PSY 203 Exam 2

Theory

set of logically consistent statements about some phenomenon that best summarizes existing empirical knowledge of the phenomenon, organizes this knowledge in the form of precise statements of relationship among variables, proposes an explanation for the phenomenon, and serves as the basis for making predictions about behavior

Productive Theory Criteria

specific hypotheses, important question, internally consistent, coherent, specifies measurement of constructs, & generates new insights

Properties of Theories

Falsifiability and parsimony

Falsifiability

predictions are precise enough to be tested and potentially refuted

Parsimony

makes as few assumptions as possible to explain phenomenon; keep it simple

Levels of a Theory

hypothesis, constructs, and variables

Hypothesis

a testable statement describing the relation between two or more constructs

Construct

abstract quality that attempted to measure

Variable

any attribute that changes values across things that are being studied

Operational definition

quantify construct; specifies interpretation of construct and how it will be measured

Topographical definition

defining a response by its form

Functional definition

defining a response by its effect or outcome on the situation; an experimentally-determined relation that shows that the dependent variable depends on or is a function of the independent variable and nothing else

Construct Validity

successful operationalization and measurement of constructs of interest; measuring the same thing repeatedly

Inductive logic

logical process of reasoning from specific events (i.e., the results of research) to the general (i.e., the theory)

Deductive logic

reasoning from a set of general statements (i.e., the theory) toward the prediction of a specific event (i.e., the results of research)

Laboratory research

a research method that uses controlled elements as an analog of the natural setting that takes place under artificial conditions

Field research

a research method that uses some controlled elements traditional laboratory settings but takes place in natural, real-world settings

Realism

the philosophical assumption that we can study something more than our subjective experiences; we have direct access through our senses to the real world

Uniformity of nature

the universe is lawful and orderly; problem for quantum mechanics

Determinism

what is observed/studied can be described by cause-and-effect relation; one variable determines something else (i.e., an effect)

Discoverability

can identify/discover causes for effects via scientific method; an exercise in problem-solving

Disconfirming evidence

data do not fit with the hypothesis

Confirming evidence

data fit with the hypothesis

Type I error

rejecting a true null hypothesis; false positive

Type II error

failing to reject a false null hypothesis; false negative

Power

probability of rejecting a false statistical null hypopthesis

Effect size

any of various measures of the magnitude or meaningfulness of a relationship between two variables

Meta-analysis

a quantitative technique for synthesizing the results of multiple studies of a phenomenon into a single result by combining the effect size estimates from each study into a single estimate of the combined effect size or into a distribution of effect sizes

Confidence interval

range of values around test statistic where population statistic can be expected and located

Null hypothesis significance testing

in hypothesis testing, a set of procedures used to determine whether the differences between two groups or models are statistically significant; used to determine whether the null hypothesis should be rejected; involves comparison of the probability statistic obtained from the test to the chosen significance level whether an observed effect may be due to chance variance

Null hypothesis

hypothesis that there is no effect of the treatment

Alternative hypothesis

hypothesis that there is an effect of the treatment

Failing to reject the null hypothesis

the obtained p-value associated with a statistical test is larger than the critical alpha level (or chosen significance level), and any experimental effects are not different from chance variance

Rejecting the null hypothesis

the obtained p-value associated with a statistical test is smaller than or equal to the critical alpha level (or chosen significance level), and any experimental effects are likely not due to chance variance

Alpha level

in significance testing, a fixed probability of rejecting the null hypothesis of no effect when it is in fact true

Beta level

in significance testing, a fixed probability of failing to reject the null hypothesis of no effect when it is in fact false

Fallacy of affirming the consequence

the assumption that finding confirming evidence means that the premise is true; observing the effect and assuming that the independent variant is the only cause

Strong inference

data that simultaneously confirm one hypothesis and disconfirm another, alternative hypothesis

Confirmation bias

tendency to search for or remember evidence consistent with your beliefs or hypotheses and actively dismissing disconfirming evidence

Experiment

a series of observations conducted under controlled conditions to study a relationship with the purpose of drawing casual inferences about that relationship; requires manipulation of an independent variable with at least two conditions, measurement of a dependent variable, random assignment of participants to conditions, and holding control variables constant

