Psychology 242, Introduction to Research Methods

Back NextHomePage

1Psychology 242Introductionto Research

Psychology 242, Dr. McKirnan Weeks 1 & 2; Introduction to scienceWeeks 1 & 2; Introduction to science.

Psychology 242, Dr. McKirnan

Introduction to research in psychology.

1/5/09Right click for “full Screen” or “end

show”. Left click to proceed,

The Class web site is: http://www.uic.edu/classes/psych/psych242

Look in the "Announcements" window on the home page for Web updates or class announcements.

Please e-mail me ASAP if you find a broken or incorrect link!

David McKirnan

Add survey measure for operationalization of depression.Try to add other examples.

Back NextHomePage

2Psychology 242Introductionto Research We will cover some basic topics in science…

What is science?

How can research answer basic questions about behavior?

How do we ensure our research is ethical?

How do we gather data that is reliable and valid?

Who do we study, and why?

How do we answer research questions when a true experiment is not possible?

How do we use statistics to evaluate our results?

How much of what we “know” is actually superstition?

Rational?

How does science differ from…common sense?religious belief?intuition?

How do we state an issue as a research question?

What is a “theory”?A “hypothesis”?An “operational definition”?

What are the basic features of a “true” experiment?

How do experiments differ from simple measurement?

How do we decide who we should be studying?

How do we make sure our sample represents the population?

How do we conduct “experiments in nature”?

How do we use surveys to test hypotheses?

How do statistical approaches affect the way we see the world?

How do we use statistics to test if our results are “significant” or meaningful?

Back NextHomePage

3Psychology 242Introductionto Research The course…

The course is based on the web site. All lecture notes are there All your readings & discussion group assignments Complete term paper instructions

The core course content is given in lectures.

Exams are (almost) completely lecture – based. You get course points via “i>Clicker” devices in

class

The term paper is central to the course.

Working on the paper will help you “get” the rest of the course materials (really!).

Begin early!~!


Back NextHomePage

4Psychology 242Introductionto Research Grades..

Grading: 3 exams, each worth 15% of your grade Lecture attendance, 10% Discussion group attendance & assignments, 15% Research paper, 30%

Exams take content from the lectures

Discussion groups require both attendance and assignment completion

Talk to your TA and begin the paper early!!


An i>clicker remote is required for this course.

You can purchase a new or used clicker through the bookstore, from other students, or possibly used on-line.

How will we use the clicker?

I pose questions on the screen during lecture.

You answer using your i>clicker remote.

Class results are tallied.

I display a graph with the class results on the screen.

We discuss the questions and answers.

You get points both for lecture participation and for answering questions correctly.

How do you vote?

Turn on the clicker by pressing the bottom “On/Off” button.

A blue “Power” light will appear at the top of the remote.

How do you vote?

When I ask a question in class (and start the timer), select A, B, C, D, or E as your vote.

I may also ask you to talk about your answer in class.

How do you know your vote was received?

Check your “Vote Status” Light: Green light = your vote was sent

AND received. Red flashing light = you need to

vote again.

Not sure you saw the light?

Just vote again.

Want to change your vote?

You can vote again as long as the timer is still going.

Registering your i>clicker

Until you register your i>clicker, your responses are tied to your clicker remote ID (located on the back of your clicker), rather than to you.

When you do register, your previously recorded voting responses will be assigned to you.

Registering your i>clicker online

REGISTER AT www.iclicker.com

1. Go to www.iclicker.com.

2. Click “REGISTER.”3. Enter these 4 details and

click “submit.”

IMPORTANT!! You MUST enter your student ID in the STUDENT ID field to ensure proper crediting.

Other tips If you bought a used clicker, replace the AAA batteries (all

of them). Do not use Duracell (they are a bit short for the casing). Do not use rechargeable batteries (they harm the clicker).

Register your clicker by the second week of classes. Before using a new clicker for the first time, pull the

plastic tab out of the battery compartment. Bring your clicker to class every day! Make sure your remote is on when voting! Do not submerge your clicker in liquid (and avoid liquid

near the clicker); like most electronics, liquid + your clicker is a bad combination.

Check out www.iclicker.com for FAQs. Contact [email protected] for help.

Back NextHomePage



The big picture: Introductory lectures

How do we “know” something?

