Research methods chapter 5
Concepts- in research methods, it is a word or symbol in language we use to convey meaning (make sure it is the same meaning)
Conception- is the mental image we have that represents a concept
Conceptualization- a process by which we make imprecise concepts more precise or specific
Conceptual definition- within a study, what is the precise definition (working definition) (the precise meaning the researcher has for the concept)
Example of this
- Concept- drugs
- Conception- an image of pills
- Conceptualization- make it known you saw pills and not something else
- Working definition- prescription drugs ( ADHD medicine)
Operationalization - describing how measurement will be made within a study
- how you are specifically going to measure your variables (how much, how often, what setting, etc.)
Ways to make observations- direct observation, indirect observation, construct observation
Direct observation- an observation you make as a researcher make first-hand (with your own eyes)
Indirect observations - are second-hand observations (use some other source of data)
EX: direct (use radar to track speed), indirect (look at speeding tickets)
Construct observation- complex concepts, no way to observe them directly or indirectly so you create a conduct
Ex: IQ (mental age divided by how old you are times 100) (mental state) (irresponsible driving)
Exhaustive- all possible attributes within a variable must be given
Mutually exclusive- each attribute is independent or distinct from one another
Example of exhaustive and mutually exclusive: favorite flavor of ice cream- exhaustive: chocolate, strawberry, orange, rocky road, cookies and cream, Oreo , other, don’t like ice cream, etc. - mutually exclusive: none of the ice creams is the same
Levels of measurement (impacts the conclusion we can draw)- always measure at the highest level of specification because a higher level of measurement can always be treated as a lower level of measurement, and a lower level of measurement can never be treated as a higher level of measurement
Nominal level- exhaustive and mutually exclusive (only characteristics)
Ordinal level- attributes are exhaustive and mutually exclusive but can be rank order as more or less of something (scale from 0-10 or 1-10)
Interval level- exhaustive and mutually exclusive, the attributes are being in order, but there is a standard or measurable (ex: asking age - 18-20, 21-23, 24-26,27-29) (DO NOT OVERLAP)
Ratio level- exhaustive and mutually exclusive, ranked order; at the ratio level, there is a true zero (ex: how many years in prison- zero)
Group example of levels of measurements: ice cream eating behaviors
- Ratio - how many days a week do you eat ice cream (0-7)
- Interval- 0-1,2-3,4-5,6-7
- Ordinal- never, occasional, often, daily
- Nominal - do you eat ice cream (yes or no)
Measurement reliability- the ability of a measurement to yield the same result when applied two or more times **(**is about consistent measure)(has to be reliable)
Ways to test for to accurate your tools are reliable - test-retest method, interrater lability, spilt-half method
Test-retest-method: to take a measurement twice (of the same thing)
Interrater liability- people gathering data; they want to make sure they are receiving the same information
Spilt-half method- like test-retest, but here you ask one set of questions to one group and the other set of questions to the other (should be measuring the same thing)
Measurement Validity - how accurate is it
Face Validity- particular empirical measures may or may not jibe with our common agreements and our individual mental images about a particular concept
Criterion-Related Validity- involves comparing a measure with some external criterion
Construct Validity- is based on the logical relationships among variables
Content Validity- refers to the degree to which a measure covers the range of meanings included within the concept