Science provides a system that allows us to construct increasingly realistic models of the world around us.

1. What is Science?

• Early in history, people attempted to explain natural phenomenon by applying human characteristics to it (sky is angry, etc).
• Others tried explaining with spirits; later this turned to authorities like religious leaders giving explanations.
• Faith is belief without evidence, while science requires evidence.
• The word science comes from the Latin Scientia, meaning knowledge.
• Science is a knowledge gained through special way of learning through systematic observation and experimentation.

1-1. How do we develop a scientific mindset?

1. Science relies on objectivity rather than subjectivity, meaning that conclusions are based on facts rather than personal opinions.

• Study by Allport and Postman (1945) showed people switching race of a person threatening someone else from white to black when they were asked to recall the photo from their memories.
• We cannot recognize when we are being subjective, so scientists cannot rely on their introspections to maintain objectivity.
• Scientific methods help maintain objectivity.

2. Science uses systematic observation, while other observations use "hit or miss".

• "Hit or miss" means making conclusions based only on whatever is happening around us - our immediate acquantances.
• College students may make "hit or miss" observations that drinking is universal for the peers of their age.

3. Science relies on observable and repeatable evidence, while every observation ignores or rejects evidence.

• Confirmation bias is our tendency to remember instances that supported our beliefs more than those that contradicted our beliefs.

1-2. The Importance of Critical Thinking

• Critical thinking is our ability to think clearly, rationally, and independently.
• Critical thinking is the foundation of scientific reasoning.
• Critical thinking gives us skills to evaluate information and the confidence to question the people in authority instead of engaging in blind obedience.

Five Critical Thinking Questions:
1. What am I being asked to believe or accept?
2. What evidence supports this position?
3. Are there other ways that this evidence could be interpreted?
4. What other evidence do we need?
5. What are the most reasonable conclusions?

Signs of Not Thinking Critically:
1. I prefer being given the correct answers rather than figuring them out myself.
2. I don't like to think a lot about my decisions, as I like to rely on gut feelings.
3. I don't usually review the mistakes I made.
4. I don't like to be criticized.

Our vunerabilities that hinder our ciritical thinking:
1. We tend to think that the most information is true as it is time and effort-wise efficient.
2. The easier the info and the more freqeuntly we are being exposed to it, the easier it is for us to believe it true.
3. Statements with pictures have more "truthiness"
4. We use inferences to bridge gaps between ideas, which can sometimes be wrong.
5. Prior knowledge can interfere with new learning.
6. We often neglect important info. (ex) "bury survivors")

1-3. The Scientific Enterprise

Science is more than collection of facts - it is a process.

Scientific Theories

• Theories are set of facts and relationships between facts that can be used to predict phenomena.
• Scientific theories generate new predictions (comes from a Latin words for "Saying before")
  • Prediction is stated in a rigorous, mathematical form that under a certain set of circumstances, a certain set of outcomes is likely to occur.
• Before we attempt to generate our own scientific questions, we should be familiar with relevant theories and previous discoveries.

Science is a process of constant refinement of observations and making new theories based on previous observations.

Generating Good Hypotheses

• Hypothesis is an educated guess based on prior evidence and logical possibilities.
• A good hypothesis links concrete variables and makes specific predictions (ex. exposure of coffee and brown fat activity).
• Scientists can never "prove" that a hypothesis is correct because using new technologies in the future may show the hypthesis to be false.

Communicating Science

• Science is a vastly collaborative enterprise.
• Scientists depend on many others in the community to help them improve their work - one way is through peer review.
  • Peer review is a review where other scientists who are experts in the same area scrutinizes the research.
  • Peer review helps dividing the responsibility to many, and checking the accuracy of the content.
• Published results should under go replication
  • Other scientists should try to replicate the results of a study, which increases the accuracy of the data.

2. How do Psychologists Conduct Research?

• There are descriptive, correlational, and experimental methods of research.
• Each method provides different point of view and has strenghts and weaknesses.
• Many studies combine several of these methods.