Science, anti-science, pseudoscience.

Where does our knowledge of the world come from?

What does science do?

The core features of a research study.

Overall Research approaches.

Authority Intuition Simple Empiricism Rationalism

Describe the world Predict behavior Test theories Test applications

of theoriesObservational or

Measurement research

Experimental studies

Simple Descriptive

CorrelationalQuasi-

ExperimentTrue

Experiment

Back NextHomePage


Introduction to science, 1


How do we know something?






Back NextHomePage



How do we know things? Gay marriage would destroy the institution of marriage. The movie will begin at 9:00.

Iran is a clear and present danger to the U.S. About 3,000 people were killed on 9/11/01.

Each of us has an intrinsic purpose that we must discover.

The earth is about 3.5 billion years old.

We will make a key distinction between beliefs and empirical statements

Beliefs may be true and important, but cannot be empirically tested or demonstrated.

Back NextHomePage



How do we know things?

What research could you do on this statement?

Gay marriage would destroy the institution of marriage. The movie will begin at 9:00.

Iran is a clear and present danger to the U.S. About 3,000 people were killed on 9/11/01.

Each of us has an intrinsic purpose that we must discover.

The earth is about 3.5 billion years old.

Back NextHomePage



All ideas have some merit and should be considered equally.

Let’s answer some knowledge attitude questions… using your i>clickers

Most any idea is worthy of study. Scientific acceptance of ideas is not

egalitarian; ideas that are coherent and have empirical support are better.

A = TrueB = I’m not sureC = False

Back NextHomePage

18Psychology 242Introductionto Research Knowledge attitudes, 2

If a lot of people believe something there is probably something to that.


Popular opinion per se is an important area of study.

Scientific is not democratic; the data “win”, not the majority of believers

Many foolish or dangerous ideas were accepted by many people, even scientists, until countered by data.


Back NextHomePage


I can just sense when something is true or false.


Intuition is an important source of hypotheses or theories

Intuition describes your emotions, not necessarily the read world.

Emotionality & subjectivity are unscientific until they are empirically tested.


Back NextHomePage


Everyone is biased, even scientists, so why shouldn’t I just believe what makes sense to me?



Preconception can be useful when based on previous research

Science is designed to not be person based – Science is about methods, not people and their bias’ Scientific method specifically works to lessen personal

bias.

Back NextHomePage



Some ideas are “better” than others. Science is based on methods and evidence, not

people. Evidence from the natural world > personal

biases or beliefs.

Logic or rational thought > emotions, fear.

Science represents some core values about knowledge

Is it logically coherent? Is it supported by evidence?

Importance of “real world” tests of ideas or plans.

Refusal to “Cherry pick” confirmatory or self-serving evidence.

Back NextHomePage










Back NextHomePage


Are we rational?

Is American society “rational”?

Are our beliefs generally scientific?


Back NextHomePage

24Psychology 242Introductionto Research Beliefs…


How much do you believe in…

ESP or Extrasensory Perception

A = I believe in this

B = I am not sure

C = I do not believe in this

Back NextHomePage

25Psychology 242Introductionto Research Beliefs, 2…


How much do you believe…

That houses can be haunted

A = I believe in this

B = I am not sure

C = I do not believe in this

Back NextHomePage



Have you ever… Been protected

from harm by an angel?

A = Yes

B = I am not sure

C = No

Back NextHomePage



Have you ever… Had a dream that

later became true?

A = Yes

B = I am not sure

C = No

Back NextHomePage


Science on the run

Science & rationality are on the run in the 21st Century


Back NextHomePage



Science & rationality on the run

Beliefs in “Para-normal” phenomena

Gallup poll, June 2005, on paranormal beliefs. N = 1,002 adults 18 and older, sampling error = ±3% @ 95% CI

% Believe

% Not sure

% Do not

believe

Psychic healing / power of the human mind to heal the body

People … are sometimes possessed by the devil

ESP or Extrasensory Perception

That houses can be haunted

Ghosts/spirits of the dead come back in certain … situations

Telepathy / communication … without traditional senses

Clairvoyance / power … to know the past / predict the future

Astrology / position of stars & planets affect people's lives

Extra-terrestrial beings have visited earth…

People can communicate mentally with someone who has died

Witches

Reincarnation … rebirth of the soul in a new body after death

Channeling / allowing 'spirit-being' to temporarily control body

55

42

41

37

32

31

26

25

24

21

21

20

9

17

13

25

16

19

27

24

19

24

23

12

20

20

26

44

32

46

48

42

50

55

51

55

66

59

70

Back NextHomePage



Science & rationality on the run, cont.

