Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Unit 2: Research Methods

No description
by

Joseph Giorlando

on 11 October 2016

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Unit 2: Research Methods

When an outlier(s) skews the mean - making it no longer "typical" of the scores
Descriptive Research
Some

descriptive research methods (
ex.
surveys) can be
generalized
CANNOT
prove cause and effect
Naturalistic Observation
Sing
le cases
(
ex.
case studies)
may be misleading
Form of
descriptive research -
study of animal or human behavior in
natural settings
rather than the laboratory
When observer knows the
goals of the study
and allows this knowledge to
influence their observations
Form of
descriptive research
- study of a single individual (or a few) to
describe
their situation by gathering as much information as possible
(
ex.
psychological tests, interviews, observation, medical records)
Steps to Correlational Research
Create
Operational Definitions
Scatter Plots
relationship
no-relationship
statistical measure of relationship between two variables
(ranges from -1.0 to +1.0)
WHY IS CORRELATION IS NOT CAUSATION?
Independent and Dependent Variables
Variable
manipulated/changed
by experimenter
Design
Experimental and Control Groups
Experimental Group
In a controlled experiment, the group
subjected to a change
in the independent variable
In a controlled experiment, this is the group
NOT
subjected to a change in the independent variable
Control Group
Placebo Effect
experimental results
caused
by expectations alone
ex.
no sugar given
Random Assignment
Randomly
assign
participants to either your control or experimental groups.
Single and Double Blind Procedure
Single Blind Procedure
Only
the participant is
unaware
of the group they are in
Both

the participant and the researcher are
unaware
of the group they are in
Descriptive Statistics
Form of statistics used to describe or summarize the results
Central Tendency
Mode
Mean
Median
Most frequently occurring score
Average of the scores
Middle score in a distribution
Range
Standard Deviation
difference between lowest and highest scores
15
-
3
= 12
measure of how much scores
vary/differ
from the mean
Inferential Statistics
Used AFTER an experiment to form
conclusions
about the
EFFECT
of the
independent
variable on changes in the
dependent
variable
APA Ethical Guidelines
Ethical Principles Governing Research with Human and Animal Subjects
Goals:
Describe descriptive research studies with regard to purpose, strengths and weaknesses by taking into account different types (case studies, naturalistic observation and surveys) random sampling, wording-effect, replication, generalization and applicable biases (ex. Social Desirability Effect).
Essential Task 2-1:
Essential Task 2-2:
Essential Task 2-3:
Essential Task 2-4:
Essential Task 2-5:
Essential Task 2-6:
Describe correlational research study with regard to purpose, strengths, and weaknesses by taking into account operational definitions, random sampling, correlational coefficient, and scatter-plots.
Apply basic statistical concepts to explain experimental research findings using: Inferential Statistics: Statistical significance (t-test and p-value)
Describe experimental research design with regard to purpose, strengths, and weaknesses by taking into account operational definitions, independent/dependent variables, confounding variables, control/experimental groups, random assignment of participants, single/double blind procedures, and applicable biases.
Apply basic statistical concepts to explain research findings using descriptive statistics: central tendency (mean, median, mode, skewed distributions), variance (range, standard deviation, and normal distributions).
Identify the APA ethical guidelines, such as debriefing and informed consent, and identify how they inform and constrain research practices.
Unit 2:
Research Methods

Descriptive Research
Correlational Research
Experimental Research
Types of Descriptive Research
Naturalistic Observation
Case Study
Survey
Strengths
Weaknesses
Methods
Use Stats
Correlation Coefficient
CORRELATION IS NOT CAUSATION
Methods
Set up
Variables
Randomly Assign
Participants
Use
Single/Double Blind Procedure
Apply
Inferential Statistics
Descriptive Statistics
Central Tendency
Variation
Types of Research Methods
What:
Research Methods Foundational Terms
What:
Case Study
Strengths
Weaknesses
What:
In-depth/nonreplicable situations
Survey
Strengths
Weaknesses
What:
CAN
be generalized IF
randomly sampled
Research Method Foundational Term
random sampling/
selection
Form of research with the purpose of describing what is going on
Example:
"What are the eating habits of ARHS students?"
Strengths
Weaknesses
Some
descriptive research methods can be done quickly
Generalized
What:
Replication
Demo 16:
Naturalistic Observation and Observer Bias
Conclusion:
D
Volunteers conducted a naturalistic observation of the students in the classroom - though there was an
observer bias
influencing the observation
What:
Observer Bias
What:
Process of repeating a study (with
SAME
methods) but different subjects/experimenters - used to
RETEST
results
MUST
be able to
REPLICATE
in order to
GENERALIZE

