Article Review and Analysis – Validity
� Briefly describe the types of validity presented. What are the critical differences among them? Assess
the authors’ performance in explaining them.
� Select a previous article that presents a quantitative study. How would you assess the study’s validity?
What information would you need in order to be able to do so, and is that information present in the
article?
� Based on Table 8 on page 415 in the article by Straub, Boudreau, & Gefen, which positivist design
contingency best describes the previous article you selected ? Which best describes your potential
Doctoral Study (Effects of HRM Practices on Employee Performance)? Explain your selections.
Article Review and Analysis – Validity 2
Introduction
To provide complete and accurate view of research work, a given construct must draw
confidence in its representation and ability to validate its own concepts based on its empirical
findings (Straub, Boudreau & Gefen, 2004).
Content Validity
The construct would be more helpful and meaningful if the validities chosen reflect the
heuristics presented in the study. Assessing content validity is very difficult as researchers are
uncertain on how to rate the degree or level of completeness of their own research work. Content
validity is basically sampling and evaluating the contents of research work for its validity
(Straub, Boudreau & Gefen, 2004).
Construct Validity
Construct validity provides the measurement between constructs or operations. Construct
validity appeals more to the content of the construct’s validity than the substance of the
construct. It raises the fundamental question as to whether the construct fits in its application by
the researcher in an attempt to capture the significance of the construct. The application of
nomological network is applied to test if there are links similar to the ones existing on the
construct literature. Construct validity focuses more on the likeness or differences of the
construct linkages and their strengths on past literature (Straub, Boudreau & Gefen, 2004).
Predictive Validity
Article Review and Analysis – Validity 3
It is also known as concurrent validity and it establishes relationship between different
measures and constructs through demonstration of posited measures directed on specific
construct correlations. Predictive validity predicts outcomes for given variables but they also
provide conceptual meaning applies to the constructs antecedents and ramifications. The goal of
predictive validity is basically prediction while also reinforcing the concept of the theory base
(Szajna, 1994).
The author has explained the differences as well as the similarities of the various validity
measures. For example, the author has explained that predictive validity validates both the
exogenous as well as the endogenous constructs based on the use of z-scores (Szajna, 1994).
The peer review research article by Klein and Olbrecht (2011) demonstrates the need to widen
and work on more exploratory research work. The article summarizes the triangulation of
qualitative and quantitative data analysis. The oral presentation and the other protocols provided
as the basis of the research work requires more cognitive frameworks to present constructive and
validated data. The intractable domains are also difficult to establish as the research relies on
limited primary data. To assess the validity of the primary data I would require the
questionnaires that were used to compile the research report (Straub, Boudreau & Gefen, 2004).
The non-theoretical work which bases its findings on the primary descriptive data would
provide a basis for predictive validity. The predictive validity predicts outcomes for given
variables but its conceptual meaning applies to the constructs antecedents and ramifications
which are collaborated with the content validity. The research work can only be validated by
confirming the constructs relationship with other research work from well established research
streams. The article confirms that the validity of scientific that is based on positivist science
needs based not only on highly biased observations but also on a series of random anecdotes that
Article Review and Analysis – Validity 4
tests the intellectual constructs of the research presented. The article requires the rigor of
intelligent, careful and thoughtful collection of primary data (Carrier et al, 1990).
The best description of my potential doctoral study is the exploratory that specifically
probe the areas in HRM that are not well understood like valuation of human resources as a key
element of a company’s capital structure.
Article Review and Analysis – Validity 5
References
Carrier, M. R., A. T. Dalessio, and S. H. Brown (1990). “Correspondence between Estimates of
Content and Criterion-Related Validity Values,” Personnel Psychology (43) 1 (spring),
pp. 85-100.
Klein, T. and Olbrecht, M. (2011) Triangulation of Qualitative and Quantitative Methods in
Panel Peer Review Research, International Journal for Cross-Disciplinary Subjects in
Education (IJCDSE), Volume 2, Issue 2, June, pg 345 -348.
Szajna, B. (1994) “Software Evaluation and Choice: Predictive Validation of the Technology
Acceptance Instrument,” MIS Quarterly (17) 3, pp. 319-324.
Straub, D., & Boudreau, M., & Gefen, D. (2004) Validation Guidelines for Positivist Research,
Communications of the Association for Information Systems, Vol 13, pg 380 – 427.