# para-normal beliefs people

endorse

% endorsing

Cumulative %

All 10 of 10

Any 9 of 10

Any 8

7

6

5

4

3

2

Any 1

None

1%

2

3

3

6

7

10

11

14

16

27

1%

3

6

9

15

22

32

43

57

73

100

Only 27% do not endorse any para-normal belief

73% endorse at least one irrational belief

57% endorse at least 2 beliefs

Beliefs such as haunted houses or demonic possession are common (~ 40%).

Our anti-empirical society:

27%27%

57%57%73%73%

Back NextHomePage


Direct paranormal experiences among Americans

55% of Americans : "I was protected from harm

by a guardian angel.“

Psychology 242, Dr. McKirnan Weeks 1 & 2; Introduction to scienceWeeks 1 & 2; Introduction to science.Baylor University nationally representative survey of 1,721 respondents

Paranormal Experiences in the United StatesPercent that report the following experiences:

Used acupuncture or other forms of alternative medicine 28%

Consulted a horoscope to get an idea about the course of your life 28%

Called or consulted a medium, fortune teller or psychic 13%

Visited or lived in a house or place believed to be haunted 22%

Consulted a Ouija board to contact a deceased person or spirit 8%

Had a dream that later came true 43%

Witnessed an object in the sky that you could not identify (UFO) 17%

Used acupuncture or other forms of alternative medicine 28%

Consulted a horoscope to get an idea about the course of your life 28%

Called or consulted a medium, fortune teller or psychic 13%

Visited or lived in a house or place believed to be haunted 22%

Consulted a Ouija board to contact a deceased person or spirit 8%

Had a dream that later came true 43%

Witnessed an object in the sky that you could not identify (UFO) 17%

Back NextHomePage


The U.S. is a deeply religious country


89% consider themselves affiliated with a religion

Of “unaffiliated” 63% believe in God (to a greater or lesser extent).

Only 4% have no affiliation and do not believe in God

Back NextHomePage



The role of religion in Anti-science attitudes

2005 Gallup poll on evolution / origins of life

53%

12%

31%

0%

10%

20%

30%

40%

50%

60%

% endorsing each belief

God created humans as the Bibledescribes it.

Evolution with no divineintervention

Evolution, with God guiding theprocess (ID)

Evolution has become a core “wedge issue” in U.S. religious & cultural politics.

Gallup data: Few endorse a wholly

scientific view of the origins of species

“Intelligent Design” version of creationism relatively common

Biblical creation view most common.

Back NextHomePage



Anti-science attitudes: the evolution debate, cont.

Americans are the least accepting of evolution in Western countries…

Key predictors of rejecting evolution: “Fundamentalist” religious

identity Politically conservative /

Republican ideology Less scientifically literate Anti-choice & stem cell

research Older, male

only Turkey scores lower.

Back NextHomePage


Studying beliefs about science

Beliefs about science can themselves be scientifically studied

Following is an example of a complex correlational study of attitudes toward evolution.


Back NextHomePage



Predictors of pro-evolution attitudes: U.S. v. Europe

In both the U.S. and Europe education strongly predicts genetic literacy,

which itself underlies acceptance of evolution.

In both areas religiosity and pro-life attitudes lead to greater political conservatism…

And to less acceptance of evolution.

Back NextHomePage



Where do ideas about science come from?Education versus politics

Religiosity and “pro-life” attitudes have much stronger influence on political ideology and attitudes toward evolution

in the U.S.

…and political conservatism predicts rejection of evolution

in the U.S.,

but not in Europe.

than in Europe.

Back NextHomePage

38Psychology 242Introductionto Research Bottom line

Conservative & religious politics are generally not supportive of empirical or scientific thought


A scientific (rational, empirical) perspective: combines rational thought with empirical

evidence is not just a “research method”, but is a

larger approach to knowledge.

Social class and health Stem cell research Abstinence-only sex Ed. Teaching of Evolution Project DARE drug prevention Global warming

Science education is increasingly crucial.

