Chapter 2
Research Methods
Research Methods
How do you do research in psychology?
Psychology uses the scientific process:
Question
Hypothesis
Data
Interpretation
Warm up with Crash Course’s Methods video
Research Methods
Hindsight bias is the tendency of people to overestimate their
ability to have predicted an outcome that could not possibly
have been predicted.
However, the goal of science is to be predictive; that is,
determine an outcome before it happens, not after.
Research Methods
Ways of doing research in psychology (research design):
There are different ways of doing research:
Experimental design (experiment)
- Laboratory experiment (usually just “experiment”)
- Quasi-experiments (like an experiment*)
- Field experiments (like an experiment*)
Naturalistic observation
Correlational design (correlation)
Survey
Case Study
Research Methods
What is the relationship between these variables:
Coffee intake and running speed?
Showering and body odor?
Studying and grades?
Mass of an object its gravitational force?
Research Method Terms
Hypothesis: Usually an ‘if, then’ statement or simply a
prediction about some event.
Theory: aims to explain a broad set of phenomena.
Hypotheses (multiple hypothesis’) support a theory.
Independent variable: The variable that influences the
dependent variable.
Dependent variable: A variable that depends on the
independent variable.
Research Method Terms
Operational Definitions
Hypothesis: If people are given money, they experience
greater happiness than if given candy.
How do we measure happiness?
Operational Definitions (IMPORTANT)
Definitions of variables in research need to be quantifiable
(countable) and observable. They need to be operationally
defined.
Example, happiness: how many times someone smiles.
Validity & Reliability
Research aims to be valid and reliable
Validity refers to whether the research measures what
the researchers set out to measure.
Reliability refers to whether the same results can be
produced under similar conditions.
Validity & Reliability Example
If you have a scale, and it says you weigh 100lbs., but on
every other scale you step on, the scale says 180lbs., the scale
would not be a ‘valid’ measure of your weight.
On the other hand, if you stepped on the scale and it said
100lbs., and then five minutes later stepped on the same scale
again, and it said 130lbs., the scale would not be a ‘reliable’
measure of your weight.
Research Method Terms
Participants are the people or subjects in your study.
Sample/Sampling: A sample is a set of participants or things
taken from a population. Sampling refers to the selection of
participants.
Population is the large set of individuals from which a
sample was taken.
Representative sample: A sample that accurately reflects the
larger population.
Research Method Terms
Example of population and sample
Research Method Terms
Example of population and sample
Understanding Terms Example
Wearing a red shirt in a large crowd gets you more attention.
What is my hypothesis?
What must be operationally defined?
Who is my sample?
Who is my population?
Is the sample representative?
Understanding Terms Example
What is a representative sample?
A representative sample better reflects the participants of a
population compared to an unrepresentative sample**
Therefore, whatever research result you find, the results will
be able to be generalized to the population.
Understanding Terms Example
Random Selection
Random selection means that every member of the
population has an equal chance of being selected.
How could we randomly select in the previous
example?
Random Selection
More on Random Selection
Stratified sampling: A process that allows a researcher to
ensure the sample represents the population on some criteria.
If I want to research whether different racial groups respond
differently to a survey, I could select 10 Caucasians, 10
Asians, 10 African Americans….
Stratified sampling takes place at the selection level. If you
want your sample to meet some criterion or criteria, then that
is stratified sampling*
Experimental Method
Psychologists prefer experiments because they can establish a
cause-effect relationship.
Laboratory experiments are conducted in a controlled
environment and use random assignment.
Field experiments are conducted in the ‘real’ world.
Researchers go out and manipulate some variable and observe
the effect. Example here.
Quasi-experiments are experiments, but do not use random
assignment.
Experimental Research
Confounding variables
A researcher must try to isolate variables or control for
“confounding” variables
Confounding variable: A confounding variable is any
difference between the experimental controls and the control
conditions, except for the independent variable, that might
affect the dependent variable.
Confounding variables
If I am studying whether the amount of time spent studying is
associated with better grades, what are variables that may
influence this relationship other than studying?
Groups
An experiment must have a control group and an
experimental group.
