Chapter 2: The Research Enterprise in Psychology

Empiricism / Empirical Method

Empiricism → A procedure for using empirical evidence to establish facts

Empirical method → A set of rules and techniques for observation

Operational definition

A description of a property in measurable terms

Power (detector)

The ability to detect the presence of differences or changes in the magnitude of a property

Reliability (detector)

The ability to detect the absence of differences or changes in the magnitude of a property

Demand characteristics

The aspects of an observational setting that cause people to behave as they think someone else wants or expects

Naturalistic observation

A technique for gathering information by unobtrusively observing people in their natural environments

Observer bias

The tendency of observers’ expectations to influence both what they believed they observed and what they actually observed

Double-blind study

A method used to avoid demand characteristics by ensuring that neither the researcher nor the participant knows how the participants are expected to behave

Frequency distribution

A graphic representation showing the number of times that the measurement of a property takes on each of its possible values

Normal distribution

• Bell curve, symmetrical, central peak, tails off to both ends, gaussian distribution

• When skewed → positive/right, negative/left

Median vs. Mode. vs. Mean

Mean →Value of the most frequently observed measurement

Median → Average value of all measurements

Mode → Value in the middle of the distribution

Range

Largest measurement minus the smallest

Standard deviation

Average difference between the measurements in a frequency distribution and the mean of that distribution

Variable

A property whose values can change across individuals and over time

Correlation and correlation coefficient

Correlation (‘r’) compares the pattern of variation in a series of measurements between variables

Third-variable problem

When two variables may be correlated to another but only because both are causally related to a third variable

Experimentation

A technique for establishing the casual relationships between variables

Independent vs. dependent variable

Independent → manipulated

Dependent → measured

Self-selection

Problem that occurs when anything about a participant determines the value of the independent variable to which the participant was exposed

Random assignment

Using a random event to assign people to the experimental or control group

Internal vs. External validity

Internal validity → results worked out, p < 0.05 (successful experiment)

External validity → experimental property where variables have been operationally defined in a normal, typical, or realistic way (used to draw broader conclusion)

Case study

Method of gathering scientific knowledge by studying a single individual (not generalizable)

Random sampling

Technique for choosing participants that ensure that every member of a population has an equal chance of being included in the sample

Type 1 vs. Type 2 Error

Type I → conclude that there’s a causal relationship between two variables when there is not (false positive/fluke)

Type II → conclude that there’s not a causal relationship between two variables when there is (false negative/flunk)

APA guidelines for humans

• Informed consent

• freedom from coercion

• protection from harm

• debriefing

• confidentiality

APA guidelines for animals

• must be trained in research methods and experienced

• minimize discomfort, infection, illness, and pain— unless no other alternative

• must be significant and beneficial to society

• surgery → anesthesia, minimize pain before, during, and after

Inferential statistics

Testing the significance of differences between groups to see if the effect we are observing is meaningful

Null hypothesis

Any observed differences between the samples are due to chance (assume this is true until rejected by statistically significant results)

Positive vs. Negative correlations

Positive correlations → increase = increase, decrease = decrease

Negative correlations → increase = decrease

Representative sample

Possesses the important characteristics of the population in the same proportions

Construct validity

A feature of operational definitions whose specified operations are generally considered good indicators of the specified properties

Why is it difficult to study humans?

Because we are more complex, variable, and reactive than almost anything else that scientist studies

Law of Parsimony

(Occam’s razor) → when there are multiple explanations for observed data, the simplest one is preferred

p < 0.05

probability that random assignment failed and that the results can be attributed to some other variable (chance) is less than 5%