Back NextHomePage










Woman with book, Pablo Picasso.

Back NextHomePage


Intuition: “I believe my Gut feelings”

Emotionality or a “hunch”


Four basic sources of knowledge or information:

Authority: “I believe what they tell me to”

Credible / powerful peopleImportant social institutionsSimple tradition

Back NextHomePage



Authority-based belief

Advantages: Stable core of beliefs or principles.

Move a field beyond the data; Visionaries, revolutionaries

Disadvantages:

Insensitive to proof or evidence

Highly susceptible to political bias

Can require evidence / science be corrupted, distorted or ignored.

Ignore or circumvent normal scientific procedures (e.g., Intelligent Design content in biology instruction).

Back NextHomePage



Intuition, emotion, superstition

Advantages:

Emotional or personal insight

Origin of novel hypotheses or theories

Move a field beyond the data

Disadvantages:

Insensitive to proof or evidence

Wishful thinking? ; often explicitly non-empirical

Emotion (e.g., fear) > rationality or evidence

Back NextHomePage



Intuition and Magical thought

Our brains may be “hard wired” for intuitive, “Magical Thought”

Back NextHomePage



Intuition, Magical Thought & science

The brain has evolved to make snap judgments about causation:

We leap to conclusions before logic can be applied.

Psychological biases can lead to distorted beliefs.

Our tendency toward for “magical thought” makes make scientific arguments a harder sell.

Emotional information processing is faster than logical or verbal processing

Simple perceptual biases precede any processing…

Co-occurrence / correlation causality

Emotional need for personal control personal causation.

Taking a rational, empirical approach can require that we suppress or reject our intuitive sense of causation

Back NextHomePage



Empiricism or simple exposure

Advantages:

Grounds knowledge in external, real world.

Confirm intuition by direct observation

Makes knowledge public (e.g., Copernican revolution)

Disadvantages / limitations:

Simple illusions / misperceptions / measurement error Confirmatory bias Oversensitive to emotional / perceptual salience Spurious correlations Anti-science use of naïve empiricism

Back NextHomePage



Limits of empiricism: optical illusions

Akiyoshi KITAOKA, Professor, Department of Psychology, Ritsumeikan University, Kyoto, Japan

http://www.ritsumei.ac.jp/~akitaoka/index-e.html

Back NextHomePage



Empiricism and the confirmatory bias

Cops and doughnuts

Yes No

Yes

No

Police officer present?

Doughnut present?

= memorability

subjective co-occurrence matrix.

Back NextHomePage



Emotional / perceptual salience

Certain experiences are more cognitively available.

Observation is never neutral or “objective”

Their perceptual salience Our emotional needs

Which is more hazardous: airline travel or automobile travel? Which is more common in women: breast cancer or cardio-

vascular disease? Conspiracy theories: could Kennedy have been killed by a single

person? (law of effect) Fear based attitude change: Saddam is an evil Arab Arabs

attacked us on 9/11 Saddam must be responsible for 9/11.

Which is more hazardous: airline travel or automobile travel? Which is more common in women: breast cancer or cardio-

vascular disease? Conspiracy theories: could Kennedy have been killed by a single

person? (law of effect) Fear based attitude change: Saddam is an evil Arab Arabs

attacked us on 9/11 Saddam must be responsible for 9/11.

Events / stimuli we pay attention to & remember are influenced by:

We only notice certain things We only “encode” certain things We only recall certain things in a given situation

Back NextHomePage



empiricism and spurious (naïve empirical) correlations:

The Japanese eat very little fat and suffer fewer heart attacks than the British or Americans.

The French eat a lot of fat and also suffer fewer heart attacks than the British or Americans.

The Japanese drink very little red wine and suffer fewer heart attacks than the British or Americans.

The Italians drink lots of red wine and suffer fewer heart attacks than the British or Americans.

Conclusion: Eat & drink what you like. It's speaking English that kills you.

Back NextHomePage



Naïve empiricism

Naïve Empiricismcan reflect anti-scientific bias

(e.g., creationism arguments; see Schafley editorial.) Science asks “why?”, not simply “what?”

