- Measurement is a process of describing some issue that presents the interest. The “object” of interest is usually assigned with the number or letter. The numbers or letters are usually assigned with the information about the property of the measured “object” or “issue”. Thus, the learners’ performance in a research class can be measured in different ways. For example, it can be measured with a letter signed to describe his/her performance or a number that will be correspondent to a percentage scale. The scales are used to define the differences in students’ performance.
- “Before the measurement process can be initiated, the business research must identify the concept relevant to the problem (Zikmund, 2003, p. 294). Concept is an abstraction and presents considerable problems for measurement. Construct is even more abstract than a concept. It does not provide a connection between the abstract and observed manifestation. In other words, it is based on the concept.
- If a researcher takes over a project after the proposal has been written by another researcher, he/she should find the things that require to be measured. The researcher can usually find them in the research question or hypothesis, as they usually define concepts and relations between them.
- There are different grading measuring scales and each represents a definite level of measurement. Nominal scale (the simplest): it assigns the categorizing events or characteristics with numbers. The scale does not allow determining an average. Ordinal scale: all the categories included in the scale have relationships with each other. It allows more flexibility in determining definite categories, “Likert scales are commonly used in attitudinal measurements” (“Measurements and Scaling”, n. d., n. p.). Interval scale allows measuring and indicating order using the units with equal intervals (“Measurements and Scaling Concepts”, n. d. p. 9). Ration scale: it has a meaningful zero point and is often used for gathering quantitative information (“Measurements and Scaling”, n. d., n. p.).
- Considering the differences between the four measuring scales, we can conclude that the ratio scale incorporates the characteristics of the previously describes scales and presents the least opportunities for error, and provides the opportunities for descriptive calculations.
- Based on a summated stress score the stress score will be 26, and the average composite scale score will be 6.5. If the 5th item will be added to the scale, in such case, the reverse-coding will be necessary, otherwise, it is not needed.
- “An attitude is as an enduring disposition to respond consistently to specific aspects of the world, including actions, people, or objects” (Zikmund, 2003, p. 289). In fact, there is no consensus concerning its definition as the components of the “attitude” are hypothetical and can vary depending on the situation or the “issue” that should e measured, thus, “To measure attitudes only the overt symbolic type of acts are taken into account because such acts alone can be observed” (“Measurement of Attitude”, 2011, n. p.)
- Rating is a commonly used attitude measurement and its technique is better than that of a ranking. It takes less time, it is not so difficult to answer when the choices are added, and does not provide limits for statistical analysis.
- Using statistical software like SAS or SPSS provides the opportunity to perform reverse recording simply using “old” and “new” scale values. There is a formula to produce this computing procedure which included New Value (NV), Scale Minimum (SMin) and Maximum (SMax) and Old Value (OV): NV=(SMin=SMax)-OV.
- Numeric Scale uses numbers to identify response position as opposed to semantic differential scales. It is more advantageous as it does not provide interval data and does not involve people’s reaction to a stimulus caused by the connotations of the word and in terms of results, “researchers have found that a scale with numerical labels for intermediate points on the scale is as effective a measure as the true semantic differential” (Zikmund, 2003, p. 315).
- There are several issues a researcher should consider when choosing a measurement scale. First of all, he/she should decide which technique fits the research the best: ranking, sorting, or rating. Next, the researcher should choose a single or index measure, furthermore, decide the number of scales to be needed and define the category of labels, should the balance, unbalance, or comparative scale be applied.
- The Likert Scale can be treated as ordinal or interval because it uses at least 5 levels and there can be equal distances between every variable included in the scale. It all depends on the manner of construction and measured issue.
- Types of scale:
- Comparative rating scale. Often used to compare an ideal situation with the actual one.
- An Unbalanced Scale. It is very specific and based on the unequal number of positive and negative categories. (Zikmund, 2003, p. 330)
- A Likert scale. It provides the opportunity for response alternatives and respondents can identify their attitudes to the issue.
- Consumers are grouped according to sociological and psychological characteristics which help to define their purchase decisions. There are several types: innovator, thinker, achiever, striver, believer, maker, survivor, and experience. I believe that I am an experience as I am motivated by self-expression.
“Measurement of Attitude” (2011). Sociology Guide. Web.
We will write a custom Research Paper on Measurements and Scaling Concepts specifically for you
301 certified writers online
“Measurements and Scaling Concepts” (n. d.) Web.
“Measurements and Scaling” (n. d.). Web.
Zikmund, W. G. (2003). Business Research Methods. New York: Thomson/South-Western.