39 terms

Marketing Research Chapter 10


Terms in this set (...)

noncomparative scale
only one object is evaluated at a time, an object is not compared to another object, respondents apply their own rating standard
types of noncomparative rating scales
continuous rating scales and itemized rating scales
continuous rating scale
respondents place a mark at any point along a line running between two extreme points rather than selecting from among a set of predetermined response categories ex) score could be a 28.637 and not just 1, 10, or 20
itemized rating scale
has a number of brief descriptions associated with each response category, categories are arranged in some logical order, most widely used
commonly used itemized rating sclaes
Likert scale, semantic differential, and stapel scales
likert scale
measurement scales with 5 response categories ranging between "strongly disagree" to "strongly agree"
advantages of likert scale
easy for researcher to construct and administer and easy for the respondent to understand, it is suitable for mail, phone, personal, or electronic interviews
disadvantage of likert scale
takes longer to complete than other itemized rating scales, it may be difficult to interpret
semantic differential scale
a 7-point rating scale on which the end points are adjectives representing opposites, respondents are typically asked to rate a store, brand
semantic differential scale set-up
negative adjective on left, favorable on the right
profile analysis
means or median values for each item are calculated, plotted, and statistically analyzed
the semantic differential is a highly popular scale because
of its versitality
major disadvantage of semantic differential
the difficulty in determining the appropriate bipolar adjectives required to construct the scale
stapel scale
named after Jan Stapel, typically presented vertically, measures attitudes that consist of a single adjective in the middle of an even-numbered range of values
data collected in a stapel scale are treated as
interval data, analyzed in the same way as semantic differential data
advantage of stapel scale over semantic differential
uses only one adjective, very simple
stapel scale is the least one of the three itemized rating scales because
it is confusing and difficult to apply
the ability to detect subtle differences in the attitude or the characteristic being measured, increasing the number of scale category increases sensitivity
balanced scale
the number of favorable and unfavorable categories or scale points is equal, in an unbalanced scale they are unequal
forced rating scale
the respondents are forced or required to express an opinion because the "no opinion" option is not provided
multi-item scale
consists of multiple items, where an item is a single question or statement to be evaluated
specific type of concept that exists at a higher level of abstraction than do everyday concepts
steps in developing a multi-item scale
develop the construct, develop a theoretical definition, develop an operational definition, develop a multi-item scale, evaluate scale reliability and validity, and apply the scale and accumulate research findings
systematic error
affects the measurement in a constant way and represents stable factors that affect the observed score in the same way each time the measurement is made
random error
arises from random changes and has a different effect each time the measurement is made
the extent to which a scale produces consistent results if repeated measurements are made
test-restest reliability
respondents are administered identical scales at two different times under as nearly equivalent conditions as possible
alternative-forms reliability
two equal forms of the scale are constructed
internal-consistency reliability
used to asses the reliability of a summated scale, where scores for seveal items are summed to form a total score for a construct
split-half reliability
simplest measure of internal consistency, scale items are randomly divided into halves and the resulting half scores are correlated
coefficient alpha
averaging the coefficients that result from all possibly combinations of split halves
the extent to which differences in observed scale scores reflect true differences in what is being measured, rather than systematic or random error
content validity
involves a systematic but objective assessment of how well a scale measures the construct or variable or interest
criterion validity
reflects whether a scale performs as expected given other variables considered relevant to the construct
construct validity
addresses the question of what construct or characteristic the sale is measuring
convergent validity
the extent to which the scale correlates positively with other measures of the same construct
discriminant validity
the extent to which a measure does not correlate with other constructs from which it is supposed to differ
nomological validity
the scale correlates in theoretically predicted ways with measure of different, but related constructs
if a measure if perfectly valid it is also perfectly
reliable. if it is unreliable, it is not perfectly valid