Measurement and Scaling Concepts, Research Paper Example

Introduction

There are various approaches in conducting academic research. A number of approaches, including that of quantitative research use scaling as a form of measurement. This paper discusses the use of scales in research and the concept of validation.

Part 1 | Use Of Scales In Research

Quantitative Research approach tends to be scientific based using mathematics and statistics and is really the opposite of the qualitative approach.  In this sense it is rigid and lacks the flexibility of qualitative research.  It often uses the basis of hypothesis as a determinant to prove or disprove as the case may be.  Experiments tend to be around a fairly fixed or standard format approach.  This provides the important context of consistency and helps in making the results free from bias or not skewed by unwelcome variables.

There are varying types of scale methods in order to analyze survey data, an example being the Likert Scale.  They can be either dichotomous providing such responses as that of agree/disagree or yes/no etc., alternatively they may be structured in the format that contains answers like:

• Disagree
• strongly disagree
• agree
• strongly agree

The concept is to obtain comparative analysis over different types of studies. The Likert scale is one of the most commonly used format in quantitative research and takes on the following simple format:

Likert Scale:(researchproposalsforhealthprofessionals.com, 2011)

Strongly agree          5

Agree                          4

Uncertain                  3

Disagree                    2

Strongly disagree    1

The use of scales in research is generally classified into that of (i) Normal scales (ii) Ordinal Scales (iii) Interval Scales (iv) Ratio scales.

Nominal Scales – Tends not to be just a number but follows a format like that of:  NUTS= 1

BOLTS=2 and WIDGETS = 3

Ordinal Scale – Tends to follow some form of rank order, frequency level of agreement or ranked into a class of achievement i.e. Excellent, Good, Poor, Bad

Interval Scales – tends to use a rank order system and equal intervals i.e. IQ test scores, temperature, test scores etc.

Ratio Scales – tends to be true zero points i.e. weight, height, speed etc.

Items 3 and 4 have little distinction in basic research terms.

Validation In Research

The main concept of research is to analyse information in order to prove or disprove hypothesis, theories and opinions. In order to accomplish this information would be useless if it was inaccurate or invalid. As such there is an importance in the concept of validity. Validity is determined by asking questions like ‘what is being measured’? In this regard quantitative research is an area where investigative questions are most appropriate.  The actual importance of validity rests upon the ‘soundness of theoretical presupposition of a study’ and as such any theory must be stable supporting logical conclusions from the research made.

Qualitative Research | The researchers rely, implicitly or explicitly,   upon different interpretation and correlation in varyingtypes of validity in the explanation of different types of interest.  (Maxwell, J.A. 2010)

Quantitative Research | The roots of validity are found in a positivist tradition and therefore positivism has been defined as a systematic form of validity. Validity provides the true determinant of measures in the research, for example: laws, evidence. Objectivity, truth, deduction, reason etc. It was Wainer and Braun (1998) who cited that validity in quantitative research is construct validity (Golafshani, N. 2003)

Empirical Research | Validity is again rooted in positivism and the works of Thomas Kuhn. There is however scope for new thought processes and validity within the confines of empirical research, particularly concerning validity and that of ideological research. (Lather, P. 1986)

The question of reliability is validity and is to do with the term ‘measurement’. For example it is not possible to perfectly control an experiment or to have a perfectly calibrated instrument. As such there is always a level of inaccuracy presented to any scenario. The scientific world views validity as a common denominator in both cause and effect; as such it is concerned about the measurement criteria or the basis for measurement. There are a number of different pitfalls with ‘validity’ as a concept. One relates to that of reliability and another considers interpretation and ambiguity. The very meaning of words leaves scope for misinterpretation and ambiguity. (Kirk, M.L. 1986).

Validation can also refer to the strength and accuracy of a particular research design. In this regard it was Campbell and Stanley (1963) who stated that there were two types of validity in research design i.e. Internal Validity and External Validity. Internal Validity was the ability to test the research hypothesis and the extent to which the independent variable can be accurately stated.  High validity is the ability to control these independent vaiable(s) and as such ensure that other factors do not interfere with the results. Threats to internal validity include:  History, Maturation, Testing, and Selection of respondents, Regression and experimental mortality.  External validation relates more to the external general acceptance of the research findings. This in turn leads to the control of many external variables.(Srivastava, T. 2011)

References

Golafshani, N. (2003). Understanding reliability and validity in qualitative research. The qualitative report, Vol 8 No 4, 597-607.

Kirk, M. (1986). Reliability and validity in qualitative research. Qualitsative Research, 20-24.

Lather, P. (1986). Issues of validity in open ideological research. Interchange, Vol 17, 63-84.

Maxwell, J. (2010). Understanding and Validity in qualitative research. Hardward Educational Review, 279-301.

researchproposalsforhealthprofessionals.com. (2011, 12 8). Likert Scale. Retrieved from researchproposalsforhealthprofessionals.com: http://www.researchproposalsforhealthprofessionals.com/likert_scale.htm

Srivastava, T. (2011). Business Research Methodology. New Delhi: McGraw Hill.