The experimental group is that which receives the
independent variable.
The control group is the group that does not receive the
independent variable.
Understanding Terms
Hypothesis: Drinking coffee before running makes one run
faster.
After gathering a sample, who is my experimental group and
who is my control group?
Terms
Assignment: Assignment is the process by which participants
are put into a group, experimental or control.
Random assignment: The process of randomly assigning
participants to either the control or experimental group.
Terms
Terms
What if in my coffee experiment, I allow
participants to join which ever group they want?
Participant-relevant confounding variables: If
participants were given the opportunity to choose
which group to be in, the results might become
biased.
Terms
Group-matching: If one wanted to match for sexual
orientation, eye color, skin color, or something other variable,
then group matching is the process of assigning individuals to
groups based on some criteria.
Situation-relevant confounding variables: When conducting
an experiment, both groups (experimental and control) must
be subject to the same environment.
Terms
Experimenter Bias: The tendency for researchers to treat
members of the experimental and control groups differently.
If the aim of my study is confirm my hypothesis, I may
interpret the results differently than someone who is not
involved or does not have an interest.
Controlling for Bias: Experimental
Designs
Single Blind: Participants do not know whether they’re in the
experimental group or the control group.
Double Blind: Neither the participants nor the research
knows who is in which group.
Placebo & Placebo Effect
Placebo is a substance that has no therapeutic effect; it is
often used in control groups for testing new drugs.
Placebo effect: Participants feel a “psychological” effect, but
have not been given a real substance that would cause “real”
physiological differences in their body.
Placebo Effect
Putting it all together
Correlational Method
Correlation is a statistical measure that indicates the extent to
which two or more variable fluctuate together.
A correlation expresses a relationship between two variables
without ascribing a cause.
Correlation does not equal causation.
Correlations
Correlational Method
Correlations
A positive correlation between two things means that the
presence of one thing predicts the presence of the other.
A negative correlation means that the presence of one thing
predicts the absences of the other.
Correlations
Predict the correlation
Drinking water and feeling thirsty
Studying and amount of free time
Colour of t-shirt worn on exam day and exam grade
Number of hours spent studying and grades
Amount of coffee drunk and talkativeness
Staying up all night and fatigue
Number of fictional books read and driving ability
Correlations
Why use correlations, why not always conduct an
experiment?
Sometimes correlations are the best research method because
you cannot manipulate the data in an experiment.
Does weather affect shop lifting? Is there any way we can
manipulate the weather?
Surveys
Conducting a survey involves asking participants to complete
a questionnaire.
Using the survey method means that one cannot control for
certain confounding variables.
Survey method is subject to the social desirability effect: the
tendency of survey respondents to answer questions in a
manner that will be viewed favorably by others.
Naturalistic Observation
Naturalistic observation: Researchers go out and observe
participants in their natural habitat without interfering at all.
Naturalistic observation is different from field experiments
because in naturalistic observation, researchers do not
interfere.
Case Studies
The case study method is often used in clinical psychology.
Case studies allow researchers to get a picture of a small
group of people (as little as one), but this means that the
findings cannot be generalized to any population.
An example of a case studies may be studying a rare
psychological disorder.
APA Ethical Guidelines
Ethical considerations are a major part of research regarding
human and animal research.
Guidelines are established by the American Psychological
Association (APA) for human and animal research.
APA Ethical Guidelines (Animals)
1. Must have a clear scientific purpose.
2. Questions of research must be important.
3. Animals chosen must be suited for the question.
4. Animals must be cared for in a humane way.
5. Animals must be acquired legally.
6. The experimental procedures must be ones which employ
the least amount of suffering possible.
APA Ethical Guidelines
(Humans)
1. No coercion: (participation must be voluntary).
2. Informed consent: (participants know and agree with
what you’re doing).
3. Anonymity/confidentiality: participants privacy
protected.
4. Risk: as little as possible (physical and psychological.
5. Debriefing: after experiment is done, you inform
participants about the nature of the experiment.
Research Methods
Slides continue
Quizlet

Methodology in Psychology for A-Level Classes

  • 1.
  • 2.
