SOPC - Lecture Notes

An umbrella term for the scientific examination of how people function (how we behave, think, and feel)

1. Describe human phenomena

2. Predict human phenomena

3. Determining the causes of phenomena

4. Explain/understand human phenomena

5. Solve applied problems –derived from 1-4

Rough Distinctions-

Basic research: designed to understand psychological processes without regard for whether it will be immediately applicable in solving real-world problems.

Applied research: designed to investiage real-world problems or improve the quality of life.

Tells us how things are.

Describes in an accurate and systematic fashion the behavior, thoughts, or feelings of a group of participants.

Tells us how things are in relation to one another

Examines the nature of the relationship between two measured variables.

• •Election days (Redelmeier & Tibshirani, 2008)
• •Super Bowl Sundays (Redelmeier & Stewart, 2003)
• •Breakfast and losing your virginity

•Gets at the causes of phenomena

Tests whether certian variables cause changes in behavior, thoughts, or feelings.

• •Independent variable (IV) and dependent variable (DV)
• •Example: Loftus and Palmer (1974)

“smashed into” 40.8 mph

“collided with” 39.3 mph

“bumped” 38.1 mph

“hit” 34.0 mph

“made contact with”31.8 mph

•Examine the effects of variables on people BUT the IV cannot be varied


or controlled (only naturally occurring)

Researcher can't assign participants to conditions or manipulate the Independent Variable.

•Useful when “true”experiment


is either impossible or unethical

How do we know things? Properties of scientific inquiry.

• Four ways:
• •Intuition
• •Logic
• •Authority
• •Observation

• Religion:
• 1. authority
• 2. intuition
• 3. logic
• 4. observation

• Government:
• 1. authority
• 2. intuition
• 3. logic
• 4. observation

• Science:
• 1. observation
• 2. logic
• 3. intuition
• 4. authority

Relying on observations or direct experience to draw conclusions about the world.

Example - Galleleo: viewing the sky over time came to the conclusion that we lived in a sun centered universe.

1. Systematic empiricism = relying on careful, organized observations to answer specific questions.

2. Public verification = public scrutiny, peer review

3. Testability

•Benjamin Rush and bloodletting

•The theory of knocking rhythms = predictions of who's knocking - needs to be specific & concret.

• •An Aquarius horoscope (Sept 12, 2005) - horoscopes very veague.
• “Working together with a certain colleague seems like a recipe for disaster, but if you restrict yourself to tackling the task at hand, you might just get more done than you'd originally hoped for. Keep that nose to the grindstone.”

“There are seven windows in the head: two nostrils, two ears, two eyes and a mouth; so in the heavens there are two favorable stars, two unpropitious, two luminaries, and Mercury…From which and many other similar phenomena of nature such as the seven metals, etc., which it were tedious to enumerate, we gather that the number of planets is necessarily seven....Besides, the Jews and other ancient nations, as well as modern Europeans, have adopted the division of the week into seven days, and have named them from the seven planets. Now if we increase the number of planets, this whole system falls to the ground....Moreover, the satellites are invisible to the naked eye and therefore can have no influence on the earth and therefore would be useless and therefore do not exist.”

•A set of propositions that attempt to explain a set of empirical observations

–How and why concepts are related

•A specific proposition derived from a theory

• Theory -->
• Hypotheses -->
• Predictions (hypotheses applied to specific research settings) -->
• Data Collection -->
• Hypotheses

• •Specifying how a concept is measured or manipulated in a particular setting
• –Sports: basketball, rules of the game
• –Physics
• –Psychology
• •Biases in how concepts are defined
• •Coren(1994)

Most Intelligent:

• 1.Border Collie
• 2.Poodle
• 3.German Shepherd
• 4.Golden Retriever
• 5.Doberman Pinscher
• 6.StetlandSheepdog
• 7.Labrador Retriever

Least Intelligent:

• 1.Greyhound
• 2.Boxer
• 3.Foxhound
• 4.Great Dane
• 5.Chihuahua
• 6.Beagle
• 7.Bloodhound

• •What case studies tell us
• –Therapies
• –Testimonials - comfort whipe
• (Case studies - describe things, what do we focus on from here? Very weak!)

The small hospital, because the sample size is smaller.

