Hypothesis vs. Theory
Hypothesis
 Testable predication
Lets us accept, refect or revise a theory
Ex.
If families do not stress gender differences then there will be fewer sex differences in siblings
Theory
EXPLANATION
Predicts behaviors or events
1. Fit the known facts
2. Predict new discoveries
3. Be falsifiable
4. Be simple
Start with observations
Observe and describe the world with
descriptive research
Form hypothesis from your observations
Test your hypothesis and retest your hypothesis
If it holds up
Theory
Descriptive Research
Purpose
:
Strengths
:
 Certain descriptive research methods can be quick

Some
descriptive research methods can be generalized (applied to more than those you sampled)
Weaknesses
:
 Cannot help you predict
 Can't give you cause and effect
Naturalistic Observation
 Single cases may be misleading
What is it?
Descriptive research method
 Study of animal or human behavior in natural settings rather than the laboratory
Strengths
Weaknesses
 Behavior is more natural than if they were in the lab
 Can not
replicate
Repeating a study using the same methods, different subjects, and different experimenters. Without replication you cannot generalize the results
 Can not
generalize
(apply to new situations)
 Observer bias
 Occurs when the observers (or research team) knows the
goals of the study
and allow this knowledge to
influence their observations
Case Study
What is it?
Descriptive research method
 Study of a single individual (or a few) to
describe
their situation
How?
Gather as much information as possible
 Psychological tests, interviews, observation, medical records
Strengths
 Take advantage of nonreplicable situations
 Indepth understanding
Weaknesses:
 Observer bias
 Cannot generalize
 Cannnot replicate
Survey
What is it?
Descriptive research method
 Questionnaires or interviews administered to a select group of people
 Carefully worded
Strengths
 Generates a lot of information for a
LOW
cost
 If
randomly sampled
You can generalize your findings to the population
Random Sampling
 Sampling in which each potential population member has an equal chance of being surveyed
ex
. Alphabetical list and select every 10th name
Why not give everyone a survey and use the responses?
NonResponse Bias

Women and Love study
 1974
 98% Dissatisfied by their marriage
 75% Extramarital affairs!
Why?
Only 4% of those who were sent surveys responded
(reality: 93% of women satisfied with 7% having had an affair)
Weaknesses
Wording Effect
Wording can change the results of a survey
Should cigarette ads be allowed on television?
Should cigarette ads be forbid on television?
Social Desirability Effect
If directly asked about a sensitive question
May alter our answer to appear more "socially acceptable"
Describe what is in reality
Demo 3
: Naturalistic Observation
Correlational Research
Purpose
:
Strengths
:
 If you know how they are related you can predict outcomes
Weaknesses
:
Show relationship between two variables
CORRELATION IS NOT CAUSATION
Operational Definition
 Make predictions (ex. relation between SAT scoes and college success)
 Exact description of how to derive a value for a characteristic you are measuring.
 Precise definition
 How collectors are to measure the characteristic
Variables:
Height:
Weight:
Salary:
Size of big toe:
(inches without shoes)
(lbs with undergarments)
(annual salary including bonuses/benefits)
(mm from top of joint to bottom of toe)
Correlational Research
 Two different dependent variables
Are these measurements unrelated to each other or are they related?
(randomly sampled participants)
Scatter Plots
 Graph
 Points generated by values of two variables
 Slope depicts the direction
relationship
norelationship
Correlation Coefficient (r=)
When one trait or behavior varies with another > we say the two correlate
r
=
+ 0.37
Correlation Coefficient is a statistical measure of relationship between two variables
+/ indicates direction of relationship (positive or negative)
Do Now:
Careers
Study of Low Self Esteem and Depression
 You conduct research because you assume the two are related
Two Variables
variable 1 =
variable 2=
Score on selfesteem test
Length of a bought with depression in months
WHY IS CORRELATION IS NOT CAUSATION?
