Tools of Psychometric Measures and Scales Order Instructions: Measures and Scales
This week spend time discussing the similarities and differences between correlation and regression, and why one would use correlation instead of regression. Consider how often there are bivariate relationships and whether a bivariate relationship will be meaningful for your DBA doctoral study as compared to a regression, which examines relationships differently.
As a researcher, you will become familiar and comfortable with the measures and scales others in choose to measure data. Understanding study measurement methods can help clarify how you might employ these measurement instruments. In this paper, you will find an example of a test and a scale used in your discipline’s literature, analyze its use in the study, and analyze the relationships among these instruments, populations, and the concepts of reliability and validity.
To prepare for this paper
• Review Lesson 38, “Item Analysis Using Reliability Procedure” In Salkind and Green. Review in particular how to report psychometric properties, including correlation coefficients (pp. 345-347)
• Search for an example of a test and an example of a scale. For each example, determine the typical population and the reliability and validity.
To search for pre-existing measures that are available in the public domain do the following:
1. Note that most measurement development is conducted in the field of psychology. Make sure that you select peer-reviewed and full-text articles. You will also want to conduct an advanced search as multiple terms will be searched.
2. In the first box type in your construct (depression, leadership, etc.)
3. In the second box type in the word “psychometric”.
4. In the third box type in one of the following words (you may have to try a few different ones) “measure” “instrument” “scale” or “survey”.
Note: Using a public measure allows you the opportunity to view the measure without purchasing it such as the Buros Institute of Mental Measurements http://www.unl.edu/buros/bimm/html/00testscomplete.html
· What considerations need to be made to ensure that the scale and test examples are reliable and valid for another population
Tools of Psychometric Measures and Scales Resources
• Websites
Trochim, W. (2006). Web center for social research methods: Selecting statistics. Retrieved from http://www.socialresearchmethods.net/
This site is an online textbook that addresses the topics of a typical graduate course in social research methods, including sampling, measurement, validity, types of designs, and analysis. Learners can navigate the textbook using a graphical research road map or simple table of contents. The textbook addresses the entire research process as well as the statistical aspects of research and is not software-specific. From the main page, click on the ” Selecting Statistics” link.
Tools of Psychometric Measures and Scales Sample Answer
The soundness of any research should use tools of measurement that psychometrically correct and sound. After that, confirmation of the validity and reliability of the tools helps one in assuring the trueness of the findings of the study. A great problem encountered is knowing the kind of psychometric property to look into (Friman, et al.,(2013). These may include the evidence that will be sufficient for the study and what statistical is used to test mean. The support for of instruments can be assessed by testing the content of the analysis, the construct, and the related concepts. Validity is the ability of the instruments to measure the different factors of the area under study. On the other hand, reliability in psychometrics is the ability of the test to measure the area or phenomenon under study in a consistent manner (Bejar, et al.,(2012).
When reporting the psychometric properties, it is important to learn the influences of other factors such as personal and environmental factors relating to a phenomenon. Instruments used to measure the outcome can be used in evaluating the perceived performance level of the analysis (Allison, et al.,(2011). These instruments present themselves either generically or specifically to the population. The quality of the output of the analysis depends on its properties from the psychometric test. For example, in a prosthetics psychometric analysis, the effect of the outcome can be classified according to test levels. The levels of outcome are four.
Nominal data that are produced when the people are arranged in categories like people amputated can be put in one category and the right-hand amputees on another. The ordinal level data involves the categories that are mutually exclusive but exhaustive. The categories tend to be ordered but not equal in intervals. An example is people who can move with a stick, those who are not able to move with the help of a stick and those who can move without a stick. The interval level data encompasses order and an equal hiatus amid units of dimension without an exact zero (Friman, et al.,(2013). An example is the temperature zero in a thermometer which does not represent that there is no temperature. The ratio level data involves array and same intervals flanked by units of measurement but with a real zero. An example is height measurement where zero represents the absence of height.
Reliability test
When the required presentation should be routine based, a rater’s input is required. The forms of reliability can be in several forms. Test-retest reliability shows the consistency in the test results. An example is an individual taking a prosthetic fit survey and repeats it a day later. The result should be the same because no change is expected. In internal consistency, the reliability is preserved for the measures in the outcome that are intended to test only one concept and assess the extent to which other concepts in the phenomenon affect this concept. An example is a depression scale where consistency is not required for stuff such as SF-36 that addresses extra concepts. Intra-rater reliability shows the consistency of a rater in administering outcome measure and scores (Allison, et al.,(2011). An example is a video where rating and re-scoring later should give consistent results. Inter-rater reliability implies that two raters agree on how to administer the outcome or result. An example is two skilled raters of a videotape whose rated result should be similar.
Validity
It is an essential but not adequate feature of an outcome measure. Validity is determined only by a question pertaining to a certain population or phenomenon. The various kinds of validities and examples are discussed below. Face validity refers to the degree to which a test appears to gauge its intention. The experiment is slanted, and an example is a clinician advocating for a universal clinical practice. Its intention should measure exactly what the clinicians need. In content validity, the test should include all the contents present in the testing phenomenon (Bejar, et al.,(2012). An example is an instrument measuring athletes that should measure jumping, running and walking. Criterion validity is assessed when the outcome can be measured against a set normal. It is a rarely used criterion. Construct validity shows the ability of the test to measure the subject of the analyzer. An example is measuring the importance of day walk to the elderly on their mobility. There is difference recorded between those who work as village volunteers daily and those already in an elderly care centers.
Considerations for reliability and validity of test and scale
For reliability, temporal stability should be considered to ensure that test and scale examples are reliable and valid for another population. One should ensure that in the same test is utilized for more than one experiment. The result should not emerge from the adaptively of the researcher hence not biased. Form equivalence should be consistent and can be used by different examinees. There should also be an internal consistency of figures (Friman, et al.,(2013). The consistency in results should form a pattern. There should be uniformity in results hence; subsequent tests should bring the same outcome.
Validity should have face validity meaning that it takes the face value. Modifications can be sought from experts in the field. Content validity is a major consideration whereby the representative of the entire population should represent the other population. The criterion should be drawn from a presumption in test scores and performance. The prediction and explanation should be consistent (Bejar, et al.,(2012). The construct validation should be uniform.
Tools of Psychometric Measures and Scales Bibliography
Bejar, I.I., Chaffin, R. and Embretson, S., (2012). Cognitive and psychometric analysis of analogical problem solving. Springer Science & Business Media.
Allison, C., Baron-Cohen, S., Wheelwright, S.J., Stone, M.H. and Muncer, S.J., (2011). Psychometric analysis of the Empathy Quotient (EQ). Personality and Individual Differences, 51(7), pp.829-835.
Friman, M., Fujii, S., Ettema, D., Gärling, T. and Olsson, L.E., (2013). Psychometric analysis of the satisfaction with travel scale. Transportation Research Part A: Policy and Practice, 48, pp.132-145.
