Quantitative Research About Likert Scales

A scale is basically a continuing spectrum or series of categories and has been thought as any series of items which are organized progressively matching to value or magnitude, into which something can be located matching to its quantification.

Likert Scales at first being developed from 1946 to 1970 with a sociologist, Rensis Likert at the University of Michigan. It was being used initially as psychological behaviour that can be assessed from qualitative standpoints into quantitative perspectives on a proper metric scale. For example, inches wide or degree Celsius true measurement scales. It's the most widely used scale in survey research that interchangeably with score level even though they are not synonymous. Respondents will answer predicated on their degree of agreement to a particular statement. It had been a bipolar level running from one extreme via a neutral indicate the contrary extreme. Things are asked to give the feedback and response based on five-point, seven-point or ten-point level. It can evaluate arrangement or disagreement on specific claims or questions. In addition, it can be called as summative scales.

Bandura (1986) advised that the best way to measure self-efficacy is to evaluate both magnitude and durability. Before, authors such as Lee and Bobko (1994) assumed that Likert scales did not follow Bandura's suggestion. However, a Likert-scale way of measuring self-efficacy should offer an approximation to a traditional measure.

Based on Tittle 1967, The Likert Size is the hottest method of scaling in the social sciences today. Perhaps this is because they may be much better to build and because they tend to be more reliable than other scales with the same number of items. A genuine Likert range creates an individual scale from all of the items, then rescales each question according to the overall scale report for each response to each item.

Likert Scales vs Likert Items

The Likert Range is a sum of responses on several Likert Items. It can be illustrated by using horizontal series by which a topic indicates his / her response by circling or verifying tick-marks). In order to avoid any misunderstanding on both, it is better, to reserve the word Likert scale to apply straight to the summated size, and Likert item to make reference to an individual item.

On a survey or questionnaire, an average Likert item often takes the following format

Strongly disagree


Neither agree nor disagree


Strongly agree

The last average score signifies overall credit score of behaviour toward the topic matter. Exemplory case of Likert Size and Likert Items are as follow

Strongly Agree




Strongly Disagree

If the price of raw materials fell firms would reduce the price of their food products.






Without government regulation the companies would exploit the consumer.






Most food companies are so worried about making profits they do not service about quality.






The food industry spends a great deal of money making sure that its manufacturing is hygienic.






Food companies should fee the same price for their products throughout the country






Semantic scales

This type of scale makes extensive use of words alternatively than quantities. Respondents illustrate their thoughts about the merchandise or brands on scales with semantic labels. When bipolar adjectives are used at the end things of the scales, these are termed semantic differential scales. The semantic scale and the semantic differential scale are illustrated the following

Semantic and semantic differential scales

http://www. fao. org/docrep/W3241E/w3241e1d. gif

The bipolar adjectives used, for instance would make use of such conditions as Good-Bad, Strong-Weak, Hot-Cold. The semantic differential level is employed to evaluate respondents' behaviour toward a specific brand, advertisement, object or person. The replies can be plotted to secure a good idea with their perceptions. This is cured as an interval scale.

The Theory of Range Types

Stevens (1946, 1951) suggested that measurements can be grouped into four different kinds of scales. They are shown in the stand below as: nominal, ordinal, interval, and ratio.

Nominal Scale

A nominal level is the one that allows the researcher to assign subject matter to certain categories or teams. For instance, if we make reference to gender, the answer is one either male or female. These two teams can be assigned code numbers 1 and 2. These volumes provide as simple and convenient category product labels without intrinsic value, apart from to assign respondents to one of two non-overlapping or mutually exclusive categories. Remember that the categories are also collectively exhaustive. Quite simply, there is absolutely no third category into which respondents would normally fall season. Thus, nominal range categorizes individuals or things into mutually exclusive and collectively exhaustive organizations. The information that may be generated from nominal scaling is to determine the ratio (or occurrence) of men and women in our test of respondents. In addition, nominal scales give attention to only needing a respondent to provide some form of descriptor as the uncooked response. Nominal Level is often used for obtaining personal data such as gender or department where one works, where grouping of individuals or objects pays to.


Please indicate your current marital status.

__Married __ Solo __ Single, never committed __ Widowed

Ordinal Scale

An ordinal size not only categorizes the variables so concerning denote variations among the many categories, it also rank-orders the categories in some significant way. With any changing that the categories should be ordered according to some desire, the ordinal scale would be utilized. The preference would be placed (e. g from better to worst; first to last) and numbered 1, 2, and so on. It allows the respondent to express "relative magnitude" between the raw replies to a question. Ordinal Level is used to ranking the tastes or usage of varied brands of a product by individual also to list order individuals, objects, or happenings.


Which one statement best represents your opinion of Intel PC processor chip?

__ Higher than AMD's Personal computer processor

__ About the same as AMD's Computer processor

__ Lower than AMD's PC processor

The ordinal level helps the researcher to determine the percentage of respondents who consider connection with others because so many important, those who consider by using a quantity of different skills as most important, etc. For example, such knowledge might help in designing jobs that would be seen as most enriched by the majority of the employees.

As of now, we can see the difference between nominal and ordinal whereby ordinal can provide more info and depth point of view when compared with nominal. However, the ordinal does not give any indication of the magnitude of the differences among the rates. This deficiency is conquer by period scaling.