 Children with big feet reason better than children with small feet
So feet size causes level of intelligence?
 AGE
Confounding Variable Problem
Relationship other than casual might exist between two variables
 Possibility of another variable or factor causing the outcome
Ice cream sales and shark attacks on swimmers are correlated
Illusory Correlation
 The perception of a relationship when none exists
Arthritis patients believe their condition was correlated with the weather
ex.
Redelmeier and Tversky (1996):
Correlation close to 0
Experimental Research
Purpose
:
Establish cause and effect relationships between variables
Strength
:
You can find out if one variable (IV) causes a change in another (DV)
Weakness
:
Possibility of confounding variables and experimenter bias
Review of Independent and Dependent Variables
Independent Variable
 manipulated by experimenter
 cause (what you are studying)
Dependent Variable
 measured by the experimenter
 effect (result of the experiment)
Independent
Dependent
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
. decaf coffee
Still acting "hyper
3
. Random Assignment
Randomly assign participants to either your control or experimental groups.
Random
Sample
= Survey
Random
Assignment
= Experiment
ex
. list of ten participants (alphabetical, everyother name)
4
. Single and Double Blind Procedure
Single Blind:
Only the participant is unaware of the group they are in
Double Blind:
Both the participant and the researcher are unaware of the group they are in
Descriptive Statistics
Used to measure and describe characteristics of groups
Central Tendency
Mode:
Mean:
Median:
Most frequently occurring score
Average of the scores
Middle score in a distribution
Skewed distribution
Lacks symmetry around their average value
Median = better measure of the center
Measures of Variation
1.
Range
2.
Standard Deviation
difference between lowest and highest scores
3
15
15  3 = 12
Measure of how much scores vary around the mean score
Small Standard Deviation = Scores tend to be similar to one another
Mean, Median and Mode are the same
ttest and pvalue
We have results from experimental research
Inferential Statistics checks if it was dumb luck
ttest
Compares results of experimental/control groups
Gives you
pvalue
pvalue
 Measures confidence in experimental results
If
pvalue
is less than 0.05 (5%)
statistically significant
Results of experiment
were not
by chance
Does caffeine improve our reaction time?
Caffeine condition has a lower mean reaction time.
We run a ttest on our samples and get:
p = 0.039
Can we be confident that the difference in the data is not due to chance?
Ethics
Ethical Principles Governing Research with Human and Animal Subjects
1.
Discuss which guidelines you believe are most important.
2.
List how ethical issues inform and constrain research practices.
Unit 2:
Research Methods
Goals:
Differentiate types of research with regard to purpose, strengths, and weaknesses:
 Descriptive Research (case studies, naturalistic observation and surveys)
 Correlational Research
 Experimental Research
Essential Task 21:
Essential Task 22:
Essential Task 23:
Essential Task 24:
Essential Task 25:
Essential Task 26:
Essential Task 27:
Describe descriptive research studies taking into account random sampling, wordingeffect 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).
Describe a correlational research study taking into account operational definitions, random sampling, correlational coefficient, and scatterplots.
Describe experimental research design 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 experimental research findings using: Inferential Statistics: Statistical significance (ttest and pvalue)
Identify the APA ethical guidelines, such as debriefing and informed consent, and identify how they inform and constrain research practices.
Descriptive Research
Correlational Research
Experimental Research
Types of Descriptive Research
Naturalistic Observation
Case Study
Survey
Strengths
Weaknesses
Strengths
Weaknesses
Strengths
Weaknesses
Methods
Stats
Scatter Plots
Correlation Coefficient
CORRELATION IS NOT CAUSATION
Methods
Set up
Variables
Randomly Assign
Participants
Use
Single/Double Blind Procedure
Apply
Inferential Statistics
After Experiment
Descriptive Statistics
Central Tendency
Variation
Types of Research Methods
Present Remotely
Send the link below via email or IM
CopyPresent 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.
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.
Unit 2: Research Methods
No description
by
Tweet