Understanding the Scaling in Research
- 1. By Dr. Shaloo Saini Assistant Professor, MKCE (CT Group of Institutions, Jalandhar, Punjab)
- 2. Scaling Scaling is the extension of the concept of measurement because the result of measurement is a scale which comprises of a set of numerals on which an object’s score is placed using a certain rule of assignment. Scaling refers to the procedure for attempting to determine the quantitative measures of subjective abstract concepts.
- 3. Scaling In Scaling numbers are assigned to various degrees of opinions, attitude and other concepts through two ways: i. By making a judgement about some characteristics of an individual and then placing him directly on a scale that has been defined in terms of that characteristics. ii. By constructing questionnaires in such a way that the score of individual’s responses assigns him a place on the scale. A Scale is a continuum where one end indicates the lowest level of the phenomena and the other end indicates the highest level.
- 4. Definition of Scaling According to Allen L Edwards “ Scaling is the procedure for assignment of numbers or other symbols to a property of objects in order to impart some of the characteristics of numbers to the properties in question.”
- 5. Scale Classification BasesThe Scales can be classified on the basis of six key decision areas: 1. Objective of the Study 2. Nature of Response 3. Degree of Subjectivity 4. Number of Dimensions 5. Scale Properties 6. Scale Construction Techniques
- 6. 1. Objective of the Study A Scale may be designed with the objective of measuring the characteristics of the respondent who completes it. For example scale design on an Organisation’s development programs. The respondent’s score may be used as an indicator of respondent’s favourable for unfavourable attitude towards the organisation. In this situation emphasis is on measuring the attitudinal difference amongst the respondents. A Scale may also be designed to use respondents as judges of the object or stimuli presented to them. Using the above example again, the respondents can be asked to score the development programme started by the Organisation. In this the objective is to measure the difference among the stimuli that is the development programme started by the Organisation.
- 7. 2. Nature of Response The scale can be classified as ‘Categorical’ or ‘Rating Scale’ and ‘Comparative or Ranking Scale.” Categorical scale obtained the responses from a respondent by asking him to score and object or attitude without any direct reference to other objects For example they may be asked to rate the taste of a newly launched soft drink on a 5 point rating scale. Comparative scales ask respondents to compare two or more objects for example they may be asked to rank two soft drinks brands on their taste giving I to the drink that taste better than the other.
- 8. 3. Degree of Subjectivity Scale may be developed on the basis of whether we measure subjective or personal preference or we measure known preference evaluation. In case of ‘Preference’ measurement the respondent is asked to choose the object solution or person that he personally favours. Whereas in ‘Non Preference’ measurement looks for an objective evaluation by the respondent. He is asked to evaluate which objective has more of some characteristics, which solution optimises the resources or which person performs the best without reflecting any personal preference towards object or solution.
- 9. 4. Number of Dimensions Scale can be classified as ‘Unidimensional’ or ‘Multidimensional’. In Case of Unidimensional scaling only one attribute or attitude of the respondent object is measured For example a teacher’s professional competence maybe judged on work experience. Several items may be used to measure his work experience and by combining all these scores into a single measure the teacher may be placed on a linear scale of work experience. In case of Multidimensional scaling the respondent or object is measured in an attribute space of dimensions For example the work experience variable may be better expressed by three distinct dimensions nature of work, place of work and length of work experience.
- 10. 5.Scale Properties The Scale can also be classified on the basis of data properties possessed by each scale. They may be classified as a Nominal Scale: which is used for just classification purpose Ordinal Scale: gives the relationship of greater than or less than but no distance or origin Interval Scale: like temperature that possesses order and distance but no unique origin, and Ratio Scale: that may possess order distance and unique origin.
- 11. 6.Scale Construction Techniques Arbitrary Approach: The scale is developed on an Ad Hoc basis and is designed largely on the researchers own subjective selection of items it is most popular scale. Consensus Approach: As against are searches subjective selection third uses a panel of judges who evaluate each item on its suitability for inclusion into the scale. Those items that find consensus among the judges form a scale. Item Analysis Approach: A set of individual items is developed for a pilot test and given to a group of respondents. Total scores of each respondent are evaluated and then individual items are analysed to determine which have the ability to discriminate between respondents with high total scores and low total scores Cumulative Scales: These scales are formed on the basis of their confirmatory to a ranking of items that is associated with ascending or descending power. A person who agrees with an item placed on the extreme end indicates that he is in agreement with all the items on less extreme Factor Scales: These scales use certain factors to establish relationship between items the correlation between the items is accounted to the presence of common factor between them.