2-1. Descriptive Methods

• Research methods designed for making careful, systematic, real-world observations with less bias.
• Descriptive methods include case studies, naturalistic observations, and surveys.

2-1-1. The Case Study

What is it?:

• Case study is an in-depth analysis of the behavior of one person or a small number of people.

When do we use it?:

• Psychologists use case study when large numbers of participants are not available or when a particular participant possesses unique characteristics.

Examples?:

• Examinations of Henry Molaison (1926-2008)'s brain is one example of case study.
  • He was left with memory deficits after his brain surgery.
  • Psychologists could learn about the brain structures and the formation of memories.
• Let's conduct a case study with an example of Michael Carneal and possible links between video games and physical aggression.
  • Gathering background info about the case by interviewing others associated with the case,
  • viewing legal and medical documents,
  • observe media accounts,
  • interview Carneal and administer personality and clinical tests.

Advantages?:

• The case study method is well suited to learning about unusual and rare situations - like mass shootings.
• A case study can contribute to science by obtaining detailed data and helping you better prepared to generate and test additional hypotheses.

Limitations?:

• We cannot make broad generalizations based on any single case.

2-1-2. Naturalistic Observation

What is it?:

• An in-depth study of a phenomenon in its natural setting.

When do we use it?:

• Need our results to be applied to more general population
• Need to observe individuals in less artificial setting.

Examples?:

• Jane Goodall's study of chimpanzees in their habitat is one example.
  • Her painstaking study showed that chimpanzees were actually not vegetarian, which was previously believed to be true.
  • Her study provided scientists with rich, accurate knowledge of the behavior of the chimpanzees.
• Video games and physical aggression could be again studied with naturalistic observation:
  • Could attend Local Area Network (LAN) parties.
  • Try observing people playing games and their subsequent aggressive behaviors.

Advantages?:

• Provides insights into natural, real-world behaviors with large numbers of participants.

Limitations?:

• Could raise ethical questions.
• Just like case studies, cannot be used to consolidate the relationship between variables because of confounding variables.
  • Taking LAN parties for example, we cannot be certain whether it's the video games or staying up all night that led to aggressive behaviors.

2-1-3. The Survey

What is it?:

• Descriptive method in which participants are asked the same questions.
• A good survey is that which require large sample that is representative of the population.

Example?:

• Video game example again:
  • Could ask questions in third person,
  • "People play violent games" rather than "I play violent games",
  • to minimize the conformity to social ideals.

Advantages?:

• Provides great deal of useful information quickly at little expense.

Limitations?:

• Survey relies on self-reporting: People may respond to answers that are socially appropriate, rather than giving honest opinions.

2-2. Correlational Methods

What is it?:

• Research methods that measure the direction and strength of the relationship between two variables.
  • measure: a method for describing a variable's quantity.
• Three possible outcomes: positive, negative, and zero correlations.

Advantages?:

• Correlations can help us learn about situations in which experiments would be unethical.
  • Cannot force pregnant women to drink alcohol,
  • instead they can write in diaries how much they drank.

Limitations?:

• Correlations do not tell us about causality:
  • Nicolas Cage appearing in file is positively correlated with people drowning in pool??? wowie
  • Two variables can influence each other:
    • Aggressive students can play video games.
  • Third variable may exist:
    • Perpetrators could have been bullied in the past.

2-3. Experimental Methods

What is it?:

• A research method that tests hypotheses and allows researchers to make conclusions about causality.
• Independent variable
  • The variable controlled by the experimenter
• Dependent variable
  • Measure used to assess the effects of manipulating the independent variables
• Control group
  • Group of experimentee that experience all procedures except exposure to independent variables.
• Experimental group
  • Group of experimentee that experience different independent variables.
• Random assignment
  • Randomly assigning each participant into any group to mitigate the effects of individual differences on the result of the experiment.
• Confounding variables
  • Variables that are irrelevant to the hypothesis being tested that can alter or distort our conclusions.
  • Could be situational: like traffic outside the building, noises, etc.
    

