In this write up, the different scales of measurement, nominal scale, ordinal scale, interval and ratio are discussed, including examples of test types that would usually employ them. Also, measures of central tendency, and measures of variability and their effect on test suitability are addressed in the second half of this piece. The Four Scales of Measurement Scales of measurement defines a framework wherein numbers are assigned to objects based on a set of rules (Thorndike & Thorndike-Christ, 2009). These authors presented four different types of scales of measurement as ordinal, nominal, ratio and interval. Nominal Scale Thorndike Thorndike-Christ (2009) described a nominal scale as a measurement type wherein each “number takes on the meaning of a verbal label” (p. 27). For example, in an assessment of students on a campus, girls could be labeled number 1, and the girls, number 2. In this case, the values are mutually exclusive, and the labels have no numerical significance. Assessment that would use a nominal scale could be practically any measure that affixes labels to constructs for assessments. For example, the Interprofessional Competency Model for Healthcare Leadership is an assessment model that measures constructs developed into competency models, such as analytical thinking, community orientation, financial skills, and initiative (Calhoun, Dollett, Sinioris, Wainio, Butler, Griffith, & Warden, 2008). Ordinal Scale Next, an ordinal scale describes “the order
2. Based on the scale of measurement for each variable listed below, which measure of central tendency is most appropriate for describing the data?
al.,2007). Using previously researched scholar articles and books, the authors were able to base their search, follow certain guidelines and compare their results with other results. Using tests such as the Kruskall-Wallis non-parametric test, Nagel et. al.(2007) were able to examine the differences in performance based on each grade group.
2. Based on the scale of measurement for each variable listed below, which measure of central tendency is most appropriate for describing the data?
Nominal data is the most basic level of measurement. It is also known as categorical. The numbers do not imply an order. Basically nominal data is used for frequency and the only number property of the nominal scale of measurement is identity. An everyday example of the use of nominal data would be classifying people according to gender is a common application of the nominal scale. When you first meet someone, an observation is generally made on the specific gender of the person you are meeting for the first time.
To view the research on a nominal scale, the research data can be drawn from the type of class. The word nominal is derived from the root word in Latin for name (Usable Stats, 2013). The name of the class, Psychological Statistics, is the nominal measurement for this research. When conducting this study, the study will only be measured during the course of this specific class. The results could drastically change when considering another type of class such as Quantitative Literacy as the cognitive understanding of such a collegic math class may be more optimal through a different course-delivery format.
32 A standardization sample is representative if the sample a has been subjected to rigorous experimental control b consists of individuals that are similar to the group to be tested c consists a great many individuals d is administered in the same way as the actual test group will be 33 When a test is administered to the general population, norms should be established using a representative sample that a has been administered the test under standard conditions b has been chosen in a completely random fashion c represents all segments of the population in proportion to their numbers d is comprised of a great many individuals 34 Administering a test with precisely the same instructions and format is called a normative conditions b standard conditions c facilitative conditions d group administration 35 Dr Johnson is trying to establish norms for his new test He determined that 50% of the people in the standardization sample should be Hispanic, 20% Caucasian, 15% Asian, and 15% African American He is creating a a normalization group b representative sample c random sample d population statistics 36 The Stanford-Binet intelligence scale was developed by a A Binet b T Simon c A Binet and T Simon d L M Terman 37 The concept of mental age was introduced in A 1900 b 1908 c 1911 d 1916
When multiple people are given assessments of some kind or are the subjects of some test, then similar people under the same circumstances should lead to scores that are similar or duplicates ("Types Of Reliability", 2011). This is the idea of inter-rater reliability. Another mode of reliability is the administration of the same test among different participants and expecting the same or similar results ("Types Of Reliability", 2011). This is known as Test-retest reliability. This method of measurement might be used to make determinations about the effectiveness of a school exam or personality test ("Types Of Reliability", 2011). Surveys and other methods of research present the appropriate avenues for data collection.
To test the above-mentioned variables Team A will need to select specific measurement scales for them. The four types of measurement scales are nominal, ordinal, interval, and ratio. The depth of sophistication increases as one moves from the nominal scale to the ratio scale. In other words, the ratio and the interval scales offer more detailed information on variables than do the ordinal and the nominal scales. That is to say, as one moves from the nominal scale to the ratio scale, one moves from the general to the more specific, and the more specific information a scale yields, the more powerful it is. The more powerful scales, provide the possibility of performing in-depth, meaningful analysis so that more
Assessment (an estimate of activity rated on a 0 to 3 visual analog scale), the most common
| Based on explicit knowledge and this can be easy and fast to capture and analyse.Results can be generalised to larger populationsCan be repeated – therefore good test re-test reliability and validityStatistical analyses and interpretation are
Multiple assessment techniques were utilized to assess variable data and compare the reliability of that data for each individual variable.
Multiple transformation were attempted to normalize the data, however, none were found to be normal as confirmed through Shapiro-Wilk tests. Therefore, the raw data was used for all further analysis. While some response values did appear to be far off the median none were considered to be outliers and no data points were omitted. Considering that the data was not normally distributed, the median and interquartile range (IQR) were used to report measures of center and spread respectably. The median (center) was found to be 94 (%) while the IQR was reported as 9 which indicates how spread out the middle values are, thus describing the
4. What measurement scales would you have used on the survey that was part of the in-restaurant product tests?
The objective of this chapter is to describe the procedures used in the analysis of the data and present the main findings. It also presents the different tests performed to help choose the appropriate model for the study. The chapter concludes by providing thorough statistical interpretation of the findings.
The above table 5.1 enumerates that the communalities of the selected 36 variables have good reliability and keenly checked that no one variable has low loading, i.e., less than 0.5. Thus finally, the 26 variables are selected for the factor analysis. The appropriateness of the data for the factor analysis is discussed in the following KMO and Bartletts’ test.