findings
Process of applying findings to
new
situations or subjects
ex.
findings of a randomly sampled survey can be
GENERALIZED
to describe the opinion of the whole school
Studying Clive Wearing is an example of a
case study
Observer bias
Can not
replicate
Can not
generalize
Form of
descriptive research
- use of carefully worded questionnaires/interviews administered to a (large) group of people
Relatively fast/cheap
What:
Step

used during
s
urveys - each member of a population is given the
same odds
of being selected to respond
ex.
alphabetical list - randomly with number generator
M
S
for
S
urvey
S
for
S
ampling/
S
election
(this can get confusing with another term we cover -
useful mnemonic
)
population
(all of ARHS)
Sample
(determined with random number generator)
When wording changes the results of survey
(although same question is being asked)
Demo 17:
Surveys and Wording and Social Desirability Effects
Conclusion:
D
Wording
influenced students responses - although the same question was being asked
What:
Social Desirability Effect
Wording Effect
ex.
Do you support
ALLOWING
or
BANNING
tobacco ads?
What:
When asked directly about a sensitive subject - we may ALTER our answer to what we think is
SOCIALLY ACCEPTABLE
Conclusion:
When asked to share responses in public - students were willing to
ALTER
their responses
Time consuming
Easy to

replicate
M
Just like
GENERALIZATION
from last unit
Behavior is more natural than if they were in the lab
Can not
replicate
Can not
generalize
Observer bias
WITHOUT
randomly sampling
Sampling Bias
Wording Effect
Social Desirability Effect
(unrepresentative results)
Correlational Research
Allows you to make
PREDICTIONS
for future research
What:
Form of research with the purpose of showing the
relationship
between two variables and making
predictions
Strengths
Weaknesses
CORRELATION
IS NOT
CAUSATION
!!!!!!!!!!!!!!!!!!!!!!!!
Correlation
does
NOT

prove cause/effect
Curiosity about the world
Now we have an idea what is going on -
without
manipulating anything

Hypothesis creation
2 variables
"Do the variables vary together?"
Select two
variables
1
(
ex.
height, weight, happiness, salary, IQ, etc.)
2
"Does happiness vary with salary?"
What:
EXACT
description of how to derive at the
VALUE
you are measuring
(
characteristics
and
HOW
to measure the
characteristics
)
Example:
Height:
Inches, without shoes)
Salary:
Average weekly earning
without bonuses/benefits
Happiness:
3
Collect
randomly sampled/selected

data
(Why do we randomly sample?)
so we can
GENERALIZE
What:
4
Graph with points comprised by values of two variables - amount of scatter shows strength of relationship
What:
r
=

+
0.70
+
1.0
0.0
-
.50
+
.50
Strength of the Relationship
Perfect
Moderate
None
Moderate
Perfect
Strong
(this is the test grade vs. AP Exam correlation coefficient)
r
=

-
0.97
VERY Strong
-
1.0
r
=

+
0.05
VERY Weak
The
CLOSER
to
-1
OR
+1
the
stronger
the correlation
-.99
is a stronger correlation than
+.98
The AP Exam LOVES to ask this.
Positive Correlation
Negative Correlation
+
.97
+
.14
Variable 1
Variable 2
Variable 1
Variable 2
(Self-Esteem)
(Depression)
(Stamp Price)
(Year)
-
.
92
Low self-esteem and depression are strongly correlated but...
Does ice cream
CAUSE
shark attacks?
NO
Ice cream sales
and
shark attacks
on swimmers are
correlated
Third Variable:
TEMPERATURE
(beach)
Was there a strong relationship?
Time to make a prediction and consider conducting an experiment
After Experiment
Experimental Research
The
ONE
and
ONLY
way to
PROVE
that one variable
CAUSES
a change in another
What:
Form of research with the purpose of establishing cause and effect between two variables
Strengths
Weaknesses
"Does X cause a change in Y?"
Confounding Variables
Experimenter bias
Experimental Research Foundation Terms
American Psychological Association
Confounding Variable
What:
Hypothesis:
Independent Variable
What:
If
a child consumes sugar
then
there will be an increase in hyperactivity.
Variable the experimenter
measures
to see if the independent variable resulted in a CHANGE
Dependent Variable
Demo 18:
Experimenter Bias
Conclusion:
D
My preconceived conclusions about the study resulted in an experimenter bias and the creation of confounding variables (
ex.
type of chair)
What:
M
Tip:
A hypothesis in this class will be presented as an
"
If
...