- 12. Scaling Techniques Scaling Techniques Comparative Scaling Techniques Paired Comparison Rank Order Constant Sum Non Comparative Scaling Techniques Continuous Rating or Graphic Rating Itemized Rating Likert Scale Semantic Differential Scale Stapel Scale Simple/Multiple Category Scale Verbal Frequency Scale
- 13. Comparative Scaling Techniques Comparative Scales involve direct comparison of stimulus objects. The data from Comparative Scale is interpreted in relative terms and are measured on ordinal scale. This technique is a non-numeric scaling technique as Ordinal data cannot be used for numeric operations. These techniques are easy to understand and apply. The respondent is forced to choose between the stimulus objects for example respondent are asked whether toothpaste of Brand A or brand B. They have to choose one of the brand even if there is very small difference in their liking of the two brands. The main disadvantage of comparative scale is its inability to generalize beyond the stimulus objects for example if we want to compare third brand of toothpaste C with the previous once we have to conduct a new study.
- 14. 1.Paired Comparison In this method the respondent can express his attitude by making a choice between two objects for example between flavour of a new soft drink and an established brand of soft drink. When the comparison is between more than two objects, there more than two judgements need to be expressed. Therefore the number of judgements required in a paired comparison is given by the formula: N= n (n-1)/2 where N= Number of Judgements and n= number of objects to be judged. For example if there are 10 suggestions for bargaining proposals available to a workers union there are 45 paired comparisons that can be made with them. Where N happens to be a big figure there is risk of respondents giving ill considered answers or they may even refuse to answer. Paired Comparison provides ordinal data but can be converted into interval scale by the method of ‘Law of Comparative Judgement’ developed by L.L. Thurstone and ‘Composite Standard Method’ given by J.P. Guilford.
- 15. 2. Rank Order Under this method of Comparative Scaling the respondents are asked to rank their choices. This method is easier and faster than the method of Paired Comparison For example with 10 items it takes 45 paired comparisons to complete the task whereas Rank Order method simply requires ranking of 10 items only. The problem of Transitivity (such as A prefers to B, B to C but C prefers to A) is also not there in case of method of Rank Order. Moreover a complete ranking at times is not needed in which case the respondent may be asked to rank only their first choice out of all the choices in the overall items.
- 16. 3. Constant Sum Constant sum scaling technique is used to assess the relative importance attached by a respondent to the stimulus objects. The respondent gives certain points to each stimulus object out of a fixed sum of points. This fixed sum is usually taken as 100 but it could be some other value also. For example a family planning a holiday fixes its budget of Rs 50000. They wish to plan the expenditure on Transport, Accommodation, Refreshment and others the constant sum of this expenditure is 50000 which could be divided as: Transportation= 10000 Accommodation= 20000 Refreshment= 15000 Others= 5000 The data obtained here is numeric but cannot be generalized beyond the list of stimulus objects. So this data is considered as ordinal data. This technique distinguishes the objects in less time. However it is not useful with uneducated people or with large number of objects.
- 17. Non Comparative Scaling Techniques In Non Comparative Scales each object is scaled independently of the other. For example the respondents are asked to give preference scores on a 1-6 scale to Brand A of the toothpaste. Here 1= not preferred at all and 6= highly preferred. Similar scores can be obtained for brand B and C. Due to the facility of the numeric data and wide applications the Non Comparative Scales are widely used in research.
- 18. 1. Continuous Rating or Graphic Rating Continuous Rating or Graphic Rating: The graphic rating scale is quite simple and is commonly used in practice. Under it the various points are usually put along the line to form a Continuum and the rater indicates his rating by simply making a tick mark at the appropriate point on a line that runs from one extreme to the other. Scale points with brief descriptions may be indicated along the line their function being to assist the rater in performing his job. For example we can use five points graphic rating scale when we wish to ascertain people's liking or disliking on any product: How do you like the product? (Put a tick mark) This type of scale has several limitations. The respondents may check at almost any position along the line which in fact may increase the difficulty of analysis. The meanings of the terms like ‘Very Much’ and ‘Some What’ may depend upon respondent’s frame of reference so much so that the statement might be challenged in terms of its equivalency. Like Very Much Like Some What Neutral Dislike Some What Dislike Very Much
- 19. 2. Itemized Rating Itemized Rating: Scale having number or brief description of each category is provided. Categories are ordered in terms of scale positions. The respondents select one of the categories that best describes the stimulus object. The commonly used itemized rating scales are: A-Likert Scale B-Semantic Differential Scale C- Stapel Scale
- 20. A. Likert Scale Likert Scales or Summated Scales are developed by utilising the item analysis approach where in a particular item is evaluated on the basis of how well it discriminate between those persons whose total score is high and those whose total score is low. Those items or statements that best meet this sort of discrimination test are included in the final instrument. Likert Scales are the Summated Scale that are frequently used in the study of social attitudes this contains number of statements which express either favourable or unfavourable attitude towards the given object to which the respondent is asked to react. The respondent indicates his agreement or disagreement with each statement in the instrument. Each respondent is given a numerical score and the score for each item is summated to measure the respondent’s attitude, in other words the overall scores represents the respondents position on the Continuum of favourable or unfavourableness towards an issue.
