- April 7, 2016
- Posted by: Dr. R. Douglas Waldo, SPHR, SHRM-SCP, CFLDP
- Category: Technical
LDP Technical Brief 164007: Supplement to Brief 164006 Psychometric Reliability
Since the earlier stages of its development, the Leading Dimensions Profile (LDP) has been administered via two forms: Standard and Perfect-Actual. The Standard Form consists of 95 questions that are used in deriving two primary composite factors, six primary dimensions (facets of the composite factors), and four secondary dimensions (these are not included in calculating the composite factors). This same approach is leveraged via the Perfect-Actual form as well, except that each of the 95 questions is asked twice; once to elicit the participant’s own answer to the question, and once to seek the participant’s perception regarding what might be the perfect response to the same question. In other words, the Perfect-Actual Form encourages participants to compare and contrast their own responses to what might be considered the perfect responses of high performers, great leaders, or of some other ideal.
The Perfect-Actual Form is thought to diminish participants’ potential tendencies to fake an assessment, by requiring them to overtly indicate a perfect response for one presentation of the question, followed by their own actual response to another presentation of the same question. While a topic of both research and practical interest, users must consider the costs and benefits associated with anti-faking measures such as the Perfect-Actual Form. While this form may cause participants to pause and differentiate a preferred answer from their own, the method may also increase the response time required for the assessment. Users should be encouraged to consider the potential implications for using each form which, depending on their specific application, may vary significantly. To assist users in rendering such an evaluation, the psychometric reliability of each form was analyzed for LDPs administered between 2012 and 2015. The following table reveals Cronbach’s Alpha coefficients for the LDP’s factors and dimensions, while comparing the Standard and Perfect-Actual Forms (n=18,623).
Table 1: Reliability Coefficients for Cronbach’s Alpha Method (LDP Form Comparison)
|Constructs (Factors and Dimensions)||Items||Standard||Perfect-Actual|
|Achievement Drive Factor (Composite)||27||.87||.85|
|Primary Achieving Dimensions (Facets)|
|Secondary Achieving Dimensions|
|Relational Drive Factor (Composite)||29||.84||.84|
|Primary Relating Dimensions (Facets)|
|Secondary Relating Dimensions|
|n = 18,623|
Cronbach’s Alpha coefficients for the LDP’s composite factors ranged from .84 to .87 for the Standard Form and .84 to .85 for the Perfect-Actual Form. This reflects only minimal differences in reliability coefficients between forms. The difference was larger when comparing the primary dimensions between forms, where the Standard Form offered coefficients ranging from .73 to .82 for the primary dimensions, compared to a range of .72 to .80 exhibited by the Perfect-Actual Form. For the secondary dimensions, this coefficient ranged from .62 to .69 and .59 to .67 for the Standard Form and Perfect-Actual Form, respectively.
Practically speaking, only minimal differences in reliability were observed between the Standard Form and the Perfect-Actual Form. However, it should be noted that when coefficients differed between forms, in every instance the higher reliability was offered by the Standard Form. Given the shorter response time required by the Standard Form, and the potential for increased reliability, users may find that the advantages of this form outweigh the potential reduction in participant faking associated with the Perfect-Actual Form. Questions regarding this brief may be directed to Dr. Douglas Waldo, SPHR, SHRM-SCP, CFLDP at firstname.lastname@example.org.