• •People are generally not good at this
• •Individual cases influence us strongly
• •“Person who” statistics
• •In science, case studies are only useful for exploring possibilities

• •We’re overly influenced by vivid memories and events
• –Airport security measures (recent bombing)
• –Worrying about negative events (MacDonald, 1990) (Parents worrying more about their kid being kidnapped than being in a car accident)
• –Purchase decisions
• –Sincere testimonials

• (ex. astrology, graphology)
• Methods not valid-claims it is based on science, but not.
• •Testimonials are typically the only evidence for the validity of claims
• •Proportion of Americans who believe in astrology:
• 31%(Harris Poll, 2003)
• 25%(Gallup, 2005)
• 25%(Pew, 2009)

• •Supporting evidence is heavily anecdotal.
• •Hypotheses generated are typically not testable.
• •Genuine scientific references are not cited.
• •Claims are stated in scientific-sounding terminology.
• •Claims ignore conflicting evidence.
• •Claims are never revised (& REAL science is all about revising claims.)

• •The burden of proof for any claim rests on the person trying to prove it (claimant.)
• •Extraordinary claims demand extraordinary evidence.

• •In US, parents of a girl are nearly 5% more likely to divorce than parents of a boy
• •Parents of 3 girls are almost 10% more likely to divorce than parents of 3 boys
• •Do American parents prefer boys or girls?
• - Divorced women w/ girls are less likely to marry than divorced women w/ boys.
• - Parents of girls are more likely to try for another child than parents of boys.
• - "Shotgun Marriages:" unmarried couples are more likely to marry if child is a boy.

1)Parents of daughters are more likely to divorce than parents of sons

2)In multi-child families, parents of daughters are more likely to try for another child than parents of sons

3)Divorced mothers of daughters are less likely to remarry than divorced mothers of sons.

Descriptive statistics = summarize people’s behaviors, thoughts, or feelings in a study

• Measures of central tendency
• •Mean: arithmetic average of all scores; communicates what you can expect from typical score
• •Median: the middle score
• •Mode: the most frequent score

• how the scores are distributed.
• •Range
• •Variance: tells us how much scores differ from their mean; symbolized by s2
• •Standard deviation: the square root of the variance; symbolized by s (or SDin APA style)

• “Frequencies” information for demographic variables
• --EX: # of males, # of females

• Descriptive statistics on scales
• --mean and SD for each variable
• --range of scores for each variable

• Student’s t-test
• –Compares two groups (e.g., M vs. F)

• Two possible decisions (based on p-values):
• Significance: when p < .05
• Nonsignificance: when p > .05

A t-test revealed a significant difference between these two groups, t(44) = 6.88, p< .001.** Always describe means after reporting significance

Correlation coefficient (r) = measures strength of association between two continuous variables.

• Possible range:
• -1.0 (Perfect inverse relationship) -> 0 (no linear relationship) -> 1.0 (Perfect direct relationship)

The correlation between these two variables was significant, r(98) = .48, p= .03.

• (1) Behavioral measures
• (2) Physiological measures
• (3) Self-report measures
• (4) Archival measures

Reliability= consistency, stability, or dependability of a measure

Validity= is the measure is an accurate representation of the construct we want to measure?

Observed score = true score + measurement error

• True score...actual score an individual should receive on a measure.
• Measurement error...result of any factor that makes observed score different from true score.

• (1) Test-retest reliability...consistency over time
• -Compare scores from one testing to another (usually 1-12 week delay)
• -General rule of thumb: correlation > .70
• -Alternate-form reliability

• (3) Interrater reliability...consistency across raters/judges-Do different observers agree on ratings?
• (2) Internal consistency...consistency across items
• -Do all the items “hang together”?
• -Item-total correlation = corr. between each item and the sum of all other items; should be > .30
• -Split-half reliability = corr. between two halves of scale; should be > .70
• -Cronbach’s alpha coefficient = equivalent to average of all possible split-half reliabilities; should be > .70
• (3) Interrater reliability...consistency across raters/judges
• -Do different observers agree on ratings?
• Correlation between two sets of ratings should be > .70

• Is the measure a true representation of the construct we want to measure?
• We need to specify our intended construct:
• - Is the SAT a valid measure?
• - Is the SAT a valid measure of future academic performance?

• Can we have reliability without validity?
• Can we have validity without reliability?

• (1) Face validity...Does measure look right?
• –Items should look like they measure intended measure.

(2) Content validity...Does measure adequately represent relevant content?

• (3) Construct validity...Does measure relate to other constructs as it should?
• •Two types of construct validity
• –Convergent validity
• –Discriminant validity

(4) Criterion-related validity...Does measure predict meaningful differences in relevant behaviors?

• Two types of criterion-related validity:
• –Concurrent validity: Does measure accurately relate to relevant current behavior?
• –Predictive validity: Does measure accurately predict relevant future behavior?

Which constructs should correlate with Need to Belong? Which should not correlate?

• Life values:
• True friendship - .36
• Mature love - .29
• Equality - .20
• Social recognition - .27
• Self-respect - .7
• Wisdom - .02
• Sense of accomplishment - .07

We need to show a measure is distinct from other established measures.

