Table of Contents:-
- Scaling Meaning
- What is Scaling?
- Criteria for a Good Scaling
- Issues in Designing Attitude Scales
Scaling may be considered as an extension of the measurement process. It entails establishing a continuum on which measured objects are positioned. It refers to the process of assigning numerical values to different levels of opinion, attitude, and other concepts. Scaling is a branch of measurement that focuses on creating instruments that link qualitative constructs with quantitative metric units.
A scale is a component consisting of the highest point (in terms of some characteristic e.g., preference, favourableness, etc.) and the lowest point along with several intermediate points between these two extreme points. The positions of these scale points are interrelated in such a way that when the first point is the highest, the second point signifies a greater degree of a specific characteristic compared to the third point, and the third point signifies a higher degree than the fourth, and so forth.
What is Scaling?
The term ‘scaling’ is applied to the procedures for attempting to determine quantitative measures of subjective abstract concepts. Scaling is a method used to assign numbers (or other symbols) to the attributes of objects, to introduce some numerical characteristics to these properties.
A scale can be described as a continuous range or a series of categories. It’s a collection of items arranged in a progressive order based on their value or magnitude, allowing us to place an item within it based on its quantified measurement. The purpose of scaling is to represent usually quantitatively, an item’s, people. or an event’s place in the spectrum.
Scaling has also been defined as a method for assigning numbers (or other symbols) to the attributes of objects, thereby imbuing these properties with some of the characteristics typically associated with numbers. It involves arranging a series of items in a progressive order based on their value or magnitude, allowing us to place an item within this series based on its quantified measurement.
Scaling thus constitutes a series of categories through marks. Its purpose is to represent quantitatively an item’s (person’s) place in the spectrum.
- nature of marketing
- difference between questionnaire and schedule
- features of marginal costing
- placement in hrm
- limitations of marginal costing
- nature of leadership
- difference between advertising and personal selling
Criteria for a Good Scaling
There are seven important criteria for ascertaining whether the scale developed is good or not:
Criteria for a Good Scaling are explained below:
Reliability testing is another crucial aspect of sound measurement. A measuring instrument is considered reliable when it consistently produces the same results. A reliable measuring instrument does contribute to validity, but a reliable instrument need not be a valid instrument.
Sensitivity refers to an instrument’s capacity to precisely measure variations in stimuli or responses. Sensitivity is not high in instruments involving ‘Agree’ or ‘Disagree’ types of response. When there is a need to be more attuned to subtle changes, the instrument is adjusted accordingly. For instance, including categories such as ‘strongly agree,’ ‘mildly agree,’ ‘mildly disagree,’ ‘strongly disagree,’ and ‘none of the above’ can enhance the scale’s sensitivity.
Validity is the most crucial criterion and signifies how accurately an instrument measures what it’s intended to measure. Validity can also be thought of as utility. In other words, validity measures how well the differences identified by a measuring instrument genuinely reflect distinctions among the individuals being tested.
The practicality characteristic of a measuring instrument can be judged in terms of economy, convenience and ease of interpretation. From a practical standpoint, a measuring instrument should be cost-effective, convenient, and easy to interpret.
The selection of a data collection method is frequently influenced by economic considerations. The rising com of personal interviewing first led to increased use of telephone surveys and subsequently to the current rise in Internet surveys. In standardised tests, the cost of test materials alone can be such a significant expense that it encourages multiple reuses.
A measuring device passes the convenience test if it is straightforward to administer. A questionnaire or measurement scale that includes clear and detailed instructions along with examples is easier to complete accurately compared to one without these features. In a well-prepared study, it is not uncommon for the interviewer’s instructions to be several times longer than the interview questions. Naturally, the more complex the concepts and constructs, the greater the need for clear and complete instructions.
Generalisability refers to the amount of flexibility in interpreting the data in different research designs. The Generalisability of a multiple-item scale can be analysed by its ability to collect data from a wide variety of respondents and with reasonable flexibility to interpret such data.
