Please enable JavaScript.
Coggle requires JavaScript to display documents.
Week 3A
Week 3A
operationalization
defintion
no concept
simple words
no synonym
contains all needed info (complete)
measurement scales
likert
ordinal- points to rank how you feel about indicators; numbered
semantic differential
word distinction (words that describe the variable ex. expensive vs cheap
two opposites
manifest variables
directly observable ex. male/female
latent variable
not directly observable ex. masculine/feminine
indicators
conceptual model
main concept (variables
independent (outcome)
dependent (outcome)
schematic overview of hypothesis
indicates realtionship
arrows and direction; types of relationships
research validity (validity of study as a whole)
internal validity
no alternative explanations
high= draw causal conclusions
low= cannot draw causal conclusions
external validity
results are generalizable
Units of analyses are representative of population
ecological validity
Representativeness of test conditions (applicable to real life situations?)
scientific statements
values
possible category per variable
units of analysis
variables
measurement validity
content
convergent validity
relating your measurements to another measurement and comparing answers
divergent validity
not wanting there to be any association between two variables/concepts
construct
concurrent validity
comparing test to well established ways of measuring
predictive validity
how well a measure can predict future outcomes/behavior
criterion
expert/panel validity
experts in the field evaluating measurements
face validity
asking participants to tell you what they think you are measuring
their answer tells you if you are measuring what you intend to be measuring, then you are good
validity
how much a measure is free of systematic error (bias)
example:
socially desirable answers
respondent knows goal of study
respondent wants to meet expectations of researcher
measurements must measure what they are supposed to measure
does the measure capture what we are interested in and does it give us significant results
reliability
how much a variable is free of random error (noise)
as many negative as positive errors (cancel each other out)
does not affect group performance
stable and consistent results
moderated relationship
third variable that explains relationship
mediated relationship
has a third variable that influences strength of a relationship between two variables
internal reliability
split had
inter coder/observer reliability
at least two researchers' independently coding the materials
established measure reliability
similar to concurrent
test re-test reliability
intra coder reliability