Exam 1 Research Methods

5 ways by which information about the world may be obtained

5 ways by which information about the world may be obtained

1. Method of tenacity (long held beliefs)

2. Method of authority (rely on experts)

3. Reason (logic & rationality)

4. Personal experience

5. Empiricism (based on the senses)

the scientific method relies on...

empiricism

the basic characteristics of the scientific method that distinguish it from other ways of knowing...

1. It is empirical - based on information perceived by the senses. It is based on data by systematic observation.

2. It is analytic - based on the idea that you are taking a complex idea and breaking it down into constituent parts.

3 basic procedures used in scientific inquiry

1. Description

2. Prediction

3. Explanation

which basic procedure used in scientific inquiry permits causal inferences to be made?

prediction

inductive reasoning

data --> theory

• uses data then draws conclusions from the data

deductive reasoning

theory --> data

• starts with a hypothesis and examines all possibilities to reach a specific conclusion

the 2 major functions of a theory

1. Organize data (ex: periodic table)

2. Generate predictions

why cant a theory be considered "proven" when its supported by experimental data?

there can be multiple factors affecting the theory. Once one factor confirms the theory, further experimentation needs to be performed to investigate other factors contributing to that theory. Theories are ever-changing and are unlike laws. Theories are constantly being evaluated and experimented on resulting in improvement and deeper explanation over time, whereas laws remain stagnant and their idea does not improve/change, as it remains the same from start to finish.

the 3 criteria for evaluating theories

1. Parsimony

2. Precision

3. Testability

parsimony

can explain many occurrences with few statements

precision

the more precise, the better (best if quantitative)

testability

if can't test the theory, it can't be falsifiable

"strong inference" in comparing different theories

A strong inference is when one theory is pitted against another theory. For example, theory A predicts outcome 1 and theory B predicts outcome 2. The two theories are pitted against each other, resulting in a confirmation on one outcome (i.e. outcome 1), which rejects the other theory (i.e. theory B).

theory

a principle that has been formed as an attempt to explain things that have already been substantiated by data.

hypothesis

usually tentative, an assumption or suggestion made strictly for the objective of being tested.

major threats to validity in naturalistic observations

• use unobtrusive observations

• use unobtrusive measures

• habituate subjects to presence of observer

• use deception as to what is really being measured

major threats to validity in case studies

• problems include forgetting, "repression", and observer bias

• get corroborative evidence

major threats to validity in surveys

• response style: habitual ways of responding

1. response acquiescence (yeah saying)

2. response deviation (nay saying)

• social desirability: use forced choice between equally desirable/undesirable alternatives

3. sampling issues

• is the sample representative of the population?

• is the response rate adequate?

• volunteer problem: volunteers differ in various ways from non-volunteers, therefore limiting generality.

The Pearson Product-Moment correlation coefficient (r)

measures the direction and strength of the linear between 2 variables.

r = 0

no correlation between the 2 variables

r = -1

perfect negative correlation

r = +1

perfect positive correlation

Why doesn't a strong correlation imply causation?

A strong correlation does not imply a causation because there may be a third variable resulting in the correlation.

Why doesn't a '0' correlation necessarily imply a lack of relationship between two measures?

A '0' correlation does not necessarily imply a lack of relation between the two measures because there could be a truncated range (too little of data) resulting in the relationship not being linear.

what 3 things does "control" denote?

1. a control condition, for purposes of comparison

2. the treatment is produced or manipulated

3. certain aspects of the experimental setting are held constant

independent variable

variable that is manipulated

dependent variable

variable being measured

control variable

variable that is kept constant by the experimenter

retrospective studies

A study that monitors people who have been exposed to an environmental hazard at some time in the past.

• extraneous variables may not be comparable

true experiments

designs in which the researcher manipulates all of the independent variables.

• the experimenter can directly control the variables at the same rate, therefore allowing for equal comparison.

Critique the following statement: When there are no significant differences between the experimental and control groups, we can safely conclude that the independent variable does not have an effect.

When there is no difference between the experimental and control groups, this is termed a null effect. It is difficult to interpret the cause and we cannot safely include that the independent variable does not have an effect. The reason may be due to:

1. IV having no effect

2. IV not adequately manipulated

3. Problem with the DV (ex: floor or ceiling effects)

4. Insufficient control of extraneous (control) variables

interaction

When effects produced by one IV are not the same across levels of a second IV (one IV depends on the level of another IV).

Who is responsible for overseeing that research at universities involving human subjects is conducted ethically?

The IRB is responsible for overseeing that research at universities involving human subjects.

What are the major criteria for defining a research project as "ethical"?

For a project to be ethical, they need to follow the APA guidelines:

1. informed consent

2. confidentiality

3. debriefing

4. deception

What is the rationale for the claim that animals have rights and should not be used as subjects in research projects?

The rational for the claim that animals should have rights and should not be used as subjects in research projects is:

1. all animals can suffer (animals = humans; animals, like humans, have rights)

2. "A rat is a pig is a dog is a boy" (PETA)

• species-ism is like racism

What is the counter-argument put forth by Cohen (1986) regarding animal rights?

