do we get colds more frequently in the
winter months? Is it because it is cold
outside? What is really going on?
- in correlational research, there is a lack of
experimenter control over other factors
- these other factors are called: extraneous
variables, third variables or confounds 
- only in experimental research can we
conclude that there is a cause and effect
relationship - by controlling for other factors
Chapter 2 – Research Methods
The human sneeze can travel at speeds well
beyond 100 mph and can reach as far as 30ft
Non-Experimental Research Methods: Matching Review
1.
2.
3.
4.
5.
6.
7.
8.
9.
an in-depth analysis of a single individual or
subject
used to measure the covert behaviors of many
people
subject to the problem of observer bias
when people behave differently when they
know they are being watched
when people lie on a survey in order to look
good to the researcher
when the experimenter intentionally looks for a
behavior to occur
when the people in your sample are different
from the population
The entire group of people that you are
interested in studying
A method used when you are trying to
determine the relationship between two
variables that you can not control
a. Observer bias
b. Observer
c.
d.
e.
f.
g.
h.
i.
j.
effect
Sample
Population
Sample bias
Social
desirability
Survey
Case study
Naturalistic
observation
Correlational
research
Non-Experimental Research Review
1.
2.
3.
4.
5.
6.
7.
8.
9.
A relationship between two variables where
one variable increases as the other
decreases
A relationship between two variables where
one variable increase as the other increases
A detailed description of behaviors observed
in their naturally occurring environment
A subset of the population
factors that were not controlled for that may
be causing the effect observed
When every element in the population has an
equal chance of being selected into the
sample
its main limitation is that the results can not
be generalized to the entire population
A type of research used to solve a problem in
the real world
A type of research used to build a theory
a. Observational
b.
c.
d.
e.
f.
g.
h.
i.
j.
research
Sample
Population
Confounds
Random sample
Case study
Basic research
Applied research
Positive
correlation
Negative
correlation
Chapter 2 – Methods
Experimental Research.
- Typically conducted in a laboratory setting
- Does not occur in the real world
- Experimenter has control over outside
factors or confounds
- Experimenter intentionally manipulates or
changes factors
- Hands-on: experimenter manipulates and
controls variables in the study.
Chapter 2 - Methods
There are two types of variables in
experimental research:
1. Independent variable (IV): the variable that
is intentionally changed or manipulated by
the researcher. [the cause]
- sometimes you have different levels of the
independent variable. For example, the
milligram levels of a new drug.
2. Dependent variable (DV): the variable that
is used to measure any change. [the effect]
Chapter 2 – Research Methods
- In experimental research, any change in
the DV is seen as a direct result of the IV
due to experimenter control
- Potential confounds are controlled for
- The independent variable usually precedes
or comes before the dependent variable in
time.
Chapter 2 - Methods
Typically, there are at least two groups of subjects in
experimental research:
1. Experimental group. This is the group of subjects
who receives the independent variable that you are
most interested in.
2. Control group. This is the group of subjects who do
NOT receive any form of the independent variable.
The control group is the standard to which all other
groups are compared.
Control groups allow researchers to rule out confounds
In a well-designed study, subjects are randomly
assigned to either the experimental or control group
Chapter 2 – Methods
Assignment: Labeling variables
In each example, label the IV, the DV, the
experimental group and the control group. If a
control group is lacking, add one to the study.
1. Researchers are interested in the effects of
violent TV on aggressive behavior in children.
100 Children in 2nd grade are randomly
assigned to watch a video. Half of the children
are shown a ½ hour video of Barney, and the
other half of the children are shown a ½ hour
video of the Power Rangers. After the videos
are shown, all children are given a ½ hour of
recess. Any violent or aggressive behavior is
recorded.
Chapter 2 - Research Methods
2.
3.
Researchers are interested in the effects of
testosterone on violent and aggressive behavior. Male
rats are given injections for 1 month of the following:
10 mg of a saline solution, 10mg of testosterone, 20
mg of testosterone, or no injections at all. Any violent
or aggressive behaviors were noted.
Researchers are interested in the effects of a new drug
on children with ADHD and their ability to pay attention
in the classroom. 200 ten year old boys with ADHD
were randomly assigned to one of four conditions: no
drug, placebo drug, 10mg of new drug, and 10mg of
ritalin (the drug that has often been used to treat
ADHD). Their ability to pay attention in the classroom
was measured after the drugs were administered.
Chapter 2 - Methods
Read the paragraph below and then see if you can
answer the following questions:
Researchers are interested in whether or not
smoking marijuana causes an increase in
appetite. 100 subjects are randomly assigned to
smoke either a marijuana cigarette, or a regular
nicotine cigarette. After the subjects finish
smoking, they are given a gallon of vanilla ice
cream, and told that they can eat as much ice
cream as they would like. The # of grams of ice
cream consumed is their measure of appetite.
What is the IV in this study? The DV?
Chapter 2 – Research Methods
- Results: Subjects who smoked marijuana
ate significantly more ice cream than those
who smoked the regular nicotine cigarette.
- Can you conclude with confidence that
smoking marijuana led to an increase in
appetite?
- What other potential confounds could be
at work here?
Chapter 2 – Research Methods
Potential confounds:
- Lactose intolerant
- Gender (Male/Female)
- When the subject last ate
- How much the subject likes ice cream
What if all confounds are controlled for. Can you
then conclude with confidence that smoking
marijuana leads to an increase in appetite?
Chapter 2 - Assignment
Critical Thinking Assignment#1: select three articles
from the links below and see if the headline is true or
false based on the actual study described in the
article. 5 pages minimum APA style.
Be sure to include: A brief summary. What type of
research method was used in each study? Was it
experimental? If yes, label any variables: IV? DV?
Experimental group? Control group? Etc. Was is
non-experimental? If yes, which non-experimental
method was used? What was the sample? How was
the sample selected? Was it random? For all studies,
what were the conclusions? Do you agree with their
findings? What are some potential confounds? Type
this up, and hand it in for credit for your paper.
Assignment #1 links
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Music lessons improve kids' brain development, memory
Authoritarian parents have fat kids
Chocolate really does make us feel better
Anorexia largely determined by heredity
Soda causes obesity, researchers assert
Spanking children fuels aggression, anxiety
Eating pizza cuts cancer risk
Breastfeeding fights arthritis
Church attendance is good for your health
Video games increase aggression
Higher beer prices cut gonorrhea rates
Early unsolicited sexual encounters leads to a life of crime
"Wash Your Hands" signs only work for women
Luckiest people born in the summer
ANY type of TV program harms toddlers
Snooze or lose: memory retention enhanced by sleep
Research Methods – just for fun 
Funny article to read about correlation and
causation.
Correlation and causation

similar documents