Survey Methods: Measurement & Scale + Ch8 Flashcards
Quantifying the qualitative steps
- Identify and develop construct
- Define construct (eg: operationally: how do you measure it?)
- Develop a measurement scale
- Test and evaluate the measurement scale
6 qualitative concepts that can be quantified (BPRIBE)
Brand awareness
Patience
Risk attitude
Implicit attitude
Brand loyalty
Extroversion
Conceptual definition
Description at a theoretical level
Operational definition
Translation into observable phenomena
Brand loyalty definition
Consumers’ emotional attachment to a brand
Risk attitude
Propensity for taking or avoiding situations with probabilities of good and bad outcomes
Popular measurement is balloon test
Measuring personality traits (big 5 model OCEAN)
Openness to experience (curiosity, imagination, flex)
Conscientiousness (discipline, organization)
Extraversion (outgoing, upbeat)
Agreeableness (sympathetic, trusting, considerate)
Neuroticism (negative feelings)
Project Implicit
Implicit social cognition - “thoughts and feelings outside of conscious awareness and control”
Goal is to “educate the public about hidden biases and to provide a virtual lab for collecting data on the internet”
Implicit Association Test (IAT)
Tests the unconscious biases and beliefs
Reliability (2 criteria)
Ability to provide consistent data
Scale does not meet criteria if the the answer is no:
Test-retest
Internal consistency
Test-retest
Does the scale give similar results over time?
Internal consistency
Do different versions or portions of the scale give similar results?
Validity (6 criteria)
Measuring intended target
Scale meets criteria (yes):
Face validity
Content validity
Scale does not meet criteria (no):
Concurrent validity
Predictive validity
Convergent validity
Discriminant validity
Face validity
Does the scale seem appropriate for what its supposed to measure?
Content validity
Does the scale include the right questions and exclude the wrong ones?
Concurrent validity
Is the scale correlated with current related behaviours?
Predictive validity
Does the scale predict future behaviour?
Convergent validity
Is the scale measuring the same thing as similar scales?
Discriminant validity
Is the scale measuring something different from different scales?
2 measures that describe categorical variables
Nominal
Ordinal
2 measures that describe numerical variables
Interval
Ratio
Nominal measures
Describing categories that can only be categorized
Math: count, mode, freq
eg: gender
Ordinal measures
Describes ordered categories, but numerical measures are limited as its still categorical
Math: count, mode, freq, median
eg: WTP, asking income range
Interval scales
Describing numerical variables
Math: count, mode, freq, median, mean, standard dev
eg: WTP, exact income input, temperature
Ratio scales
Describing numerical values + having a true zero
Math: count, mode, freq, mean, standard dev, percent diff
eg: 0kg = lack of weight
eg: 0 degrees C is not ratio, but interval. There is no lack of temp
Graphic rating
Continuum anchored by extremes
eg: Yuck - Awesome
Itemized rating
Limited number of ordered categories on a scale
eg: Yuck - meh - good - awesome
Semantic differential scale
Interval scale that translates a person’s qualitative judgements into metrics that estimates is the semantic differential scale (good way to measure brand image)
Stapel scale
Numbers that range from a minus end to a corresponding plus end with or without a zero as the midpoint
