2nd Quiz (Measurements) Flashcards
Conceptualization
- The process of specifying what we mean by a term
- Some concepts are straight forward
- Can find definitions from dictionary
Operationalization
Process of connecting concepts to observation Includes: 1. Measure 2. Method of data collection 3. Scoring information 4. Interpretation of the scores
Nominal measurement
Additional
- Value: Just “name”
- No mathematical interpretation
Ex: gender, ethnicity
Lowest level of measure
Ordinal measurement
Additional
- Rank-ordered
- Interval between values don’t have meaning
Ex: 1. low 2. medium 3. high
Interval Measurement
- Same distance between values
- No absolute 0
Ex: Temperature
Ratio Measurement
- Has absolut 0
2. Highest level of measure
Reliability
(Additional)
Consistency of measurement
Types of Reliability
- Internal consistency
- Test-retest reliability
- Alternative-form reliability
- Inter-rater reliability
Validity
Accuracy of a measurement
- Scale’s ability to measure what is design to measure
Types of Validity
- Face validity
- Content validity
- Criterion validity
- Construct validity
Construct Validity
Degree to which a measurement measures a similar concept or a different concept
Types of Construct Validity
- Convergent Validity
2. Discriminant Validity
Single Indicator
- Single item question
- Low reliability and validity
- Can’t calculate consistency reliability for a single item measure
Scale Indicator
Combination of set of items
-Better reliability and validity than Single Indicator
Face Validity
Lowes level of validity
Systematic Error
- Systematic error = Bias
- Affects average performance for the entire group
- Predictable
Ex: Construction on building/Everybody’s exam score goes down
Random Error
- Does not affect average performance for the group
- Unpredictable
- Can’t be controlled
Ex: You feel bad today. Your exam score goes down, the rest of the class does not.
Cronbach’s alpha
- Measure to consider acceptable internal consistency reliability should be >.7
Ways to reduce measurement errors
- Pilot: test the instrument you are going to use to measure
- Triangulate: Use different ways of measure (Exam, Paper, Presentation)
- Train Interviewers or observers
- Use unobtrusive measurements