Experiment & Evaluation of Quantitative Studies

Experiment

determine whether a cause-effect relationship exists between 2 variables

IV

causes change in other variable

DV

measured after the manipulation of the IV

Control

researcher manipulates the IV, all other variables stay the same

extraneous variable

variable other than IV causes unwanted changes in DV

Reactivity

participants expectations of research and researcher can affect trustworthiness of idea

Representational generalization

findings from qualitative research can be applied to populations outside the aimed population

Theoretical generalization

theoretical concepts derived from study can be used to develop further studies or other theories

Null

states IV will have no effect on DV or any change in DV due to chance

Reject the null

when you have enough statistical evidence to show a difference or an association

Operationalized

written in clearly measurable way

One-tailed

directional prediction

Two-tailed

no directional prediction

Experimental designs

organization of groups and conditions within a research

What are the 3 types of experimental designs

Independent groups

Repeated measures

Matched pairs

Independent measures

Randomly allocation into experimental and control groups

Independent measures advantage

confounding variables cancel out ; random allocation allows cause-effect inference

Repeated measures

All participants take part in all conditions of the experiment

Repeated measures advantage

participant variability

Repeated measures limitation

order effects (which condition comes first may impact next)

Repeated measures combat limitation

counterbalancing: reverse order of conditions

Matched pairs

a technique whereby each participant is identical to one other participant in terms of a third variable

What are the types of experiments?

Laboratory, field, natural, quasi

Lab

researcher to control for extraneous variables

Field

real world (cannot control for extraneous variables and cannot be easily replicated)

Quasi

participants are grouped based on a trait or behavior cannot be randomized

Natural

IV that is environmental in nature and outside of the control of the researcher

Sampling

process of finding and recruiting individuals for a study

Generalizability

extent to which results from the study can be applied beyond the sample

Types of effects

Expectancy

Screw You

Social desirability

Expectancy

participant attempts to discern the experimenter's hypothesis with goal of helping the research -> acting in a certain way or giving the right answer

Screw You

participants discern the experiment's hypothesis but only to destroy credibility

Social Desirability

participants answers in a way for them to look good to research ; avoid embarrassment and judgement

Types of biases

Ascertainment

Confirmation

Ascertainment

Systematic failure to represent equally all classes of cases or persons supposed to be represented in a sample

Confirmation

tendency to search for information that confirms one's preconceptions

Types of external validity

Population validity and ecological validity

Ecological (mundane realism)

extent to which tasks and manipulations of a study similar to real world contexts

Population validity

degree to which the study results can be generalized to and across the people in the target population

Inter rater reliability

measurement of agreement among observers on how they record and classify an event

Bidirectional ambiguity

no IV manipulated, no cause-effect relationship can be determined

Positive correlation

where as one variable increases, the other also increases as one decreases so does the other

Negative correlation

relationship between two variables in which one variable increases as the other decreases

Threats to internal validity

Selection

History

Maturation

Testing effect

Instrumentation

Regression to the mean

Experimental mortality

Demand characteristics

Experimenter bias

Selection

occurs if groups aren't equivalent at start of experiment ; differ from IV

History

outside events that happen to participants during experiment

Maturation

natural development processes

Demand characteristics

participants act differently simply because they know they're in a study

Testing effect

first measurement of DV may affect the second measurements

Instrumentation

instrument measuring DV changes slightly between measurement

Regression to mean

initial score on the DV is extreme

Experimental mobility

same participants drop out but not due to randomness

Experimental Bias

situations researcher exerts on influence of results

Confounding variable

cause-effect inferences ; factor other than IV might produce an effect

Types of Validity

construct, internal, external

construct

quality of operationalizations

Internal

methodological quality of experiment

external

generalizability of findings

Population - extent findings can be generalized from sample to population

Ecological - extent which findings can be generalized from experiment to settings

Types of triangulation

method, data, researcher, theory

Method

checking reliability of data thru diff data collection methods

Data

diff types of data from diff sources to make findings of study more reliable

Time

research is repeated in diff times to eliminate # of outside variables that could possibility impact the info collected during study

Triangulation

using multi. Methods of investigation to study same phenomenon and get rich data

Theory

involves looking at data using diff theoretical perspectives

Target population

group to which findings are generalized

Types of sampling

Random

Stratified

Convenience

Self-selected

Purposive

Snowball

Representative

Random

every member of target population has equal chance to be apart of sample

Stratified

deciding essential characteristics sample must reflect, study distribution on target population ; equal representation

convenience

recruit people must be available ; lead to biased results and it is problematic to generalize from studies that use opportunity sampling

Self-selected

volunteer based

Purposive

Targets a particular group of people ; obtain a sample quickly to investigate an urgent problem or if the desired population is rare or difficult to locate

Snowball

participants recruit other participants from among their friends and acquaintances

Representative

generalized to target population as whole