then
"
statement
If
a child consumes
sugar

then
there will be an increase in
hyperactivity
.
FOLLOWS
"IF"
-
Independent
FOLLOWS
"THEN"
-
Dependent
1
Select
operationally defined
independent and dependent variables (
need to operationally define so the experiment can be replicated)
What:
ex.
participants
GIVEN
sugar
If
a child consumes sugar
then
there will be an increase in hyperactivity.

What:
ex.
participants
GIVEN
a
PLACEBO
sugar
2
3
What:
Experimental Group
Control Group
ex.
alphabetical list - random group generator
Why:
4
What:
Double Blind Procedure
What:
Why?
Variables that a researcher
FAILS
to control for
OR
eliminate
Experimenter Bias
Errors in a research study due to the prior beliefs of the experimenter
What:
ex.
my clear bias in the previous demo
Helps
REDUCE

c
on
founding variables and experimenter bias
What if the researcher put all of the gender in the control group? Adds a CONFOUNDING VARIABLE.
Random Assignment reduces that
M
Once you have participants - the next step is to put them into groups...
or in the other words - put their
into groups
random
ass
ignment
CONS
lie and these make your results a
CON
CON
founding Variable
M
Reduces
BOTH
experimenter bias
AND
placebo effect -
essential
when testing new drug treatment
5
What:
Representative
Population
(Random) Sample
Can we make conclusions?
Inferential Statistics
t-test and p-value
What:
A calculation conducted after an experiment to determine
IF
the results were
CAUSED
by manipulating the independent variable
OR
by dumb luck
you will
NEVER
be asked to calculate this
BUT
They
LOVE

asking what the p-value means
a
t-test
provides the
p-value
.100
.50
DUMB LUCK
.051
.05
Statistical Significance
What:
The difference found in the experiment was
REAL
and
NOT
by chance
IF
a p-value is
5% AND BELOW (.05)
- the results were
STATISTICALLY SIGNIFICANT
In a study of power and self-image, participants were not told the true purpose of the study; instead, they believed they were participating in a business simulation. Researchers randomly assigned participants to a high- power (n = 44) or low-power (n = 44) condition. In the high-power condition, participants recalled a time when they had power over others, and in the low-power condition, they recalled a time when others had power over them. Participants were asked to adjust the height (in centimeters) of an electronic graphical image (an avatar) of themselves to reflect their personal appearance.
Results indicated a statistically significant difference in participants’ perceptions of their own height across the two conditions.

Participants in the high-power condition created taller self-images (mean = 6.0, standard deviation = 1.5) than participants in the low-power condition (mean = 4.0, standard deviation = 1.0).
AP Psychology FRQ - 2014
(Mean: 3.42 of 7)
- Describe the levels of the
independent variable.

- Describe how the researchers measured the
dependent variable
.

- Create a
bar graph
illustrating the results of the study. Correctly
label
each axis.

- Explain why the researchers can conclude that there is a
cause-and-effect
relationship between the independent and dependent variables.

- Explain what statistical significance means in the context of the study.

