Chapter 10: Marketing Scales

Balanced Scale Example

1. How important is it to you to wear the latest fashions? __Very unimportant __Somewhat unimportant __Neutral __Somewhat important __Very important 2. Surfing the internet is: __Extremely good __Very good __Good __Bad __Very bad __Extremely bad

Unbalanced Scale Example

1. How important is it to you to wear the latest fashions? __Very unimportant __Somewhat unimportant __Somewhat important __Very important __Extremely important 2. Surfing the internet is: __Extremely good __Very good __Good __Somewhat good __Bad __Very bad

Non-Forced Choice Scale

A fixed-alternative rating scale that provides a "don't know" or "no opinion" category or allowing respondents to indicate that they cannot say which alternative is their choice.

Forced-Choice Rating Scale

A fixed-alternative rating scale that requires respondents to choose one of the fixed alternatives.

Semantic Differential Controlling Halo Effect

A seven-point rating scale with end points associated with bipolar labels that have semantic meaning. The negative adjective or phrase sometimes appears a the left side of the scale and sometimes at the right. Controlling the tendency of some respondents, particularly those with very positive or negative attitudes, to mark the right- or left-hand sides without reading the labels.

Graphical Rating Scales

A type of non-comparative scale. Place a response anywhere on a continuous line. Scales are normally anchored with antonyms - produces interval level. Means and standard deviations can be reported.

Itemized Rating Scales

A type of non-comparative scale. Respondents choose a response from a select number of items or categories - scales may use words or pictures for categories. Easy for respondents to understand & use. Most produce interval level data when it can be assumed that equal distance between category responses exists. Some produce ordinal data if equal distance cannot be assumed.

Rank-Order Scale Example

All of the following restaurants are located in the Oakview Mall area. Please rank the restuarants in terms of your personal preference from 1 being your most preferred to 7 being your least preferred.

No Opinion & Don't Know Opinions

Center point of odd-numbered points. Presence encourages its use. Usually have no opinion or don't know.

Rank-Order Scales

Comparative scale that tends to mimic reality. Consumers are familiar with the items being ranked. Best used for 5 or fewer items. Graphing results - ranking of one of the items, or ranking top choice.

Q-Sort Scales

Comparative scale to rank large sets of items. Sort items into groups based on some criteria. Quota size for each group may be determined by research resulting in normal distribution curve. Ordinal data is produced by this technique.

Non-Comparative Scales

Each object is scaled independently of the others in the stimulus set. The resulting data are generally assumed to be interval or ratio scaled.

Semantic Differential Scales

Finite number of choices; anchored by dichotomous words/phrases; can be comparative or non-comparative scale. Scales usually have 5 or 7 points, producing interval data. Scale anchors must be bipolar opposites. Easily answered when proper anchors are chosen. Often used to assess brand image or personality. Have high reliability, but halo effect may occur.

Paired Comparison Scale Example

For each of the following pairs of criteria in purchasing a laptop computer, indicate which item in each pair is most important to you by placing a checkmark on the appropriate line: ___Price or Size of computer___ ___Price or Physical Appearance___ ___Technical specifications or Price___

Non-Comparative Scale Examples

Graphical rating; Itemized rating

Itemized Rating Scale Examples

How likely are you to purchase a Dell computer? Extremely Unlikely (1) (2) (3) (4) (5) Extremely Likely

Comparative Scales

Involve the direct comparison stimulus objects. Comparative scale data must be interpreted in relative terms & have only ordinal or rank order properties.

Disadvantages of Rank-Order Scales

List may not be categorically exhaustive, or the respondent may not have knowledge or experience with all items listed. Difficult to rank middle items in a long list, and the criteria used in the ranking may not be clear. Produces ordinal data, not interval.

Constant Sum Scale Example

Listed below are the restaurants allocated in the Oakview Mall area. Please allocate 100 points among the seven restaurants based on your overall preference for each restaurant.

Considerations in Creating Marketing Scales

Number of categories; balanced vs. unbalanced scales; no opinion & don't know options; forced vs. on-forced choices; type of category description.

Forced vs. Non-Forced Choices

Odd number of points allow neutral position, while even number of points forces an opinion choice.

Semantic Differential Scale Examples

Please evaluate the last purchase you made at Home Depot and the experience you had at the retail store: Unfriendly Staff ( ) ( ) ( ) ( ) ( ) Friendly Staff Poor Selection ( ) ( ) ( ) ( ) ( ) Excellent Selection Poor Value ( ) ( ) ( ) ( ) ( ) Excellent Value

Graphical Rating Scales Example

Please evaluate the quality of service at Olive Garden by placing a large "X" on the line at the spot that most closely corresponds to your evaluation: Poor Service-------------------------------------Excellent Service

Likert Scale Example

Please indicate your level of agreement or disagreement with each of the following statements: 1. Wearing the latest fashion is important to me: Strongly Disagree; Disagree; Neutral; Agree; Strongly Agree 2. The brand name is important to me: Strongly Disagree; Disagree; Neutral; Agree; Strongly Agree

Scale Development

Process of assigning numbers to abstract concepts. Can be unidimensional or multidimensional scales.

Comparative Scale Examples

Rank-order; Q-sort; Paired comparison; Constant sum

Characteristics of a Good Scale

Relatively easy for respondents to understand; clear & concise; provides useful data; discriminates well; limited response bias; valid & reliable.

Number of Categories

Reliability increases with more categories. 3 categories lack discriminate ability, while 10 categories may be too many. 5 to 7 categories is optimal. Odd or even number categories.

Scale Category Considerations

Research objectives; information needs; research participants; mode of administration.

Constant Sum Scales

Respondents allocate points among various attributes - 10 or fewer items should be ranked. Relative distance between ratings can be assessed; ratio level data is produced. Used to assess brand preference or importance of attributes, benefits, or other characteristics.

Paired Comparison Scales

Respondents choose between two. Easier for respondents than ranking a series of items. Overcomes order bias associated with rank-order scales. All possible combinations must be listed. Limited number of items can be compared. Combinations [(n) * (n-1) / 2]

Non-Comparative Scaling Techniques

Respondents evaluate only one object at a time, so these scales are often referred to as monadic scales. Can be continuous or itemized rating scales.

Likert Scales

Respondents indicate level of agreement or disagreement with a series of statements. Very popular in marketing research because it is easy to create, and easy for respondents to understand & answer. Can be comparative or non-comparative. Uses 5 or 7 points, and produces interval level data. Scales should include reversed items.

Rating Scales

Scales that utilize evaluative labels, such as satisfied/dissatisfied, poor/excellent, etc.

Non-Comparative or Comparative Scale Examples

Semantic differential or Likert

Relative Advantages of Comparative Scales

Small differences between stimulus objects can be detected. Same known reference points for all respondents. Easily understood & can be applied, involving fewer theoretical assumptions. Tend to reduce halo or carryover effects from one judgement to another.

Taste Testing

The most common method is paired comparison. The consumer is asked to sample two different products & select the one with the most appealing taste. The test is done in private & a minimum of 1000 responses is considered an adequate sample.

Type of Category Description

Verbal or written description; number (all positive numbers or negative & positive numbers); anchor descriptions only.