Scientific data are often not directly observable

Empirical evidence can involve a large time scale

Microscopic / atomic level

Indirect evidence via instrumentation or accretion

Theory & hypothesis testing > simple data

Paleontology, geology, astrophysics

Developmental Psychology

(I won’t believe it unless I can directly see it myself)

Back NextHomePage



Advantages:

Develop larger, coherent principles or theories.

Accept conclusions that correspond to other knowledge

Disadvantages: Correspondence to empirical world?

Susceptible to ideological bias or authority – based belief systems

Rationalism

Back NextHomePage



Bottom line: Ways of knowing and science

Authority / authoritarianism Provides stable, core principles or beliefs Limits empirical evidence or alternative views

Intuition / subjective “hunch” Important source of novel hypotheses / theories / scientific

approaches Emotion-based “wishful thinking” or “magical thought” can

make us irrational or ignore / distort empirical facts.

Empiricism Grounds knowledge in “real” world, provides important

hypothesis-testing perspective Our perceptions are subject to cognitive / emotional biases.

Rationalism / theory Central purpose of science: coherent explanation of “why” or

“how” nature works. When subject to political pressure can limit hypothesis testing

or lessen respect for empirical evidence.

Back NextHomePage



How do we “Know” something?

Science: Integration of..Rationalism

TheoryHypothesis

Empiricism Objective observationControlOperational definitions Replication

Developing theories – explanations of how / why behavior works – is a core purpose of research.

Empirical data helps us:describe the world test hypotheses & develop theory.

Back NextHomePage










Back NextHomePage




Describe the world Initial approach to scientific study: “what is it” Leads to hypotheses

Predict events Core feature of a hypothesis: if “X” then “Y”. Often still descriptive rather than experimental.

Test theories Cause and effect questions involving hypothetical

constructs. Often controlled experiments or complex correlation

designs. Test applications of theories Using theory to model change Testing interventions or policy

Back NextHomePage



Science: A. Describing the world

Taxonomies or behavioral categories

Major personality "types"?

Categories of mental illnesses

"Types" of drug users.

Epidemiology rate of behavior or status x a population

Distribution of HIV/AIDS: sifts by time, place, demographics

Uniform crime rates

Distribution of drug use types across ages…

Direct behavioral description typically qualitative

Consumer decision making processes.

Actual mechanics of drug acquisition & use...

Back NextHomePage



Science: B. Prediction of events

Simple prediction

What test score or personal attribute predicts college success?

How can I predict which employees will develop a drug problem?

“Method of similarity” (Correlation)

What child rearing style correlates with extroversion?

What personality types correlate with drug use?

“Method of differences" (Experiment)

1. Two groups differ in one attribute

(Independent variable)-- an existing condition / behavior-- an imposed treatment

2. Do they also differ in a second attribute?

(Dependent var.)

solve a practical problem

test a theory

Test efficacy of a heroin agonist v. placebo in treating drug addicts

Back NextHomePage



Science: C. Testing theories

Direct cause & effect questions

What causes individual differences in academic ability?

How does personality create vulnerability to drug use?

Identifying basic psychological processes

How is language consolidated in the brain?

What brain & behavioral changes underlie drug tolerance?

Showing how processes are relatedMediationMediation:: Do drugs lead to risk by making people more

impulsive?

ModerationModeration:: Do drugs lead to risk primarily among men who are depressed? (Does depression create vulnerability to drug-related risk…?)

Back NextHomePage



Testing theory: Mediating effects

Simple empirical effect:

Mediating (theory testing) hypothesis:

Drug useRisky

behavior

Drug use ImpulsivityRisky

behavior

How does an effect “work”? Why or How does drug use lead to risk?

Where / how might we change it?

Back NextHomePage



Testing theory: “Moderating” effects

Moderating (theory limiting) hypothesis:

When or among whom does an effect “work”?

Where / how might we change it in different groups?

Drug use RiskDepressed men

Non-depressed men

Drug use Risk

Back NextHomePage



Science: D. Testing applications of theories

Using theory to design an intervention

Using basic learning theory to “teach” people to no longer have phobias.

Designing alternatives to drug use for people with high “sensation seeking” disposition…

Using an intervention study to actually test a theory

Comparing drug treatment to cognitive-behavioral treatment for depression.

Testing social network approaches to drug prevention among college students.

Back NextHomePage



What does science do: Summary?

