How to Get Started in Social Media for Art League City
Week 5 measurement, scaling, and questionnaire design
1. A formalized set of questions for obtaining information from respondents
Definition
Translate information needed into a set of specific questions
Questionnaire must uplift, motivate, and encourage the respondent to become
involved in the interview, to cooperate, and to complete the interview Objectives
Should minimize response error
1. Specific the information needed
2. Type the interview method
Questionnaire
3. Individual question content Assigning numbers or other symbols to characteristics of objects
according to certain pre-specified rules
4. Overcome inability & unwillingness to answer
Measurement
5. Choose question structure Definition
Questionnaire Design Checklist Involves creating a continuum upon which measured objects are located
6. Individual question wording
Scaling
7. Determine the order of questions
8. form & layout Classification
9. Reproduce the questionnaire Nominal Ex.: brand, production type (import, local)
10. Pretest
Classification
Order: diurutkan
Test / retest Ordinal
Ex.: ranking (1/2/3, education (SD/SMP/SMA)
Alternative forms
Reliability
Internal consistency Classification
Order
Content Primary Scales of Measurement
Interval Distance: jarak bermakna
Creation
Ex.: IP, TOEFL, rating, temperature
Convergent Internal Validity
Discriminant Construct Classification
Validity
Nomological MEASUREMENT, SCALING, Order
Scale Evaluation
& QUESTIONNAIRE DESIGN
Distance
Generalization Ratio
External Validity
Natural Origin: punya nilai 0 (nol) yang berarti tidak memiliki besaran
If a measure is perfectly valid, it's also reliable Ex.: jumlah uang, satuan ratio keuangan, tinggi badan, panjang, jarak, umur
If a measure is unreliable, it can't be perfectly valid
Involve direct comparison of stimulus objects
If a measure is perfectly reliable, it may or
may not be perfectly valid Relationship between reliability & validity Definition Must be interpreted in relative terms and have only ordinal or
rank order properties
Reliability is necessary, but not sufficient
(condition for validity)
Paired Comparison
Comparative Scales
Rank Order
Characteristics Classification
Constant Sum
Judges Research objective
Q-Sort
Scaling Techniques
Rating
Each object is scaled independently of the others in the stimulus set
Ranking Definition
The resulting data are generally assumed to be interval or ratio scaled
Categorization Research types
Sorting Continuous Rating Scales
Selecting a measurement scale Noncomparative Scales Likert
Data properties Classification
Itemized Rating Scales Semantic Differential
Number of dimension
Stapel
Balanced vs. Unbalanced
Forced vs. Non-forced
Number of scale point
Rater errors
week 5 Measurement, Scaling, and Questionnaire Design.mmap - 07/04/2011 - Mindjet