Researchers have affirmed the existence of a relationship among tests, scales, populations, reliability, and validity (Kolen & Brennan, 2014). Scales often rely on unidimensionality because they use one item, which defines the unit of the construct. Franzen (2013) defines this unit as the level of severity within the construct under assessment. Scales and tests share a close relationship because the latter relies on scales as an instrument to test a population, or characteristic, in a research study. This is why scales are instrumental in organizing data for purposes of assessing reliability and validity in research. Therefore, researchers use scales to order subjects or characteristics of the testing (Kolen & Brennan, 2014). This is in lieu of the degree that scales use to display specific characteristics of a study.
Scales also give researchers the opportunity to rank their research units according to varied severity levels. This possibility also gives them an opportunity to change ordinal and nominal data to interval level scales (Franzen, 2013). These actions delimit nominal and ordinal data for use in other aspects of a research study. The samples that characterize the units of measurement in the scale determine how easy, or difficult, it is to generalize a measure across different populations. However, Kolen and Brennan (2014) say that generalizing the measures of a scale depends on the ability to generalize a scale across different types of populations. It is possible to affirm issues of validity and reliability using scales by comparing them to other data that measure the same constructs. Such a correlation should show the validity and reliability of the instrument used in a study. High reliability measures mean that most researchers could extrapolate the scales and tests to varied populations.
Application to Sampled Case Study
This section of the paper investigates the relationship among scales, populations, reliability, and validity through the works of Lin, Hsu, Li, Mathers and Huang (2010) when they developed an instrument to measure public health nurse competencies in Taiwan. In the same study, they also measured the nurses’ psychometric properties. As a demonstration of the relationship between scales and tests, the researchers used the Public Health Nurse professional Competency Scale as the instrument for assessing public health nurse competencies (as the test). The scale had 38 items, categorized into four groups/ domains.
Franzen (2013) says the construction of scales should consider issues of validity and reliability to affirm the usefulness of the scales. To do so, Lin, Hsu, Li, Mathers and Huang (2010) used a 7-member professional panel to determine the validity and reliability of the Public Health Nurse professional Competency Scale. The researchers assessed the content validity measures in each of the four domains created and found that the scale had strong reliability and validity measures. Based on the validity and reliability tests underpinning the development of the Public Health Nurse professional Competency Scale, it is plausible to believe that other researchers could apply the test and scale to other populations.
Suggestion and Validation of Idea
In my experience doing public health research studies, I believe that the relationship between tests and scales is direct because most tests use scales as the unit of assessment. Based on this extent of analysis, tests are reliant on scales as the nominal unit of assessment. However, their relationship with populations and reliability is indirect because researchers are limited to some tests when assessing different population dynamics (especially in public health studies). For example, it would be fallacious to use one test across varied population groups, of different socioeconomic classifications, to understand a research phenomenon. In fact, Kolen and Brennan (2014) say that for such generalizations to occur, researchers need to undertake further research to affirm the use of common tests and scales. Nonetheless, the validity and reliability of each scale and test depends on the reliability and validity tests used by the researcher. However, since there is a direct relationship between tests and scales, does this mean the same relationship exists across all levels of measurement in research?
References
Franzen, M. (2013). Reliability and Validity in Neuropsychological Assessment. New York, NY: Springer Science & Business Media.
Kolen, M., & Brennan, R. (2014). Test Equating, Scaling, and Linking: Methods and Practices. New York, NY: Springer Science & Business Media.
Lin, C., Hsu, C., Li, T., Mathers, N., & Huang, Y. (2010). Measuring professional competency of public health nurses: development of a scale and psychometric evaluation. J Clin Nurs, 19(21), 3161-70.