Descriptive studies “who what where…”

Predict events Correlational studies Experiments / Hypothesis tests

Test theories“How / why it works

Testing applications of theories

Back NextHomePage










Back NextHomePage



The core features of a research study

Back NextHomePage



Basics of research: what is a theory?

Hypothetical Constructs Abstract statements of psychological processes…

““stress”, “depression”, “learning”, “attraction”…stress”, “depression”, “learning”, “attraction”…

…that cannot be directly observed: we observe their effects only.

Interview data, behavioral symptoms, questionnaire…Interview data, behavioral symptoms, questionnaire…

That are linked as a proposition. Specifies how one construct is related to another…

stress + genetics stress + genetics depression depression

…and Generally specifies what causes an outcome

Theory:Abstract statement of how two processes

relate to each other…

Answers why or how the phenomenon “works.”A theory has two core ingredients…

Back NextHomePage



How do we use theory in research?

Test a theory: does stereotype threat actually exist and govern performance?

Compare theories: Which best explains women’s statistics performance: stereotype threat or social role learning

Extend an established theory to a new outcome or phenomenon: can stereotype threat help us explain athletic as well as academic performance?

Apply a theory to change behavior: can I create instructions that relieve stereotype threat for women during statistics.

Back NextHomePage




Back NextHomePage



Basics of research: hypothesis

An hypothesis is a Prediction It links variables derived from the theory.

It implicitly specifies your idea of cause and effect.

Hypotheses are expressed in control terms for experiments.

If X then Y: if I make people relaxed their fear and if I make people relaxed their fear and loathing of statistics will decrease…loathing of statistics will decrease…

…and as a simple relation for measurement studies.

People who are (already) relaxed will tend to fear People who are (already) relaxed will tend to fear statistics lessstatistics less

…that is potentially falsifiable (see text for discussion)

Can be conceivably / logically shown to be untrue

Specific enough to be tested

Hypothesis A concrete statement of how processes relate to each other..

about variables derived from your theory.

Back NextHomePage




Back NextHomePage



Basics of research: methods

Core element of scientific approach Objective; designed to separate data from person Public: Copernican Revolution / Galileo

Replicable: others can repeat or expand the study

Test variables via operation definition Specify operations that express construct Define / understand variable in terms of operations

Methods

Depression“vegetative”; sleep, eating

Verbal behavior

AppearanceSuicide, drug use, work…

etc…

Turn variables into research procedures.

Back NextHomePage



What does it mean to operationalize a variable?1. Specify a manipulation that creates the

variable. Typical of the independent variable in experiments To relate stress to memory I may create “stress” in the lab via…

Threatening information. Shock. Requiring a difficult performance in front of others.

2. Specify a measurement to capture a variable For measurement studies and the dependent variable

in experiments. Measurement-based operational definitions of stress may be:

A questionnaire scale Heart rate

Anxious behavior, sleep loss, appetite change…

Back NextHomePage



Why use operational definitions?

Any theory must be operationalized to be heuristically useful

Generate concrete & testable hypotheses

Test or eliminates jargon, pop psych, new age constructs

Many ψ concepts are abstract, so their real meaning critically depends on an operational definition.

Attitudes “ Cognitive load”

Operational definitions orient us toward real world in theory development.

“Stress” I.Q.

Back NextHomePage



The limits of “operationism”

Science consists of theories & explanations, not just measures. Measures that do not have the goal of explaining a ψ

process are vacuous.

Science wants general laws, not measure-specific findings. A concept pertains to a class of measures (e.g., diverse

measures of depression, motivation, etc.), not one specific measure.

Ultimately, science does not care about the measures or the numbers

Back NextHomePage



Methods of operationally defining variables

Some variables are easy to operationalize; e.g., the effect of a drug dose on hypertension. IV = drug1 v. drug dose2 v. placebo

DV = blood pressure, serum measure, etc.

Some constructs can only be roughly operationalized. Variables such as “future orientation”, “identity integration”…

Some constructs have diverging operational definitions. How do you operationally define “stress”?

…motivation?

Some domains may not be operationalizable. String theory…

Relativity v. quantum mechanics views of gravity; indirect derivations can be tested, but not the core construct

“Spirituality”? “Happiness”?

Behavior?

Self-perception?

Physiological?

