SCIENTIFIC METHOD

ASSUMPTIONS OF SCIENCE

Nature is:

Lawful: every event can be understood as a predictable sequence of cause and effect.

Deterministic:  behaivor is influenced soley by natural causes.

Understandable:  we assume we will eventually understand it.

ATTITUDES OF SCIENTISTS

Uncertain:  no one already knows how a behavior operates

Open-Minded:  any explination may be correct

Skeptical:  any statement may contain error

Causious: any conclusion is not fact

Ethical: should not do harm

CRITERIA FOR SCIENTIFIC EVIDENCE

Empiracle:  Learned by observation

Objective:  observations are made without bias

Systematic:  observations are made in a methodical fashion

Controlled:  elliminating factors that might influence the behavior being observed.

GOALS OF PSYCHOLOGICAL RESEARCH

Describe:  each behavior and the conditions under which it occurs

Explain: behaivors in terms of their causes

Perdict:  behaivors, know when it will occur or what will bring it about

Control: behaivors, we should be able to create the situation in which it occurs

SCIENTIFIC HYPOTHESIS

A formally stated expectaion about how a behavior operates

CRITERIA FOR SCIENTIFIC HYPOTHESIS

Testable:  It is possible to devise a test of the hypothesis

Falsifiable:  The test can potentially show the hypothesis is incorrect.

Percise: Terms are clearly defined

Rational:  It logically fits with what is already known

Parisimonious:  It is as simple as possible.

HYPOTHETICAL CONSTRUCT

 is an abstract concept used to relate and describe behaviors according to their underlying features.

EX:  We describle ppls mental capacities using the construct of intelligence.

OPERATIONAL DEFINITION

Defines a construct or variable in terms of the operations used to measure it.

EX: operational def of intelligence is a score on the wais.

INDEPENDENT VARIABLE

is a variable that is directly changed by the examiner.  We think it causes a change in behavior.

DEPENDENT VARIABLE

reflects some aspect of the participants or their behavior

DESCRIPTIVE STATISITICS

INFERENTIAL STATISTICS

Descriptive Stats:  are used to organize summarize and describe data.

Inferential Stats: are for deciding whether a sample represesnts a particular relationship.

EXPERIMENTS

Test causal hypotheses.

Researcher demonstrates a relationship by manipulating the independent variable.

A true experiement participants can be randomly assigned. A quasi esperiment they cannot be randomly assigned bc they already belong to a condtion based on some inherent characteristic.

EXTRANEOUS VARIABLES

are variables that can influence results, but are not intended to me studied. 

They can influence scores and alter sample relationships.

Lead to errors when interpreting bc they are opperating in addition to the variables we think are opperating.

4 TYPES OF EXTRANEOUS VARIABLES

Participant:  personal characterisitics and experiences of the particiapnt

Examiner: behaviors and charactisitics of the researcher.

Measurment:  aspects of stimuli presented or measurement procedure.

Enviornemental:  aspects of the enviornement

RELIABILITY

Is the extenet to which the same event or behavior produces the same score each time it is measured.

3 TYPES OF RELIABILITY

Correlation Coefficent is used to show reliability.

Inter-Rater:  the extent to which raters agree in the scores assigned to a behavior

Test-Retest:  Particpants tend to obtain the same score when tested at different times

Split-Half:  partipants scores on some trials consistently match their scores on other trials.

VALIDITY

The extent to which a procedure measures what it is intended to measure.

VALIDITY OF A MEASUREMENT

Content Validity:  is the degree to which a measurement reflects the variable of intrest.

Construct Validity:  the degree a measurement reflects the hypothetical construct of interest

Face Validity:  extent to which a measurment apears to measure what it is intended to measure.

VALIDITY OF A RELATIONSHIP

Internal Validity:  degree to which the relationship found reflects only the relationship between the intended variables.

External Validity:  degree to which the relationship found generalizes to other individuals and situation

Ecological Validity:  extent to which an experiement situation can be generalized to natural settings.

3 THREATS TO INTERNAL VALIDITY

selection:  differences bw groups exist that may interact with the independent variable, such as sex, weight, personality.  Controlled by random assignment and select.

history  events outside of the study, ex: large scale events that effect attitudes and behaviors. controlled thru random assignement and selection.

experiemental bias: ind's conducting the study affect the outcome by behaving differently to memebers of groups.  controlled by double blind study.

3 THREATS TO EXTERNAL VALIDITY

demand characteristics:  a cue that guides or biases a participants behavior. ex they may fake good. controlled by blind study so they dont know what they are being tested for and control group.

Hawthorne Effect: the presence of others watching causes change in performance. control group is used bc they are being watched also.

HOW DO YOU TEST FOR CONSTRUCT, CONTENT AND FACE VALIDITY?

Construct: convergent validiation, high correlation with other tests that measure same construct and factor analysis.

Content:  Expert judgement is primay method.

Face: ask anyone

SCALES OF MEASUREMENT

Nominal:  numbers name or identify a quality or characteristic. (male and female, assign males a 1 and females a 2)

Ordinal:  numbers indicate rank order. So 1 means the most and 2 means the second most.

Interval:  numbers measure a specific amount, and there is an equal distance bw separating any 2 scores, allows for negative numberfs, no zero

Ratio:  numbers measure a specific amount, and 0 truly means zero amount of something.

MEASURES OF CENTRAL TENDENCY

3 types

summarizes the location of a distribution of scores on a variable.

Mode: most frequently occuring score, sumarizes nominal data.

Median:  score at 50th percentile used with ordinal

Mean: average score located at the center of the distribution, used with interval and ratio

AFTER PERFOMING INFERENTIAL STATS IN ALL STUDIES DESCRIBE THE RELATIONSHIP BY:

SAMPLE

A subset of the population that is selected to represent the population.

PROBABLITY SAMPLING

any sampling technique where every potential participant in the population has an equal chance of being selected

NONPROBABLITY SAMPLING

any sampling technique where every potential particpant does NOT have an equal chance of being selected

3 TYPES OF SAMPLING

Stratified random: identification of imp subgroups in the pop. followed by the proportionate random selection from each sub group.

Cluster:  certian groups are randomly selected and all participants in each group are observed.

Convenience:  researcher studies the particiapants that are conveniently available.