Example?:

• Confounding Variables:
  • People are more aggressive under hot temperatures.

Limitations?:

• Artificiality:
  • Participants know they are in an experiment, which may affect their behaviors.
  • Making experiments more realistic may raise ethical issues.
• Need to be aware of defining and implementing the variables in a concrete fashion (Need to carefully operationalize).
  • Operationalization: translating abstract independent and dependent variables into measurable forms.
    • 1. Identify the concept to be measured specifically.
    • 2. The quantitative measures of the variables must be determined.
  • There are too many ways to operationalize variables:
    • Few direct comparisons can be made among many studies.

2-3-1. Quasi Experiment

What is it?:

• "Quasi" means to resemble.
• Resembles true experiment, but without random assignment to groups.

Example?:

• Gender as an independent variable:
  • No random assignment to groups involved.
• Head Start programs affecting later school achievements?:
  • Children's families should have income below proverty threshold to join Head Start programs.
  • These children have different socio-economic status, which may make these children different from others.
  • May cloud the impact of attending the program.

Limitations?:

• Does not do as good a job as the true experiment at controlling individual differences.

2-3-2. Meta-Analyses

What is it?:

• A statistical analysis of many previous experiments on a single topic.

Advantages?:

• Often provides a clearer picture than do single experiments observed in isolation.

Limitations?:

• Might be subject to Publication Bias, making the results of any meta-analysis misleading:
  • The possibility that published studies are not representative of all work done on a particular phenomenon.
  • Only the studies with some significant results are published.
  • "File drawer" problem is a manifestation of publication bias:
    • studies with no significant results are "filed away" in drawers.

2-3-3. The Importance of Multiple Perspectives.

• Using the core perspectives in chapter 1 can provide different results.
• Converging evidence from descriptive, correlational, and experimental research provides us confidence in our conclusions.

2-4. How do we study the effects of time?

2-4-1. Cross-sectional study

What is it?:

• Data are obtained simultaneously from different age groups of people.

Advantages?:

• Quick and cost-effective
  • Data are collected at a single point of time, all at once.

Limitations?:

• Cohort effects:
  • Generational effects of having been born at a particular point in history.
  • People who was born with internet may react differently to games itself than those who was born without it.

2-4-2. Longitudinal Study

What is it?:

• data are obtained from the same individuals over a long period of time.

Advantages?:

• Few logical drawbacks.

Limitations?:

• Expensive and time consuming.
• Participants my drop out because they move or lack of incentive.
  • Those left must still comprise representative sample.

2-4-3. Mixed Longitudinal Design

What is it?:

• Combines the cross-sectional and longitudinal approaches by observing a cross-section of participants over a shorter period than longitudinal studies.

Advantages?:

• Faster and less expensive than longitudinal method.
• avoids some cohort effects caused by cross-sectional study.

3. How do we draw conclusions from Data?

3-1. The Importance of Valid and Reliable Measures

• Reliability and Validity

3-1-1. Reliability

• refers to the consistency of a measure.
• Test-retest, interrater, inter-method, internal consistency.

  • Test-retest: Retesting should not change the score so variably.

  • Interrator: different observers should have similar interpretation of the measure.

  • Inter-method: different methods of measuring a feature of an individual should have positive correlation.

    • SAT and highschool grades are positively correlated, making these measures more reliable.

  • Internal-consistency: measures within a single test are positively correlated with on another.

    • "I like to ride bicycles" and "I've enjoyed riding bicycles in the past", and disagreement with the statement "I hate bicycles"

    

3-1-2. Validity

• means that a measure leads to a correct conclusions and evaluates the concept it is designed to do.
• To check validity, use convergent validity
  • whether a measure correlates with other existing, established measures of the same concept.
  • What is the god damn difference between inter-method and convergent validity?????