Back NextHomePage




Back NextHomePage



Basics of research: Data & Analyses:

Provide a Numerical representation (or

operational definition…) of reality

Rating scales: ratio, interval, ordinal, categorical

Statistics can be Descriptive

Simply characterize a phenomenon: “what is it?”.

Test a theory: “how does it work?”

“Statistical reasoning” is central to interpreting research.

We use the normal distribution & probability judgements to determine whether observations are meaningful

Statistics

or Inferential.

Back NextHomePage



Basics of research: Results Descriptive or measurement studies

typically address… A simple “empirical question”…

What % of adolescents use X or Y drugs? Demographic profile of an “undecided” voter?

Or an exploratory account of a question… What are the correlates of college success?

Experiments (and some measurement studies) always test a hypothesis:

How do we know if the hypothesis was supported? What statistical criteria did we use? Are there alternative explanations for the results?

Back NextHomePage




Back NextHomePage



Basics of research: Discussion Core issue: What are the implications of the

results for our theory. What does it mean that the hypothesis was (was not) supported? What future research does this lead to? What other hypotheses might these data support?

Study limitations: what are the boundaries on what this study can tell us?

Internal validity: How well did we model or represent the hypothetical constructs we were

interested in? Quality / nature of operationalization & design.

External validity: Our sample? Our manipulation or measurement of the independent variable(s)? Our assessment of the dependent or outcome variable(s)? The research setting itself

How representative was…

Back NextHomePage



Core features of a research study:

Hypothesis

Theory

Methods

Data & Analysis

Discussion

Hypothetical constructs In important relationship

More specific variables Falsifiable prediction

Operational definition Internal & external validity

Numerical representation Normal distribution Probability

Meaning of these results for the theory Limitations of methods: sample, setting, variables

Results Descriptive: Empirical question or exploration Hypothesis: Statistical significance

Back NextHomePage



Basic Elements of a Research Project

MethodsMeasurement v.

experimental

Move from the “big question” and theory…

Each element of the project corresponds to a later / earlier issue…

ConclusionsFuture research?

…then back to larger issues.

PhenomenonBig picture / question

Theory Hypothetical Constructs

Causal explanation

Hypothesis Operational definition

Specific prediction

Data / Results• Descriptive data• Test hypothesis

DiscussionImplications for theory

To specific methods, the core of a scientific study…

…to a concrete hypothesis…

To actual data…

Back NextHomePage










Back NextHomePage



Overall research strategies

Observation or Measurement Experiments

Simple Description Correlational Studies

Quasi-experiment

s

“True” experiment

sQualitative Quantitative

Explore the actual process of a behavior.

Describe behavioral or social trends.

Relate measured variables to each other to test hypotheses.

Test hypotheses or study naturally occurring event in field studies or less controlled experiments.

Test specific hypotheses via controlled “lab” conditions. Rich / detailed

description using direct observation, interviews, or existing text.

Typically small samples that are highly targeted, e.g., specific risk groups.

Computer analyses can link parts of text.

Simple counts, “blocked” by, e.g., age, gender, ethnicity.

Use probability or highly targeted non-probability sampling.

May use existing archival data as “markers” of psychological processes.

Often uses non-probability / targeted methods to sample specific groups.

Key: standard, reliable & valid scales (e.g., of attitudes) or behavioral reports (e.g., smoking).

Experimental design, but…

no control over Independent Variable

groups non-equivalent (not blind, not randomly assigned, self-selected…).

Manipulate Independent Variable, measure effects on Dependent Variable.

Control the IV and all observations, randomly assign participants, etc.

Back NextHomePage



Overall research strategies: Drug use



Quasi-experiment

s



Methods: Surveys, face-

to-face interviews, archival data (e.g., drug arrests, ER visits..)

Block by demographic variables (age, ethnicity…)

Methods: hypothesis-

oriented surveys or interviews (potentially with targeted samples: people in rehab., etc.).

Test ψ variables (motivation, emotions, attitudes…)

Methods:

Experimental-like design comparing two treatment groups.

Groups are non-equivalent (not blind, not randomly assigned, self-selected…).

Methods: Experimental design:

Operationalize drug “craving” in rats (DV),

Stimulate specific brain areas (IV) to map brain structure onto craving / drug-seeking.