Issues in Designing Attitude Scales
No single scaling device is considered the best choice for all measurement situations. We will discuss the most commonly used scales alongwith some for their use. The following are some commonly used scales for variables of interest for professional service marketers. Researchers using one of these scales face several choices:
1) Odd Number Versus Even Number of Options
Is a scale with an even number of points preferable to an odd number, or is the odd number preferable? No hard evidence supports either choice. An odd number on the scale offers a midpoint, whereas an even number compels respondents to make a clear choice, leaning toward one end of the scale or the other. This remains a matter of research preference.
By creating a scale with an odd number of categories, a researcher will be leaving a mid-point, which acts as a neutral option for respondents to select. Without it, respondents will be forced to pick an option on either the lower or higher end of the rating scale. Odd numbered scales are generally regarded as allowing for a “neutral” option such as ‘neither agree’ nor ‘disagree’, or ‘do not care’. However, advocates of even-numbered scales argue that in reality, people are never neutral on issues and always have an opinion, even if they had not previously conceived of it.
2) Number of Scale Points
Research suggests that scale reliability improves to a certain extent when there are more item statements and scale points. So, e.g., a questionnaire that uses twelve attitude item statements is more reliable than one that uses only three item statements, but thirty-item statements are probably less reliable than twelve. Why? Because of the fatigue factor- respondents get fatigued in answering too many questions and find their responses are made more to “get it over with” rather than as a true representation of their state of mind. The same is true for response points on the scale – five is better than three, but more than ten points do not improve reliability. Typically, the number of scale points varies between five to seven.
3) Forced Versus Unforced Scale
An unforced-choice rating scale allows participants the option to indicate no opinion when they cannot make a clear choice among the provided alternatives. A force-choice scale requires that participants select one of the offered alternatives. Marketing researchers often exclude the response choice “no opinion”, “undecided”, “do not know”, “uncertain” or “neutral” when they know that most participants have an attitude on the topic. It is reasonable in this circumstance to constrain participants so that they focus on alternatives carefully and do not idly choose the middle position.
However, when many participants are undecided and the scale does not allow them to express their uncertainty, the forced-choice scale biases the results. Researchers discover such bias when a larger percentage of participants express an attitude than did so in previous studies on the same issue. Some of this bias is attributable to participants providing meaningless responses or reaching to questions about which they have no attitudes. This affects the statistical measures of the mean and median, which shift toward the scale’s mid-point, making it difficult to discern attitudinal differences throughout the instrument. Understanding neutral answers is a challenge for marketing researchers. In a customer satisfaction study that focused on the overall satisfaction question with a company in the electronics industry, an unforced scale was used.
4) Verbal and Pictorial Description of Response Categories
The nature and degree of verbal description associated with scale categories vary considerably and can affect the responses. Scale categories can include verbal, numerical, or even pictorial descriptions. Furthermore, the research must decide whether to label every scale category, some scale categories, or only extreme scale categories. Interestingly, offering a verbal description for each category may not necessarily enhance the accuracy or reliability of the data. Yet an argument can be made for labelling all or many scale categories to reduce scale ambiguity. The category description should be placed as near as possible to the response categories.
A scale can have numerical verbal or pictorial descriptions associated with the scale points. In some cases, researchers label extreme scale points. In some other cases, the researchers label every scale point. As a general rule, the description of the scale point should be close to the concerned point. Furthermore, labelling all the scale points helps researchers avoid any ambiguity in the scale. These are the general recommendations, although the final decision is a matter of the researcher’s wisdom.
5) Balanced Versus Unbalanced Scale
A balanced rating scale has an equal number of categories above and below the mid-point. In general, rating scales should be balanced, comprising an equal number of favourable and unfavourable response choices. However, scales may be balanced with or without an indifferent or mid-point option. A balanced scale might take the form of “very good, good, average, poor, very poor”. An unbalanced rating scale has an unequal number of favourable and unfavourable response choices. An example of an unbalanced scale that has only one unfavourable descriptive term and four favourable terms is “Fair- very good excellent”. The scale designer anticipates that the average ratings will cluster around the ‘good’ point and that responses will be distributed symmetrically around that mark. However, the scale does not provide a way for participants with unfavourable opinions to express the intensity of their attitudes.