preview

Neo-R Personality Assessment

Decent Essays
Open Document

I am to assume the role of career counselor at a college, and proffer career/vocational recommendations to one John Lee, a Freshman at the college. These recommendations are to be based upon results from Mr. Lee’s completion of the “Neuroticism, Extroversion, & Openness-Revised” (NEO-R) personality assessment instrument, in addition to review of career goals questionnaire, school/work history, and personal interview with the student (Kirwan, 2014). The NEO-R is to serve only as a template to evaluate and consider vocational aspirations for Mr. Lee, and further careful deliberation is paramount in the selection of the occupation that will engage him for remainder of his vocational life.

The NEO-R instrument consists of 240 items presented …show more content…

This is to be expected, as self-reporting measures proceed from subjective data gathering modality. This places the validity and reliability of the instrument (and the results) at risk for misinterpretation, and caution should be exercised in the examination and analysis of the sum data. There are numerous methods by which to assure validity and reliability in the administration and interpretation of assessment instruments. The constructs and framing of reliability and validity are disengaged into types, rendering scrutiny more accurate and duplicable. With respect to personality assessment inventories, types of validity would translate into an extension of types of reliability, the two most principal of which would be “parallel-forms” reliability (Frazier, Naugle, & Haggerty, 2006), and “inter-rater” reliability (Ingenhoven & Abraham, 2010). It is largely accepted that the phenomenon of “faking” responses is a factor in the self-reporting assessment administration, and can even be attributed to inadvertent (accidental) impulsive (spur-of-the-moment) categorical responding on the part of the participant, “Persons may either consciously or unconsciously provide inaccurate responses to instrument inquiries” through “faking” of responses to said inquiries (Groth-Marnat, 2003, p. 65). Inter-rater reliability involves the judgment of multiple professional interpreters of the instrument and the results (Ingenhoven & Abraham, p. 237), which can be useful in that more than one professional can diversify the evaluation of the assessment, advancing a general consensus, to the exclusion of discursive interpretation. Parallel-forms reliability requires that the assessment inquiries (questions) be randomly displaced (moved around) from test to test (assessment form) (Frazier, Naugle, & Haggerty,

Get Access