Sie sind auf Seite 1von 37

Measurement

Assigning Numbers To Reflect the Degree or Amount of a Characteristic

Measurement Scales
Series of items that are arranged progressively according to value or magnitude A series into which an item can be place according to its quantification

Be Careful.
The type of scale used will dictate:


the type of analysis that can be done the conclusions that can be drawn

MEASURES OF CENTRAL TENDENCY


MODE:

MEDIAN:

MEAN:

Nominal Scale
Identification only No order or meaning to the numbers Examples:

Measure of Central Tendency:

Ordinal Scale
Order to the data Distance between two numbers is unknown and uneven

Examples:

Measure of Central Tendency:

Interval Scale
Order to the data Equal distance between numbers No natural zero


Examples:


We assume a lot of scales are interval

Measure of Central Tendency:

Ratio Scales
Order to the data Equal distance between numbers natural zero where zero means none


Examples: Measure of Central Tendency:

YOUR TURN --Write a question for each type of scale


Nominal Ordinal Interval Ratio

Converting Theories into Measurement Instruments


Theory establishes the relationship among constructs


The constructs must be related to observable data

C C

M M M

Observable data

Need to convert the conceptual definition to an operational definition:


Conceptual definition:


Operational Definition:


Example: Operationalizing Consumer Sentiment Toward Marketing


Product Quality
The quality of most products that I buy today is as good as can be expected I am satisfied with most of the products I buy

Price of Products

Most products I buy are overpriced Most prices are reasonable considering the high cost of doing business Most advertising provides consumers with essential information Most advertising is annoying Most retail stores serve their customers well Because of the way retialers treat me most of my shopping is unpleasant .

Ads for products

Retailing

Gaski and Etzel Index of consumer sentiment toward marketing in Marketing Scales Handbook (2000)

Criteria For Good Measurement


Reliability-

Validity-

Sensitivity-

1) Reliability of Scales
Internal Consistency:
  

Coefficient alpha (inter-correlation among items) Are the results on questions measuring the same thing consistent? Single item scales more suspect to random error

Test/retest:


Are consistent results found on repeated measures

NOTE: Reliability is a necessary condition for validity but a reliable instrument may not be valid

Inter-judge reliability


Agreement between coders/judges evaluating an object

2) Validity of Scales
Content validity (Face validity)


Everything related to the construct is captured by the instrument

Construct Validity
 

Does the measure actually measure the construct, concept or trait that it is suppose to be measuring Look at convergent validity, discriminant validity, nomological validity

Pragmatic (Predictive) validity




Do the results of the measurement predict some other characteristic or behavior

3) Sensitivity
Refers to an instruments ability to accurately measure variability in stimuli or responses Example: I love to eat chocolate:
Agree Disagree OR

Strongly agree Mildly agree Neither agree or disagree Mildly disagree Strongly disagree

Types of Scales

General Distinction:
Dichotomous Scale:


Multichotomous Scale:


Open-ended Ratio:


Open-ended Nominal:


Continuous Graphic Rating Scales


Place a mark on the line indicating how important it is to have each of the following at your vacation resort: Alpine slides _____________________ unimportant important 5 inch line
 

127 mm 1/20 inch

Scaling?

Other Graphic Rating Scales


Happy faces

Thermometer

Uses??

Itemized Rating Scale (Category Scales)


Several categories from which the respondent can choose Example of a General Itemized rating scale:
How likely are you to buy a Sony DVD player in the next 3 mos. definitely will buy Probably will buy Might buy Probably will not buy Definitely will not buy

Common Forms of Itemized Rating Scales:


Likert Semantic Differential Staple

Likert Scales
Sentences with which the respondent agrees or disagrees Indicate your level of agreement or disagreement with the following statement: It would be cool to have a candy-red 1965 convertible Mustang SD D Neither A SA

Likert Scales
Code such that higher numbers indicate more or better things


May need to Reverse Code

Can create summated scales to form an index if the coefficient alpha is high Assume __________ scale

EXAMPLE of LIKERT SCALES

PLEASE X ONE BOX FOR EACH STATEMENT BELOW. I really want my heart to get better .. I would be very happy if my heart got better... I desire that my heart gets better .

STRONGLY DISAGREE 1 1 1 2 2 2 3 3 3 4 4 4

STRONGLY AGREE 5 5 5

It is important to me that my heart gets better Whether or not my heart gets better doesnt really matter to me... I feel like my heart is not as healthy as I would like..

1 1 1 1

2 2 2 2

3 3 3 3

4 4 4 4

5 5 5 5

I have been without a healthy heart for a long time.

Semantic Differential
Series of attitude scales where repeated judgments about a concept are made Opposite adjective words or phrases Use several of these and sum them if the coefficient alpha is high
Fast Bad Service Tasty Food __:__:__:__:__:__:__ __:__:__:__:__:__:__ __:__:__:__:__:__:__ Slow Good Service Not Tasty Food

Semantic Differential
Code such that higher numbers indicate better things or more of something


May need to Reverse Code

Make an overall score--sum the items Develop a snake diagram (image profile) to compare competitors Assume _____ scaling

Staple Scale
Use +5 (describes completely) to -5 (does not describe at all) Easy to construct Good for phone May look difficult for respondent
Service is Fast Food is Tasty -5 -4 -3 -2 -1 +1 +2 +3 +4 +5 -5 -4 -3 -2 -1 +1 +2 +3 +4 +5

Assume _______ scaling

Considerations for Itemized Response Scales


7 +/- 2 categories work best


Odd vs even number scales

Equal number of positive and negative categories Consider reverse scaling but use sparingly Distinguish undecided from neutral
Put undecided or no opinion at the end

Comparative Scales
Compare one set of objects directly with another
  

sensitive easy can create artificial differences

Paired Comparison
Which do you prefer?
Barry Manilow Red Hot Chili Peppers Barry Manilow Tom Petty Tom Petty Red Hot Chilli Peppers

Paired Comparison Table (Cross Tab)


Manilow Manilow RHCP ----0.10 RHCP 0.90 ---0.40 Tom Petty 0.85 0.60 ----

Tom Petty 0.15

* % of time that the column was chosen over the row

Calculation of Rank-Order Values


Manilow RHCP Tom Petty Manilow 0 0 0 RHCP 1 0 1 Tom Petty 1 1 2

*If the % score was 50% or more = 1 *If the % score was less than 50% = 0 Higher score indicates greater preference

Rank-order Scales
Respondents are simultaneously presented with several objects/attributes that they rank order Please rate the following from 1=most preferred/important to 4= least preferred/important
price cleanliness location food selection

Rank-order Calculations
Calculate frequencies of responses for each item Multiply frequency by response Lower numbers indicate greater preference See Example - Handout

Comparative Continuous Graphic Rating Scale


Similarity ratings used for perceptual maps Pitt and WVU _________________________ Exactly Completely the same different

Constant Sum Scales


Assign chips or points to attributes Very careful with instructions Difficult for the respondent use with high involvement/education

Das könnte Ihnen auch gefallen