    Research Methods How doyou do research in psychology? Psychology uses the scientific process: Question Hypothesis Data Interpretation Warm up with Crash Course’s Methods video
  • 3.
    Research Methods Hindsight biasis the tendency of people to overestimate their ability to have predicted an outcome that could not possibly have been predicted. However, the goal of science is to be predictive; that is, determine an outcome before it happens, not after.
  • 4.
    Research Methods Ways ofdoing research in psychology (research design): There are different ways of doing research: Experimental design (experiment) - Laboratory experiment (usually just “experiment”) - Quasi-experiments (like an experiment*) - Field experiments (like an experiment*) Naturalistic observation Correlational design (correlation) Survey Case Study
  • 5.
    Research Methods What isthe relationship between these variables: Coffee intake and running speed? Showering and body odor? Studying and grades? Mass of an object its gravitational force?
  • 6.
    Research Method Terms Hypothesis:Usually an ‘if, then’ statement or simply a prediction about some event. Theory: aims to explain a broad set of phenomena. Hypotheses (multiple hypothesis’) support a theory. Independent variable: The variable that influences the dependent variable. Dependent variable: A variable that depends on the independent variable.
  • 7.
  • 8.
    Operational Definitions Hypothesis: Ifpeople are given money, they experience greater happiness than if given candy. How do we measure happiness? Operational Definitions (IMPORTANT) Definitions of variables in research need to be quantifiable (countable) and observable. They need to be operationally defined. Example, happiness: how many times someone smiles.
  • 9.
    Validity & Reliability Researchaims to be valid and reliable Validity refers to whether the research measures what the researchers set out to measure. Reliability refers to whether the same results can be produced under similar conditions.
  • 10.
    Validity & ReliabilityExample If you have a scale, and it says you weigh 100lbs., but on every other scale you step on, the scale says 180lbs., the scale would not be a ‘valid’ measure of your weight. On the other hand, if you stepped on the scale and it said 100lbs., and then five minutes later stepped on the same scale again, and it said 130lbs., the scale would not be a ‘reliable’ measure of your weight.
  • 11.
    Research Method Terms Participantsare the people or subjects in your study. Sample/Sampling: A sample is a set of participants or things taken from a population. Sampling refers to the selection of participants. Population is the large set of individuals from which a sample was taken. Representative sample: A sample that accurately reflects the larger population.
  • 12.
    Research Method Terms Exampleof population and sample
  • 13.
    Research Method Terms Exampleof population and sample
  • 14.
    Understanding Terms Example Wearinga red shirt in a large crowd gets you more attention. What is my hypothesis? What must be operationally defined? Who is my sample? Who is my population? Is the sample representative?
  • 15.
    Understanding Terms Example Whatis a representative sample? A representative sample better reflects the participants of a population compared to an unrepresentative sample** Therefore, whatever research result you find, the results will be able to be generalized to the population.
  • 16.
  • 17.
    Random Selection Random selectionmeans that every member of the population has an equal chance of being selected. How could we randomly select in the previous example?
  • 18.
  • 19.
    More on RandomSelection Stratified sampling: A process that allows a researcher to ensure the sample represents the population on some criteria. If I want to research whether different racial groups respond differently to a survey, I could select 10 Caucasians, 10 Asians, 10 African Americans…. Stratified sampling takes place at the selection level. If you want your sample to meet some criterion or criteria, then that is stratified sampling*
  • 20.
    Experimental Method Psychologists preferexperiments because they can establish a cause-effect relationship. Laboratory experiments are conducted in a controlled environment and use random assignment. Field experiments are conducted in the ‘real’ world. Researchers go out and manipulate some variable and observe the effect. Example here. Quasi-experiments are experiments, but do not use random assignment.
  • 21.
    Experimental Research Confounding variables Aresearcher must try to isolate variables or control for “confounding” variables Confounding variable: A confounding variable is any difference between the experimental controls and the control conditions, except for the independent variable, that might affect the dependent variable.
  • 22.
    Confounding variables If Iam studying whether the amount of time spent studying is associated with better grades, what are variables that may influence this relationship other than studying?
  • 23.