A problem with ordinal scales is that the difference between categories on the scale is hard to quantify. For example, excellent is preferable to good but how much is great better?

Interval Scale

An interval scale allows us to perform certain arithmetical businesses on the info collected from the respondents. Whereas the nominal level allows us only to qualitatively distinguish organizations by categorizing them into mutually exclusive and collectively exhaustive units and the ordinal range to rank-order the choices, the interval scale lets us measure the distance between any two points on the size. This can help us to compute the means and the typical deviations of the reactions on the variables. Quite simply, the interval scale not only organizations individuals corresponding to certain categories and taps the order of the groups; it also steps the magnitude of the distinctions in the preferences among the list of individuals. Quite simply, interval scales demonstrate the absolute differences between each range point. Interval Scale can be used when replies to various items that assess a variable can be tapped on a five-point (or seven-point or any other amount of points) range, which can thereafter be summated over the items.

Interval scales allow comparisons of the dissimilarities of magnitude (e. g. of behaviour) but don't allow determinations of the genuine power of the magnitude

Ratio Scale

The ratio scale overcomes the drawback of the arbitrary source point of the period scale, in that it has an absolute (as opposed to an arbitrary) zero point, which is significant dimension point. Thus, the proportion size not only measures the magnitude of the distinctions between details on the level but also taps the proportions in the variances. Ratio scales enable the identification of absolute differences between each size point, and absolute comparisons between fresh responses.


Please circle the amount of children under 18 years currently surviving in your home.

0 1 2 3 4 5 6 7 (if more than 7, please identify ___. )

Practically, Percentage Scales are usually found in organizational research when exact figures on objective (instead of subjective) factors are called for. Another good example is Johan scale of temperature. This scale has an absolute zero. Thus, a temperatures of 300 Johan is doubly high as a temperature of 150 Johan.


Type of Scale

Numerical Operation

Descriptive Statistics



Frequency in each category, ratio in each category, mode


Rank Ordering

Median, range, percentile ranking


Arithmetic Procedures on Intervals between numbers

Mean, standard deviation, variance


Arithmetic Businesses on real quantities

Geometric mean, coefficient of variation

Scale Type

Permissible Statistics

Admissible Level Transformation

Mathematical structure

nominal (also denoted as categorical)

Mode, Chi-Square

One to 1 (equality(=))

Standard set framework unordered


Median, Percentile

Monotonic increasing (order (<))

Totally purchased set


Mean, standard deviation, relationship, regression, evaluation of variance

Positive linear (affine)

Affine line


All statistics permitted for interval scales in addition to the following: geometric mean, harmonic mean, coeeficient of variance, logarithms

Positive similarities (multiplication)


These constitute a hierarchy where in fact the lowest range of way of measuring, nominal, has far fewer numerical properties than those further up this hierarchy of scales. Nominal scales give data on categories; ordinal scales give sequences; period scales commence to uncover the magnitude between items on the scale and proportion scales clarify both order and the overall distance between any two details on the level.


The reliability of an measure shows the scope to which it is without bias (error free) and hence ensures consistent measurement across time and across the various items in the device. Quite simply, the reliability of a measure is an indication of the stableness and consistency with that your instrument steps the concepts and helps to access the goodness of the measure.

Stability of Measures

The ability of your measure to remain the same as time passes - despite uncontrollable testing conditions or the talk about of the respondents themselves - is indicative of its balance and low vulnerability to improve in the problem. This attests to its goodness because the idea is stably measured, no matter when it's done. Two sets of stableness are test-retest trustworthiness and parallel-form dependability.

In test-retest trustworthiness, the respondents will be analyzed at two differing times using the same sets of scale item to be able to look for the degree of similarity of the two measurements.

In alternative-forms consistency, the same scale is built and respondents being measured at two different times whereby every time using different form.

Internal consistency stability determines the extent to which different parts of a summated scale are constant in what they suggest about the attribute being assessed.

The Cronbach's alpha or coefficient alpha, calculating the internal uniformity which determines the amount of correlation between a couple of items as an organization. It isn't a statistical test but a coefficient of dependability or persistence.


Validity is the power of a level or measuring device to evaluate what it is intended to evaluate (e. g. is absenteeism from work a valid way of measuring job satisfaction or is there other influences like a flu epidemic which is keeping employees from work).

Content validity is a subjective but it being used to ascertain on content of level that match measurement process.

Criterion validity reflects whether a scale performs as expected with regards to other variables chosen (criterion factors) as important criteria.

Construct validity addresses the question of what characteristic the range is - measuring. Construct validity includes nomological, discriminant and convergent validity.

Convergent validity is the analysis to which the scale correlates positively with other options of the same build.

Discriminant validity is the magnitude to which a strategy will not correlate with other constructs from which it is supposed to vary.

Nomological validity is the extent to that your scale correlates in theoretically forecasted ways with methods of different but related constructs.

Also We Can Offer!

Other services that we offer

If you don’t see the necessary subject, paper type, or topic in our list of available services and examples, don’t worry! We have a number of other academic disciplines to suit the needs of anyone who visits this website looking for help.

How to ...

We made your life easier with putting together a big number of articles and guidelines on how to plan and write different types of assignments (Essay, Research Paper, Dissertation etc)