- 21. B. Semantic Differential Scale The SD scale was developed by Charles E. Osgood, G.J. Suci and Tannenbaum (1957). This scale measures the psychological meaning of an object to an individual. This scale is based on the presumption that an object can have different dimensions of connotative meanings which can be located in multidimensional property space, or what can be called the semantic space in the context of S.D. scale. This scaling consists of a set of bipolar rating scales, usually of 7 points by which one or more respondents rate one or more concept on each scale item. For example the S.D. Scale items for analysing candidates for leadership positions candidates for leadership position may be compared and may be scored from +3 to-3 for each parameter:
- 22. C. Stapel Scale Stapel Scale: This scale was developed by John Stapel. This is a unipolar rating scale which usually 10 categories numbered from – 5 to+5. This scale does not have the Zero or the neutral point. The respondents rate how each term describes the object by selecting the appropriate number. Positive number means the term describe the object accurately, while the negative number implies that the term describes the object inaccurately. +5 means the highest degree of accuracy while-5 means highest degree of inaccuracy. For example respondent may be asked to rate how accurately the terms (i) tasty food (ii) fast service and (iii) good ambience describe the quality of the restaurant. Method is applicable when the responses are rated on a single dimension. The method is very economical and data can be collected over telephonic interview also. However some researchers think that this method is confusing and is of not much use.
- 23. 3.Single/Multiple Category Scale The scales are known as dichotomous scales. These have two or more mutually exclusive responses. Like yes and no /true and false to choose of the given categories. For example Do you like to play cricket? Yes or No; Which vehicle do you have? Car, Bike, Scooty or any other. The collected data is on nominal scale. This method is very easy and very popular on internet survey.
- 24. 4. Verbal Frequency Scale This scale is used when the respondent is unable or unwilling to give the exact numbers in the answer. For example: How often do you eat out? Frequently/ Sometimes/ Rarely/ Never. This Scale provides only an approximation of frequency and so the data is on ordinal scale.
- 25. Summary Scaling refers to the procedure for attempting to determine the quantitative measures of subjective abstract concepts. According to Allen L. Edwards “Scaling is the procedure for assignment of numbers or other symbols to a property of objects in order to impart some of the characteristics of numbers to the properties in question.” The bases for the scaling procedures or the number assigning procedures can be broadly classified as objective of the study, Nature of Response, Degree of Subjectivity, Number of Dimensions, Scale Properties and Scale Construction Techniques. The Scaling Techniques are broadly classified into Comparative Scaling Techniques and Non Comparative Scaling Techniques. Comparative Scales involve direct comparison of stimulus objects and in Non Comparative Scales each object is scaled independently of the other. The Comparative Scaling Techniques are Paired Comparison, Rank Order and Constant Sum. The Non Comparative Scaling Techniques are Continuous or Graphic Rating Scales, Itemized Rating, Simple/Multiple Category Scale and Verbal Frequency Scale.
- 26. References Bajpai N. (2015). Measurement and Scaling. In Business research methods (pp. 43). Nodia: Pearson Education. Cooper, D. R., Schindler, P. S., & Sharma, J. K. (2012). Measurement scales. In Business Research Methods (8th ed., pp. 341). New Delhi: Mc. Graw Hill Education (India). Edwards, Allen L.(1957) Techniques of Attitude Scale Construction, Appelton- Century Crofts, New York. Gupta S.K. & Rangi P.(2017). Measurement and Scaling. In Research Methodology (4th ed., pp4.20).Punjab: Kalyani Publishers(India). Kothari C. R. & Garg G.(2019). Measurement and Scaling. In Research Methodology Methods and Techniques( 4th ed., pp.66-87) New Delhi: New Age International Publishers(India). Torgerson W. S. (1958). Theory and Methods of Scaling, New York: John Wiley and Sons Inc.
- 27. Thank You This Power Point Presentation has been made while referring to the research books written by eminent, renowned and expert authors as mentioned in the references section. The purpose of this Presentation is to help the research students in developing an insight about the Scaling in Research. I hope the students will find this presentation useful for them. All the Best Dr. Shaloo Saini