    Groups An experiment musthave a control group and an experimental group. The experimental group is that which receives the independent variable. The control group is the group that does not receive the independent variable.
  • 24.
    Understanding Terms Hypothesis: Drinkingcoffee before running makes one run faster. After gathering a sample, who is my experimental group and who is my control group?
  • 25.
    Terms Assignment: Assignment isthe process by which participants are put into a group, experimental or control. Random assignment: The process of randomly assigning participants to either the control or experimental group.
  • 26.
  • 27.
    Terms What if inmy coffee experiment, I allow participants to join which ever group they want? Participant-relevant confounding variables: If participants were given the opportunity to choose which group to be in, the results might become biased.
  • 28.
    Terms Group-matching: If onewanted to match for sexual orientation, eye color, skin color, or something other variable, then group matching is the process of assigning individuals to groups based on some criteria. Situation-relevant confounding variables: When conducting an experiment, both groups (experimental and control) must be subject to the same environment.
  • 29.
    Terms Experimenter Bias: Thetendency for researchers to treat members of the experimental and control groups differently. If the aim of my study is confirm my hypothesis, I may interpret the results differently than someone who is not involved or does not have an interest.
  • 30.
    Controlling for Bias:Experimental Designs Single Blind: Participants do not know whether they’re in the experimental group or the control group. Double Blind: Neither the participants nor the research knows who is in which group.
  • 31.
    Placebo & PlaceboEffect Placebo is a substance that has no therapeutic effect; it is often used in control groups for testing new drugs. Placebo effect: Participants feel a “psychological” effect, but have not been given a real substance that would cause “real” physiological differences in their body.
  • 32.
  • 33.
  • 34.
    Correlational Method Correlation isa statistical measure that indicates the extent to which two or more variable fluctuate together. A correlation expresses a relationship between two variables without ascribing a cause. Correlation does not equal causation.
  • 35.
  • 36.
  • 37.
    Correlations A positive correlationbetween two things means that the presence of one thing predicts the presence of the other. A negative correlation means that the presence of one thing predicts the absences of the other.
  • 38.
    Correlations Predict the correlation Drinkingwater and feeling thirsty Studying and amount of free time Colour of t-shirt worn on exam day and exam grade Number of hours spent studying and grades Amount of coffee drunk and talkativeness Staying up all night and fatigue Number of fictional books read and driving ability
  • 39.
    Correlations Why use correlations,why not always conduct an experiment? Sometimes correlations are the best research method because you cannot manipulate the data in an experiment. Does weather affect shop lifting? Is there any way we can manipulate the weather?
  • 40.
    Surveys Conducting a surveyinvolves asking participants to complete a questionnaire. Using the survey method means that one cannot control for certain confounding variables. Survey method is subject to the social desirability effect: the tendency of survey respondents to answer questions in a manner that will be viewed favorably by others.
  • 41.
    Naturalistic Observation Naturalistic observation:Researchers go out and observe participants in their natural habitat without interfering at all. Naturalistic observation is different from field experiments because in naturalistic observation, researchers do not interfere.
  • 42.
    Case Studies The casestudy method is often used in clinical psychology. Case studies allow researchers to get a picture of a small group of people (as little as one), but this means that the findings cannot be generalized to any population. An example of a case studies may be studying a rare psychological disorder.
  • 43.
    APA Ethical Guidelines Ethicalconsiderations are a major part of research regarding human and animal research. Guidelines are established by the American Psychological Association (APA) for human and animal research.
  • 44.
    APA Ethical Guidelines(Animals) 1. Must have a clear scientific purpose. 2. Questions of research must be important. 3. Animals chosen must be suited for the question. 4. Animals must be cared for in a humane way. 5. Animals must be acquired legally. 6. The experimental procedures must be ones which employ the least amount of suffering possible.
  • 45.
    APA Ethical Guidelines (Humans) 1.No coercion: (participation must be voluntary). 2. Informed consent: (participants know and agree with what you’re doing). 3. Anonymity/confidentiality: participants privacy protected. 4. Risk: as little as possible (physical and psychological. 5. Debriefing: after experiment is done, you inform participants about the nature of the experiment.
  • 46.