Independent variable

what the experimenter manipulates; must have at least 2 levels; suspected cause of an effect

Manipulation

experimenter determines what treatment each individual participant receives and when

Comparison

treatment of interest compared to “reference” group or condition (e.g., control group); sometimes a separate group & sometimes within-subject

Control variables

what the experimenter holds constant or equates across conditions

Field experiment

a research method that uses some controlled elements traditional laboratory settings but takes place in natural, real-world settings

Situational variable

experimenter manipulates environmental/situational conditions (or features of the experiment); different environmental conditions for task

Task variable

experimenter manipulates problems or assigned questions/work; different problems to solve

Instructional variable

experimenter manipulates task instruction (but not the task); different instructions for the same task

Experimental group

participants in this group experience the “present” condition of the independent variable; receive the intervention

Control group

participants in this group experience the “absent” condition of the independent variable; do not receive the intervention

Extraneous variable

variable the experimenter is not interested in that is left free to vary and could potentially affect the dependent variable; potential confound

Dependent variable

what the experimenter measures or records; value depends on level of independent variable

Confound

something other than the independent variable(s) which systematically varies across groups and values of dependent measure are expected to increase with intervention

Ceiling effect

confound; measurement issue; no change in dependent variable because already at the maximum value of dependent variable, and values of dependent measure are expected to increase with intervention

Floor effect

confound; measurement issue no change in dependent variable because already at the minimum value of dependent variable, and values of dependent measure are expected to decrease with intervention

Subject variables

subject/participant attributes that cannot be manipulated (i.e., randomly assigned) or changed arbitrarily in an experiment (i.e., gender, age, socio-economic status, education, etc.)

Selection bias

selection effects; any effects are due to initial group differences (prior to study) and not independent variable

Internal validity

independent variable (and nothing else) was responsible for differences/changes in the dependent variable; determine if study methodology is sound

Posttest only design

subjects measured once, after the treatment (e.g., therapy and then rating test); includes manipulation

Pretest-posttest design

subjects measured before and after the treatment (e.g., rating test, therapy, and a second rating test); includes manipulation and comparison

Pretest-posttest with nonequivalent control group

participants not randomly assigned to groups; Group 2 (control) does not get the treatment/intervention; both groups measured before Group 1’s intervention and after

History

other variables operating with treatment; only those occurring once independent variable has begun and operating prior to assessment

Maturation

participants get older long-term/longitudinal studies

Regression to the mean

extreme scores on an initial test become less extreme on a second test

Testing effect

repeated measurement (i.e., testing) causes observed change in performance

Attrition

difference because more participants drop out of one group than another (e.g., 10 drop out of Group 1 but 40 drop out of Group 2)

Subject reactivity

know that you are part of a study; not behaving normally — being helpful or unhelpful or responding based on expectations

Quasi-experiments

quasi= almost; at least one non-manipulated variable (i.e., subject variable)

Sample

subset of a population

Population

the set of all individuals with shared characteristics of interest in a study

Random assignment

experimenter randomly determines how participants from the sample will be assigned to groups/conditions

Probability sampling

all individuals have a definable, nonzero chance of participating in a study

Random sampling

all individuals have a definable, nonzero chance of participating in a study

Random assignment

experimenter randomly determines how participants from the sample will be assigned to groups/conditions

Probability sampling

all individuals have a definable, nonzero chance of participating in a study

Quota sampling

experimenter has a specific number of each type of individual/characteristic that they wanted to find and stops collecting each type once quota is filled

Snowball sampling

experimenter finds one individual with relevant characteristics and used that individual to connect to similar individuals for the study

Subject/Participant representatives

subjects not randomly selected from population and any effects found could be due to existing differences between participants and not independent variable (i.e., gender, age, socio-economic status, education, etc.)

Reliability

characteristic of a test that refers to the extent to which a test results in consistent scores

Types of Reliability

test-retest, alternative forms, internal consistency, and interrater

Test-Retest Reliability

same test taken at different times gives same score

Alternate Forms Reliability

different forms of the same test give same result

Internal Consistency Reliability

individual items within the same test give same result

Inter-Rater Reliability

different test examiners and scorers (observers) rate the same behavior similarly