The counter-argument put forth by Cohen is that animals do not have a moral "sense".

Explain two advantages of using animals as research subjects.

1. Neuroscience

• recovery from brain damage

• fetal alcohol syndrome

• effects of aging on learning and memory

• Alzheimers and Parkinsonism

2. Alleviation of pain- discovery of pain suppression system- endorphins

who is responsible for overseeing that research at universities with animals is conducted ethically AND sets the standards governing proper use and care of animal subjects?

A. IACUC B. APA Guidelines C. NIH Guide for the Care and Use of Laboratory Animals D. U.S. Dept. of Agriculture Regulations

method of tenecity

A method of acquiring knowledge in which information is accepted as true because it has always been believed or because superstition supports it

long held beliefs

method of authority

A method of acquiring knowledge in which a person relies on information or answers from an expert in the subject area.

rely on experts

the scientific method

A series of steps followed to solve problems including collecting data, formulating a hypothesis, testing the hypothesis, and stating conclusions.

the scientific method is both...

empirical and analytical

The scientific method is empirical as it is...

based on information perceived by the senses. It is based on data obtained by systematic observation.

systematic observation

careful observation of one or more specific behaviors in a particular setting

Causal Observation

engage on a daily basis

ex: walking by animals at the zoo

causal inference

A statement about cause and effect that claims that a change in one variable is the cause of a change in another variable.

causality requires...

1. covariation

2. temporal order

3. other factors to be ruled out (i.e., controlled)

self-correcting

a system of challenges by which scientific claims can be verified.

~ importance of replication ~

The scientific method is analytic as it is

taking a complex phenomena and breaking it down into constituent parts.

It can be confirmed by synthesis (qualitative and quantitative analysis).

• Real ex: in a chemistry lab

• Theoretical ex: motivation for hunger, sex, and money by studying behavior

experimentation

manipulate variables thought to underlie the phenomenon (i.e. IV - the manipulated variable; DV - the measured variable/the effect)

experimental group

independent variable + extraneous variables

covariation

the amount of change in one variable that is consistently related to the change in another variable of interest

as x changes, y changes

control group

no independent variable + extraneous variable

temporal order

the change in x precedes the change in y

extraneous variable

Something unwanted or unexpected that might affect the dependent variable.

null effects

a finding when there are no differences between the experimental and control groups

does confirmation prove the theory?

no, confirmation does not "prove" the theory; only disconfirmation is useful (falsifiability)

advantages of multiple IV's

• more efficient than multiple single experiments

• better experimental control (easier to hold control variables constant with multiple single experiments)

• allows for interactions

Falsifiability

a feature of a scientific theory, in which it is possible to collect data that will prove the theory wrong

ceiling effect

all the scores are squeezed together at the high end

floor effect

all the scores cluster at the low end

why use animals in research?

1. ethics

2. experimental control

goals of observational research

1. to describe behavior as first step in future experimental studies (i.e. predatory behavior of cats; mating behavior in stickleback fish)

2. to answer specific questions (i.e. do parents treat their sons and daughters differently in science museums?)

Types of uses for animal research

A. Biomedical / behavioral research B. Teaching C. Drug & product testing

threats to validity

Possible problems in a study's design

1. observer bias

2. reactivity

observer bias

systematic errors in observation that occur because of an observer's expectations

objection to animal research

1. animal welfare

2. animal "rights"

to prevent observer bias...

• use well-developed coding system and operational definitions

• use rigorous observer training

• assess inter-rater reliability

• keep observers "blind" to experimenter's hypotheses

moderates

persons who do not hold extreme political views

militants

People for the Ethical Treatment of Animals (PETA); against use of animals in research

correlational research

a statistical association between variables; variables are measured, not manipulated as in experiments (ex: ovarian cancer and tea)

terrorists

Animal Liberation Front (ALF); effectively end animal testing with violence

Institutional Review Board (IRB)

A committee at each institution where research is conducted to review every experiment for ethics and methodology.

informed consent

an ethical principle that research participants be told enough to enable them to choose whether they wish to participate.

deception

Sometimes, it may be necessary to withhold some information from participants. This is to minimize participant's reactions.

• deception is not permissible if it inflicts physical or psychological damage on participants (Milgram shock)

risk/benefit ratio

analysis of whether the research is important enough to warrant placing participants at risk.

vulnerable populations

• children

• prisoners

• mentally and physically handicapped

• pregnant women and fetuses

• animals??

scientific fraud

• Plagiarism

• Exaggerating data to make it look more significant

• Fabricating data

• Excluding data that goes against your hypotheses

Tuskegee Syphilis Experiment

Researchers in this experiment were studying the effects of syphilis, but once a cure was developed, they did not provide it to the study participants.

Institutional Animal Care and Use Committee (IACUC)

A committee at each institution where research is conducted to review every experiment involving animals for ethics and methodology.

APA Guidelines

1. informed consent

2. confidentiality

3. debriefing

4. deception