3-2. Descriptive Statistics

• organizing data into meaningful patterns and summaries
• Most likely Frequency Distribution

3-2-1. Central Tendency

• Central Tendency means identifying "average" scores of our data sets.
• 3 types of central tendency: Mean, Median, and Mode.
• Mean:
  • Numerical average of a set of scores.
• Median:
  • Halfway mark in the data set.
  • Meaningful in the presence of outliers.
• Mode:
  • Score that occurs most frequently
  • Meaningful in cases like bimodal distributions.

3-2-2. Variance

• Variance indicates how clustered our scores are.
• Standard Deviation measures how tightly clustered a group of scores is from the mean.
  

3-2-3. Normal Curve

• Symmetrical: Half of data above mean, half below.
• Most scores occur around the mean:
  • +-1 standard deviation: 68% of scores.
  • +-2 std: 95%
  • +-3 std: 99%
    

3-2-4. Descriptive Statistics with Two Variables

• We can use a scatterplot to analyse the relationship between two variables or more.
• Correlation Coefficient can be used to descirbe the strength of relationship.
  • +1 means the variables have strong positive correlation
  • 0 means no systematic correlation
  • -1 means strong negative correlation.

3-3. Inferential Statistics

• Statistical methods that allow experimenters to extends conclusions from samples to larger populations.
• We must be cautious when making generalizations (making conclusions outside our research sample).
• Conclusions could be rejecting a null hypothesis or failing to reject null hypothesis.

3-3-1. Null Hypothesis & Significance testing

• Null Hypothesis means that there is no relationship between variables.
• We fail to reject null hypothesis if p >= 0.005, or we have more than .5% of the occurrences of tests that contribute to concluding that there is no relationship; in other words if there is more than or equal to .5% occurrences of null hypothesis being true.
• When p <= 0.005, we can say that the observed result is statistically significant.
• There is a limitiation:
  • Significance testing does not show the effect size.
  • It only determines if there is a relationship, but does not determine how much one influences the other.
    • Facebook's experiment showed that news feed does impact a user's status, meaning that the result was statistically significant,
    • but the impact was trivial; changes were not so big.
    • it had tiny effect size.

4. How can we conduct ethical research?

• Research institutions that receive federal fundings must seek approval from Institutional Review Boards (IRBs) or Institutional Animal Care and Use Committees (IACUCs).
• Private corporations that do not receive federal fundings are not being regulated - efforts are being made to bring them into compliance with federal guidelines as well.

4-1. Human Participants.

• Human participants must be voluntary.
• Ethics is in higher priority than our ability to generalize research results:
  • It is acknowledged that voluntary participants are different from others, so it can be hard to generalize the results to other members of the population.
• Here are several things to consider to keep the experiment ethical.

Incentives

• Incentives must not be so extreme that they become the primary motivation for prospective volunteers.

Informed Consent

• Researchers must provide an Informed Consent form which explains what the experiment will entail.
  • Facebook experiment did not solicit informed consent.

Qualified Guardian

• Many psychologists' interests are with people who are unable to sign their consents to join the experiment, like a person with alzheimer's disease.
• If so, experiments should seek qualified guardian's permission.

No irreversible harm

• Quite obvious.

Deceptions are told to the participants

• Any existence of deception for the experiment should be written in the informed consent, although it does not have to be too specific.

Privacy and Confidentiality

• Researchers should seek parcitipants' agreements to use their private data, like medical records.
• Parcipants have their rights to keep their data confidential - revealing their information only under their permissions.
• Collecting data anonymously, so that researchers do not know the participant's identity, is the best way to keep it both private and confidential.

4-2. Animal Subjects

• 3Rs in the UK: Reduce (the number of animals used), Replace (do not use animals if experiment can be done other ways), and Refine (adjust methods to be humane)
• Research using animals must demonstrate clear purposes, like benefitting animal and human healths.
• Animals must be provided with exellent housing, foods, and veterinary.