Question

is criterion because it uses a category l40 Catholics, Jews, Buddhists, Atheists, and so forth) that terms of s may De p for some p include broad for meaningful research Scale D in Table 1.3 is probably the most common way preference has been measured in North America, and this scale w reli re ate in many situations. However, it may be too general for so for legal abortion would probably have to distinguish among the var Protestant denominations, and an effort to document religious div certainly would need to add categories for Buddhists, Muslims, and nu rposes. For example, an investigation of moral questions such as su diversit other religious faiths. inal Level of Measurement Variables measured at the ordinal level are more sophisticated than nominal-levelT variables. They have scores or categories that can be ranked from high to low, s0 in addition to classifying cases into categories, we can describe the categories in terms of "more or less" with respect to each other. Thus, with ordinal-level vari- ables, not only can we say that one case is different from another; we can also say that one case is higher or lower, more or less than another. For example, the variable socioeconomic status is usually measured at the ordinal level, often using categories such as those in Table 1.4. Individual c can be compared in terms of the categories into which they are classified. Thus, an individual classified as a 4 (upper class) would be ranked higher than an indi- vidual classified as a 2 (working class), and a lower-class person (1) would rank lower than a middle-class person (3). Other variables that are usually measured at the ordinal level include attitude and opinion scales, such as those that measure prejudice, alienation, or political conservatism. ases TABLE 1.4 Measuring Socioeconomic Status If you were asked to describe yourself, into which of the folowing classes would you say you belong? Score Class Lower class Working class Middle class Upper class

Answer 1

Ordinal Data and Analysis  

Ordinal scale data can be presented in tabular or graphical formats for a researcher to conduct a convenient analysis of collected data. Also, methods such as the Mann-Whitney U test and Kruskal–Wallis H test can also be used to analyze ordinal data. These methods are generally implemented to compare two or more ordinal groups.

In the Mann-Whitney U test, researchers can conclude which variable of one group is bigger or smaller than another variable of a randomly selected group. While in the Kruskal–Wallis H test, researchers can analyze whether two or more ordinal groups have the same median or not.

Nominal Level of Measurement: In nominal level of measurement, variables are differentiated by their namings. These variables have no order or hierarchy associated with them.

Numbers associated with the names are mere tags with no mathematical aspect linked to them. These variables are descriptive in nature. In terms of statistics, nominal scale is the easiest to understand and implement. These variables have minimum two divisions such as Male/Female, Yes/No.

This scale has no numerical value, for example – gender, ethnicity, race etc.

Ordinal Level of Measurement: In ordinal level of measurement, the order of variables is critical. The difference between these variables is not established and is not an integral aspect of this measurement scale.

The variables are identified and described along with allotting a value to each of these identified variables. In market research, ordinal scales are used to analyze relative perceptions, choices, and feedbacks, i.e., marketers can evaluate the degree of customer satisfaction or happiness, understand whether their newsletters should go out more often, etc.

Factors	Nominal Scale	Ordinal Scale
Description	The variables of this scale are differentiated by their nomenclature and none other factors. There is no implied sequence in which variables exist in nominal scale	These variables have a naturally occurring order present between them yet the difference between variables is unknown. The value of difference between two variables on this scale cannot be calculated. For instance, the order of size is small, medium, large, extra large. But Small – Medium ≠ Large – Extra Large.
Degree of Quantitative Value	There is no quantitative value associated with variables on this scale. Instead, it is a qualitative measurement scale.	Quantitative values are linked to ordinal variables but arithmetic evaluation cannot be conducted on these variables.
Key Differentiators	These variables cannot be ordered. The variables of this scale are distinct. Nominal data is not quantifiable.	Numbers are assigned to the variables of this scale. No arithmetic calculation can be done on these variables. The difference between variables cannot be calculated.
Examples	Sex (Male, Female) Marital Status (Married, Divorced, Unmarried, Widowed etc.) Religion (Christian, Jew, Muslim) Race (Red Indian, South-east Asian etc.)	Rank in a class test (first, second or third) Customer satisfaction ratings (On a scale of 0-10) Socio-economic status Customer satisfaction degrees (Very satisfied, satisfied, neutral, dissatisfied, very dissatisfied) Education qualification