Methods: Direct observation of “shooting galleries” or corner drug markets, in-depth interviews with drug users…

Research Question:

How does drug use actually occur?

Research Question:

Who tends to use drugs, how often, etc.? (epidemiology of drug use).

Research Question:

What social or ψ variables are associated with drug use?

Research Question:

Does one form of drug treatment work better than another?

Research Question:

What brain centers control “drug craving”?

Back NextHomePage



Overall Research strategies: measurement v. experiments

Many areas not amenable to true experiments, e.g., medical, educational, policy studies… Readings: Diet & health, mammography, maternal employment

Key: degree of control over variables

ExperimentHigh control / ‘lab’ conditions

Internalvalidity

Determine “cause and effect”: validly interpret data

MeasurementLess control; ‘research in nature’

Externalvalidity

Data can generalize to “real world” & capture more complexity

Core issue: controlled experiments are “gold standard” for testing hypotheses or treatments.

Back NextHomePage



Overall Research strategies: Validity



Quasi-experiment

s




Describe a behavioral or social trend.


Test hypotheses in naturally occurring events or field studies.

Test specific hypotheses via controlled “lab” conditions.

External validityInternal validity

Less control:

Observe / test phenomenon under natural conditions.

More accurate portrayal of: “how it works in nature” complexity of phenomenon

Less able to interpret cause & effect

More control:

Isolate (or create) the phenomenon in a lab or controlled environment

Addresses more specific questions or hypotheses

More ability to interpret cause & effect

Back NextHomePage



Research Strategies: Key issues

Key research issues

Observation or Measurement Quasi

Experiment

True Experimen

tSimple

Descriptive Correlational

Uses

Tests a Hypothesis?

Variables

Control

Statistics

Internal Validity

External Validity

Test causality, theory

Naturally occurring events or groups.

Non-experimental theory test

Exploration & description, epidemiology.

AlwaysTypicallyYes, or complex description

Often not, descriptive only

Manipulated & measured

Measured and/or manipulated

Subjective ratings, behavior

Behavior, text, status markers

High, via I.V. & exp. procedures

Moderate to high, except sampling

Moderate, via context or stats.

Little to moderate

Analyses of variance

Analysis of variance

Complex correlations

None or simple descriptive

Very highModerate to HighModerate, high in some designs

Often low to moderate

Often lowModerate to highModerate to highHigh (given sampling)

Back NextHomePage

88Psychology 242Introductionto Research overview

Overview


Back NextHomePage



Core course topics

How do we know things? What does scientific method tell us that other

methods (political, religious thought) do not?

What does science do? Describe the world

Taxonomies Epidemiology Qualitative research

Predict events Simple predictions Correlational studies Experiments

Test theories Cause & effect Identify basic processes Show how processes are

related

Test applications of theories

E.g., behavioral interventions

Back NextHomePage

90Psychology 242Introductionto Research Ways of knowing


Authority / authoritarianism Provides stable, core principles or beliefs Limits empirical evidence or alternative views

Intuition / subjective “hunch” Important source of novel hypotheses / theories / scientific

approaches Emotion-based “wishful thinking” or “magical thought” can

make us irrational or ignore / distort empirical facts.

Empiricism Grounds knowledge in “real” world, provides important

hypothesis-testing perspective Our perceptions are subject to cognitive / emotional biases.

Rationalism / theory Central purpose of science: coherent explanation of “why” or

“how” nature works. When subject to political pressure can limit hypothesis testing

or lessen respect for empirical evidence.

Back NextHomePage



key terms

Features of research: Key terms Theory Hypothetical construct Hypothesis Variable Operational definition Internal & external validity Independent v. Dependent variables

Back NextHomePage



Basic Elements of a Research Project

MethodsMeasurement v.

experimental

ConclusionsFuture research?

PhenomenonBig picture / question

Theory Hypothetical Constructs

Causal explanationHypothesis

Operational definitionSpecific prediction

Data / Results• Descriptive data• Test hypothesis

DiscussionImplications for theory

Back NextHomePage

93Psychology 242Introductionto Research Basics of major forms of research.




Quasi-experiment

s




Describe a behavioral or social trend.


Test hypotheses in naturally occurring events or field studies.

Test specific hypotheses via controlled “lab” conditions.

External validityInternal validity

Documents

Psychology 242, Introduction to Research Methods