Self-perception is linked to a variety of psychosocial outcomes and its measurement has become a priority across a several disciplines. The Self-Perception Profile for Children (SPP-C) is commonly utilized to measure both global self worth and several important sub-domains of self-perception. Although much research has suggested this instrument possesses good internal consistency, previous investigations have primarily employed Cronbach's α; to estimate the stability of responding across items. This represents an important limitation, as α; is vulnerable to mis-estimation in the presence of correlated errors and non-τ-equivalent indicators, neither of which have been ruled out for the SPP-C. The present investigation initially examined the SPP-C responses from 106 girls, aged 8-12 to assess whether the assumptions underlying Cronbach's α; could be justified for this instrument. The investigation then re-estimated the internal consistency of the SPP-C using a covariance structure modeling approach. Results show that not a single scale of the SPP-C met all of the requirements for accurate estimation with α;. In two cases, α; was found to be meaningfully different from the reliability estimated with more durable methods. Lastly, the reliability of the Physical Appearance (PA) sub-scale was not significantly greater than the .70 cutoff for use as a research instrument. Discussion is centered on the appropriateness of Cronbach's α; for estimating the reliability of the SPP-C and recommends revision of the PA sub-scale.


Sara K. Sifers

First Committee Member

Lisa Perez

Second Committee Member

Cindra Kamphoff

Date of Degree




Document Type



Master of Arts (MA)




Social and Behavioral Sciences

Creative Commons License

Creative Commons Attribution-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-No Derivative Works 4.0 License.