Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 3 (Part 2): Research Design - Coggle Diagram
Chapter 3 (Part 2):
Research Design
Reliability
measure indicate the extent to which it is without biased (error free) & ensure consistent measurement across time & across various item in the instrument.
Measurement
The assignment of numbers or other symbols to the characteristic (or attribute) of objects according to a prespecified set of rules.
Characteristics
gender, length, weight, diversity, service
Objects
persons, companies, countries, bicyles, balls
Operationalising the concept:
Reduction of abstract concept to render them measurable in a tangible way
Step 1: Define the variable
select measurable phenomena
Step 2: Design the content of measure
develop a set of mapping rules/ items/ questions/ instrument
Step 3: Determine the response format
Apply mapping rule to each phenomenon
eg: 5-point rating scale
Step 4: Assess the validity and reliability of measurement
Scale
A tool or mechanism by which individual are distinguished as to how they differ from one another on the variables of interest to our study
1. Nominal Scale (Classification)
Allow the researcher to assign subject to certain categories or groups.
Mutually exclusive
Collectively exhausting categories
Exhibits only classification
Eg: gender, religion, hometown, nationality
2. Ordinary Scale (rank & classification)
Rank orders categories in meaningful way
does not give indication of magnitude of the difference among the ranks.
Eg: age group, level of education, preference rating
3. Interval Scale (numerical)
Characteristic of nominal & ordinal scale
measure the distance between any two points on the scale.
Eg: likert scale (strongly agree to strongly diasgree, temperature
4. Ratio Scale (range)
Overcome disadvantage of the arbitrary origin point of the interval scale, that it has absolute zero point which is meaningful point.
characteristic of nominal, ordinal, inverval scales
absolute zero
eg: weight, height, age, income
Validity
A test of how well an instrument that is developed measures the particular concept it is intended to measrure
1. Content Validity
ensure the measure include adequate & representative set of item that tap the concept.
2. Face Validity
considered by some basic & minimum index of content validity.
indicate the item are intended to measure concept.
3. Criterion-related Validity
Concurrent validity
Predictive validity
4. Construct Validity
Testify how well the result obtained from the rest of measure fit the theories around which is test is designed.
Convergent Validity
Discrimination Validity
~ when 2 variables are predicted to be uncorrelated = correlation analysis, factor analysis
Stability of Measures
Ability of measure to remain same over time, despite uncontrollable conditions
1. Test - Retest Reliability
The reliability coefficient obtained with a repetition of the same measure on a second occasion.
2. Parallel- Form Reliability
Responses on two comparable sets of measures tapping the same construct are highly correlated.
Internal Consistency of Measures
Indicative of homegenity of the items in the measure that tap the construct.
1. Interitem Consistency Reliability
Test of consistency of respondents' answers to all the item in a measure.
Most popular test is the
Cronbach's coefficient alpha.
2. Split-Half Reliability
Split half reliability reflects the correlation between halves of an instrument,