Definition of Measurement
From a simplistic perspective, measurement can be defined as the assignment of numbers to the parameters of an occurrence, commodity, or events among other things. On the basis of a critical view, however, measurement cannot be restricted to the assignment of real numbers to an object since there are measurements that involve qualitative assessment rather than quantitative measurement. It is, therefore, critically correct that measurement is a process incorporating observation, recording and quantification of a parameter including performance, objects and occurrences (Zikmund, 2003).
During class researches, the instructors need the students to investigate on specified fields and write a research paper reporting the findings. This research paper provides the basis of measuring the performance that is assessed through various parameters including language, organization of ideas, scope of research and clarity among others. A maximum mark is assigned to the individual parameter forming the scale within which the marks are granted. When the instructor assigns the marks, they are summed up to give the overall marks reflecting performance.
Concepts and Constructs
Although we seek to identify the potential difference between a concept and a construct, there is a thin line between them since they are both abstract. However, a concept is an attribute involved in the organization of abstract thoughts into classifications. For example, the security agents are classified according to the expectations of performance and authority in a logical manner ensuring that they have a classified hierarchy of law enforcement.
On the other hand, constructs are abstract pictures that are formed mentally to facilitate the formation of learning and teaching bases. For example, language is a construct that is represented symbolically, yet it needs explanation in order to understand within the same context. It is, therefore, clear that concepts are logical while constructs are mental. Also, the above description shows that a construct involves actual cases while concepts incorporate the actual and possible cases.
In the actual research, the researcher is needed to measure various variables that are contained in the proposal to obtain the data required for analysis. In a research proposal, these variables should be included in the data collection part of the proposal. This implies that the researcher should obtain these variables from that part in order to facilitate measurement. In some case, these variables are embedded on the questionnaires involved in data collection.
Levels of Scale Measurement
Scale measurement has 4 levels distinguished in the following sentences (Holl, 2012). Nominal scale is used to represent qualitative data by placing people into defined categories that can be compared in enhance analysis. For example, a nominal scale can categorize people into men and women. Similarly, the ordinal scale places data into categories that can be ranked in an ascending or descending order (Waltz, Strickland & Lenz, 2005).
Ordinal scale is, therefore, more helpful than nominal scale in terms of data analysis since it shows the relationship between data points. Interval scale exhibits the characteristics of the ordinal and nominal scale since it places data into categories while showing the relationship of data sets to each other. However, it shows the distance between the data point such that the analyzer can identify the distance one point sets to another. The ratio scale incorporates all the characteristics of the three scales where data is placed into categories that are ranked increasingly (Time & Payne, 2008). However, it has an absolute point acting as the reference point for all measurements.
Grading Measuring Scale
The first grading system follows the ordinal scale where the grades are categorized into A, B, C, D, E and F, but they have an attribute of hierarchy. The second grading system uses the interval scale showing the distance between each grade to another. The third system exhibits a ratio scale since the intervals start from zero. Lastly, the fourth system is a nominal scale that categorizes the grade into two ignoring the intervals (Roberts, 2008).
- On the basis of summated ratings, the score would be 8 since it indicates the highest level of stress incurred
- The composite score would be the summation of the scores adding up to 26.
- None of the three statements require reverse coding since they are all stated positively.
Attitude refers to the response, which can be negative or positive, towards an object, event or an occurrence showing disfavor or favor.
Rating and Ranking
Rating and ranking are used in measuring the human attitude towards objects, commodities and people. Rating seeks to determine the magnitude of the response that is exhibited by the person towards an object, event or person. On the other hand, ranking seeks to classify the objects in the order of preference as portrayed by the person being tested (Petraczi & Aidman, 2009).
Reverse Coding in SPSS
Mental reverse coding might lead to an error since the analyzer might make a mistake in the process. This makes it important to use software like SPSS. While using SPSS, recoding is done by entering the data in the normal way looking like the initial data. Click on transform in order to have two options that allow recoding for different and same variables alongside storing the reversed codes as extra variables.