(1) Nominal scale: data values are just labels or categories

(2) Ordinal scale: involves the rank ordering of a set of behaviors or characteristics

(3) Interval scale: precisely-spaced order of measurement

(4) Ratio scale: interval scale plus a true zero

• Social desirability bias:
• *Sensitive topics
• -Finances
• -Religion
• -Voting
• -Sexual habits
• -Health-related questions
• -Drug use
• -Racial attitudes

• Social desirability bias
• -Interviews vs. self-administered responses
• -Bogus pipeline
• -Pledges and closing questions

• Knowledge problem:
• Unfamiliar w/ issue; no strong opinion

• Do you favor or oppose providing taxpayer funded loans to help keep General Motors, Ford, and Chrysler in business? (Rasmussen)
• 28% favor, 53% oppose, 19% not sure

• Do you think it is the right thing or the wrong thing for the government to spend billions of dollars In loans to General Motors, Ford and Chrysler to keep them in business. (Pew)
• 39% right thing, 54% wrong thing, 7% don't know

• As you may know, the three U.S. automakers --General Motors, Ford and Chrysler --appeared before Congress to say that their companies are on the verge of bankruptcy and to ask for taxpayer-funded loans to help them survive. Do you think the government should or should not rescue the three U.S. automakers? (LA Times/Bloomberg)
• 57% should rescue, 42% do not rescue, 11% don't know

• The major U.S. auto companies have asked the government for a program that would provide them with several billion dollars in assistance. The auto companies say they may go into bankruptcy without that assistance. Based on what you have read or heard, do you favor or oppose this program? (CNN/Opinion Res. Corp.)
• 36% favor, 61% oppose, 2% unsure

• Do you approve or disapprove of the federal government providing money to the big three U.S. automakers if it would help prevent them from going out of business or declaring bankruptcy? (CBS News)
• 45% approve, 44% disapprove, 11% don't know

• Decisions to make:
• What question format to use?
• -Open-ended questions - more options
• -Closed-ended questions - fewer options
• *Major difference between open/closed ended questions.

• Decisions when creating questions:
• (1) Keep it short and simple.
• (2) Use precise terminology.
• (3) Avoid unnecessary negatives.
• (4) Avoid questions that don’t produce variance.
• (5) Be aware of question order. - G. Bush satisfaction
• (6) Ask only 1 question at a time.
• (7) Avoid leading questions.
• (8) Avoid false choices.
• (9) Pretest your questions.

• Similarities to Self-Report
• -Need to establish reliability and validity
• -Participant may distort behavior
• Differences from Self-Report
• -Can gather data from people of any age (e.g., infants), or other species
• -Can capture properties of the dyad or group
• -Greater generalizability; no memory biases

• Decisions to make
• -Where will observation occur?
• *Naturalistic observation
• *Contrived observation
• Example:
• - Testing whether certain visual skills are innate
• - Depth perception (Eleanor Gibson) - Baby, Visual Cliff!

• Decisions to make when designing studies
• - Will participants know they are being observed?
• *Undisguised observation
• --Problem:REACTIVITY
• *Should try to use unobtrusive measures
• *Disguised observation
• --Middlemistet al. (1976)
• --Dolinski(2000)

• Decisions to make:
• - How will participant’s behavior be recorded?
• *Checklists (tally sheet)
• *Temporal measures (duration and latency)
• *Rating scales

Are observers biased?

• Decisions to make:
• How will interraterreliability be maximized?
• -Coding system must be specific and precise.
• -Raters should be trained and allowed to practice.
• -Pilot test coding system with real data.
• -Check interrateragreement periodically.

• -Self-perception theory
• -The foot-in-the-door effect
• *Compliance increases after doing someone a favor
• -What about after trying to do someone a favor, but failing?
• *Asking directions to a fictitious street (“ZubrzyckiegoStreet”)

2nd request: agree to watch a bag...

• People who failed to give directions - 58% agreed to request
• Control (2nd request only) - 34% agreed to request

2nd request: agree to watch a bag...

• People who were shown an illegible address - 54% agreed to request
• People who were shown a nonexistent address - 62%
• Control (2nd request only) - 28% agreed to request

2nd request: agree to watch bike...

• People who gave directions (foot-in-the-door) - 32% agreed to request
• People who failed to give (illegible address) - 58%
• Control (2nd request only) - 36% agreed to request

2nd request: agree to watch bike...

• People who gave directions (easy FITD) - 38% agreed to request
• People who drew map (hard FITD) - 54% agreed to request
• People who failed to give (illegible address) - 54%
• Control (2nd request only) - 36% agreed to request

Analyzing data from records left behind.

• Advantages:
• -Inconspicuous; no reactivity
• -Can study unobservable phenomena
• -Cost/time effective