- Explain why
debriefing
would be necessary in the study.
Scored
This would NOT be enough - when on to SPECIFY what that means
Scored
"Participants in the high-power condition created taller self-images."
Unlike 2015 - this year actually HAD a difference in results
(6 vs. 4)-
so the graphs needed to differ
Scored:
Did NOT Score:
As we know -
only way
to prove cause-and-effect
Scored:
Statistical significance is shown by the p-value If
BELOW 5%
(.05) - the experiment
WAS
statistically significant and
NOT due to chance
Scored:
Always connecting back to the the prompt
2007 AP Exam
Found statistically significant results?
Looks like you have yourself a theory
Other Stats: Descriptive
What:
Placebo
10 mg
100 mg
Activity
Treatment
Pre-Treatment
Post-Treatment
What:
Way of checking what is average/typical
3 Forms
Skewed Distribution
What:
ex.
adding Tom Brady to the mean income of those in this classroom would result in a
positive skew
l
E
ft
n
E
gative skew
r
I
ght
pos
I
tive skew
M
When distribution is skewed:
MEDIAN
is used to describe central tendency
Positive Skew
What:
15
12
12
8
5
3
range
What:
Small
Standard Deviation =
Scores
tend to be similar
to one another
F Block has the
smallest SD
- meaning the scores in F Block
tended to be CLOSER
than in D/E Block
(often displayed on a BAR GRAPH)
Normal Distribution (Bell) Curve
What:
Distribution curve where the mean, median and mode are all the same
ex.
intelligence tests are
designed for results to be
normally distributed
Normal distribution curves always look like this:
You will need to memorize
".02, 2, 13 , 34"
For baseball fans:

"little" Jeter, "big" Jeter, A-Rod and Ortiz
.02%
.02%
.02 2 13 34
For non-baseball fans:
Going backwards
"4, 3, 1... you forgot about the deuces!"
Step One:
Input Mean
.02%
.02%
100
High point represents the
MEAN
Example:
Mean = 100
Step Two:
Input Standard Deviation
.02%
.02%
100
Represent
"standard deviations above/below the mean"
When given the SD (ex. SD = 15)
0
1
2
3
-1
-2
-3
Add 15
each SD to the right and
subtract 15
each SD to the left
100
115
130
145
85
70
55
100 + 1 SD (15)
115 + 1 SD (15)
130 + 1 SD (15)
100 - 1 SD (15)
85 - 1 SD (15)
70 - 1 SD (15)
100
115
130
145
85
70
55
100 + 1 SD (15)
115 + 1 SD (15)
130 + 1 SD (15)
100 - 1 SD (15)
85 - 1 SD (15)
70 - 1 SD (15)
Step 3:
Using the Curve
0
1
2
3
-1
-2
-3
100
115
85
100 + 1 SD (15)
100 - 1 SD (15)
0
1
-1
34 +34 =
68%
Answer
115
85
100 + 1 SD (15)
100 - 1 SD (15)
0
1
-1
96
Answer
96
115
85
100
1
-1
Mean
= 100
SD
= 15

Which of the following scores is
within one standard deviation
of the mean?

(A) 132
(B) 77
(C) 96
(D) 62
(E) 140
Answer:
96
What percentage of scores fall between 115 and 85?
What percentage of scores fall two standard deviations above and two standard deviation below the mean?
70
130
0
-2
2
13 + 34 +34 + 13 =
95
Answer
The answer on the AP Exam will be 95 - we remove the decimals for ease
For example
- adding 10 points to
EVERY
score wouldn't change this
(though it would change ALL central tendencies )
APA Ethical Guidelines
What:
Still acting
"hyper"
What:
Highlights:
informed consent
What:
confidentiality -
unless agreed upon in advance
debriefing
What:
In
dependent is what you put
"IN"
the cup
What:
Founded in 1892 - largest scientific/professional organization of psychologists
Original President:
The
"Original G"
-
G
. Stanley Hall
M
Second President (first female):
Academic research must be approved by APA ethics board
(based on APA Ethical Guidelines)
Margaret
F
loy
W
ashburn
(
F
-
W
for
F
irst
W
oman)
To Note:
Percentiles
.02%
.02%
If you ever see the term "percentile" regarding this curve - they are just referencing the %'s in each deviation
50th percentile
84th percentile
100th percentile
(100%)
(50%)
(5.1%)
(5%)
the process of describing the research
AND
obtaining the subjects' consent based on an understanding of the project's methods and goals -
IF
possible

(deception is allowed)
Process of giving participants in a research study a complete explanation of the study after the study is completed
Do Now
Which APA ethical guideline do you believe is the most important and why?
Full transcript