Numerical Scales over Semantic
There are various advantages of numerical scales over the semantic differential scales (Zikmund, 2003). First, numerical scales are easy to interpret as opposed to semantic scales whose interpretation can be affected by the opinions of the analyzer. Unlike semantic differential scale, the results obtained from numerical scales can fit into a direct decision that makes it easy to make conclusions. Lastly, the numerical scales provide results that are more comparable than the semantic scales (Miller & Davis, 2008).
While choosing the type of measurement scale, the researcher must consider the nature of data that are collected from the field. Quantitative data requires numerical scales to provide quantitative results for analysis and interpretation. Secondly, the researcher should consider the interpretability of the results that are obtained due to the chosen scale in accordance to the research variables. These are the most crucial factors since they affect the processes of data collection, analysis and interpretation.
Absolutely, a Likert scale possesses distinct ordinal properties. Ordinal scale possesses an attribute of order and rank without identifiable level of separation between items. For instance, there could be a grading system identifying students who perform well with a letter ‘A’. On the other hand, poorest performing students are identified with a letter ‘E’. Assuming that the remaining classes of students are ‘B’, ‘C’ and ‘D’, the distance separating these categories is not evident. Similarly, a Likert scale cannot indicate the distance between the levels of acceptance as described in the following statement. Collected data could indicate the level of acceptance to a provided question (Malecukov, 2008). Probably, there could be choice such as ‘strongly accept’, ‘accept’ and ‘do not accept’. It cannot be identified how much acceptance lies between the choices.
Types of Scales
- Simple attitude scales call for an outright agreement or disagreement on a stated case. The 2 target choices could be accompanied by a choice for those who do not side. The question provides a room for two possible answers. The interviewee either opposes or favors the proposal (Lockhart, 2012). Otherwise, the third answer provides a base to disregard the interviewee response.
- The question has provided more than two choices that categorize the level of favorability. Additionally, the question gathers more information about the favor exhibited by the people (Holl, 2012). Lastly, the question is phrased properly to determine the differences in personal favors.
- This question exemplifies the contents of questionnaires that measure the negativity and positivity of people towards given attributes. In addition, the five choices approve this question to exemplify Likert scale. Finally, the question determines the strength of agreement like other Likert scale questions (Herod, 2011).
Likert scale contains positive and negative statements to cover answers from positivity to negativity. If statements are all positive, they would not satisfy the attribute that show a range from positive to negative (Feldt, 2008). Consequently, it can be stated that Likert scale could contain positive and negative statements. However, single sided statement cannot exemplify Likert scale.
Bulsara, C., Styles, I., Ward, A., & Bulsara, M. (2006). The Psychometrics of Developing the Patient Empowerment Scale. Journal of Psychosocial Oncology, 24(2), 1-16.
Feldt, R. C. (2008). Development of a Brief Measure of College Stress: The College Student Stress Scale. Psychological Reports, 102(3), 855-860.
Herod, A. (2011). Scale. London: Routledge.
Holl, S. (2012). Scale. Zurich: Lars MuÌller Publishers.
Lockhart, P. (2012). Measurement. Cambridge, Mass.: Harvard University Press.
Malecukov, A. L. (2008). Scales. Leipzig: Inst. fuÌr Linguistik, University.
Miller, J., & Davis, R. C. (2008). Unpacking Public Attitudes to the Police: Contrasting Perceptions of Misconduct with Traditional Measures of Satisfaction. International Journal of Police Science and Management , 10(1), 9-22.
Petraczi, A., & Aidman, E. (2009). Measuring Explicit Attitude toward Doping: Review of the Psychometric Properties of the Performance Enhancement Attitude Scale. Psychology of Sport and Exercise, 10(3), 390-396.
Roberts, C. (2008). Attitude measurement. Los Angeles: SAGE.
Time, V., & Payne, B. (2008). School Violence Prevention Measures: School Officials’ Attitudes about Various Strategies. Journal of Criminal Justice, 36(4), 301-306.
Waltz, C. F., Strickland, O., & Lenz, E. R. (2005). Measurement in nursing and health research (3rd ed.). New York: Springer Pub.
Zikmund, W. G. (2003). Business research methods (7th ed.). Mason, OH: Thomson/South-Western.